Change search
Refine search result
12 1 - 50 of 52
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Rows per page
  • 5
  • 10
  • 20
  • 50
  • 100
  • 250
Sort
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
Select
The maximal number of hits you can export is 250. When you want to export more records please use the Create feeds function.
  • 1. Andersson, Thord
    et al.
    Romu, Thobias
    Karlsson, Anette
    Norén, Bengt
    Forsgren, Mikael F
    Smedby, Örjan
    KTH, School of Technology and Health (STH), Medical Engineering, Medical Image Processing and Visualization. KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Linköping University.
    Kechagias, Stergios
    Almer, Sven
    Lundberg, Peter
    Borga, Magnus
    Leinhard, Olof Dahlqvist
    Consistent intensity inhomogeneity correction in water-fat MRI2015In: Journal of Magnetic Resonance Imaging, ISSN 1053-1807, E-ISSN 1522-2586, Vol. 42, no 2Article in journal (Refereed)
    Abstract [en]

    PURPOSE: To quantitatively and qualitatively evaluate the water-signal performance of the consistent intensity inhomogeneity correction (CIIC) method to correct for intensity inhomogeneities

    METHODS: Water-fat volumes were acquired using 1.5 Tesla (T) and 3.0T symmetrically sampled 2-point Dixon three-dimensional MRI. Two datasets: (i) 10 muscle tissue regions of interest (ROIs) from 10 subjects acquired with both 1.5T and 3.0T whole-body MRI. (ii) Seven liver tissue ROIs from 36 patients imaged using 1.5T MRI at six time points after Gd-EOB-DTPA injection. The performance of CIIC was evaluated quantitatively by analyzing its impact on the dispersion and bias of the water image ROI intensities, and qualitatively using side-by-side image comparisons.

    RESULTS: CIIC significantly ( P1.5T≤2.3×10-4,P3.0T≤1.0×10-6) decreased the nonphysiological intensity variance while preserving the average intensity levels. The side-by-side comparisons showed improved intensity consistency ( Pint⁡≤10-6) while not introducing artifacts ( Part=0.024) nor changed appearances ( Papp≤10-6).

    CONCLUSION: CIIC improves the spatiotemporal intensity consistency in regions of a homogenous tissue type.

  • 2.
    Astaraki, Mehdi
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Wang, Chunliang
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Buizza, G.
    Toma-Dasu, I.
    Lazzeroni, M.
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Early survival prediction in non-small cell lung cancer with PET/CT size aware longitudinal pattern2019In: Radiotherapy and Oncology, ISSN 0167-8140, E-ISSN 1879-0887, ISSN 0167-8140, Vol. 133, p. S208-S209Article in journal (Refereed)
  • 3.
    Astaraki, Mehdi
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Karolinska Inst, Dept Oncol Pathol, Karolinska Univ Sjukhuset, SE-17176 Stockholm, Sweden.
    Wang, Chunliang
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Buizza, Giulia
    Politecn Milan, Dept Elect Informat & Bioengn, Piazza Leonardo da Vinci 42, I-20133 Milan, Italy..
    Toma-Dasu, Iuliana
    Karolinska Inst, Dept Oncol Pathol, Karolinska Univ Sjukhuset, SE-17176 Stockholm, Sweden.;Stockholm Univ, Dept Phys, SE-10691 Stockholm, Sweden..
    Lazzeroni, Marta
    Karolinska Inst, Dept Oncol Pathol, Karolinska Univ Sjukhuset, SE-17176 Stockholm, Sweden.;Stockholm Univ, Dept Phys, SE-10691 Stockholm, Sweden..
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Early survival prediction in non-small cell lung cancer from PET/CT images using an intra-tumor partitioning method2019In: Physica medica (Testo stampato), ISSN 1120-1797, E-ISSN 1724-191X, Vol. 60, p. 58-65Article in journal (Refereed)
    Abstract [en]

    Purpose: To explore prognostic and predictive values of a novel quantitative feature set describing intra-tumor heterogeneity in patients with lung cancer treated with concurrent and sequential chemoradiotherapy. Methods: Longitudinal PET-CT images of 30 patients with non-small cell lung cancer were analysed. To describe tumor cell heterogeneity, the tumors were partitioned into one to ten concentric regions depending on their sizes, and, for each region, the change in average intensity between the two scans was calculated for PET and CT images separately to form the proposed feature set. To validate the prognostic value of the proposed method, radiomics analysis was performed and a combination of the proposed novel feature set and the classic radiomic features was evaluated. A feature selection algorithm was utilized to identify the optimal features, and a linear support vector machine was trained for the task of overall survival prediction in terms of area under the receiver operating characteristic curve (AUROC). Results: The proposed novel feature set was found to be prognostic and even outperformed the radiomics approach with a significant difference (AUROC(sALop) = 0.90 vs. AUROC(radiomic) = 0.71) when feature selection was not employed, whereas with feature selection, a combination of the novel feature set and radiomics led to the highest prognostic values. Conclusion: A novel feature set designed for capturing intra-tumor heterogeneity was introduced. Judging by their prognostic power, the proposed features have a promising potential for early survival prediction.

  • 4.
    Batool, Nazre
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Chowdhury, Manish
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Moreno, Rodrigo
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Estimation of trabecular bone thickness in gray scale: a validation study2017In: International Journal of Computer Assisted Radiology and Surgery, Vol. 12, no Supplement 1Article in journal (Refereed)
  • 5.
    Bendazzoli, Simone
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems.
    Brusini, Irene
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Karolinska Inst, Dept Neurobiol Care Sci & Soc, Alfred Nobels Alle 23,D3, S-14152 Huddinge, Sweden..
    Damberg, Peter
    Karolinska Inst, Dept Clin Neurosci, Tomtebodavagen 18A P1 5, S-17177 Stockholm, Sweden..
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Andersson, Leif
    Uppsala Univ, Dept Med Biochem & Microbiol, Sci Life Lab Uppsala, Biomedicinskt Ctr BMC, Husargatan 3, S-75237 Uppsala, Sweden..
    Wang, Chunliang
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Automatic rat brain segmentation from MRI using statistical shape models and random forest2019In: MEDICAL IMAGING 2019: IMAGE PROCESSING / [ed] Angelini, ED Landman, BA, SPIE-INT SOC OPTICAL ENGINEERING , 2019, article id 1094920Conference paper (Refereed)
    Abstract [en]

    In MRI neuroimaging, the shimming procedure is used before image acquisition to correct for inhomogeneity of the static magnetic field within the brain. To correctly adjust the field, the brain's location and edges must first be identified from quickly-acquired low resolution data. This process is currently carried out manually by an operator, which can be time-consuming and not always accurate. In this work, we implement a quick and automatic technique for brain segmentation to be potentially used during the shimming. Our method is based on two main steps. First, a random forest classifier is used to get a preliminary segmentation from an input MRI image. Subsequently, a statistical shape model of the brain, which was previously generated from ground-truth segmentations, is fitted to the output of the classifier to obtain a model-based segmentation mask. In this way, a-priori knowledge on the brain's shape is included in the segmentation pipeline. The proposed methodology was tested on low resolution images of rat brains and further validated on rabbit brain images of higher resolution. Our results suggest that the present method is promising for the desired purpose in terms of time efficiency, segmentation accuracy and repeatability. Moreover, the use of shape modeling was shown to be particularly useful when handling low-resolution data, which could lead to erroneous classifications when using only machine learning-based methods.

  • 6.
    Brusini, Irene
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Carneiro, Miguel
    Univ Porto, Ctr Invest Biodiversidade & Recursos Genet CIBIO, InBIO, P-4485661 Vairao, Portugal.;Univ Porto, Dept Biol, Fac Ciencias, P-4169007 Porto, Portugal..
    Wang, Chunliang
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Rubin, Carl-Johan
    Uppsala Univ, Sci Life Lab Uppsala, Dept Med Biochem & Microbiol, S-75236 Uppsala, Sweden..
    Ring, Henrik
    Uppsala Univ, Dept Neurosci, S-75236 Uppsala, Sweden..
    Afonso, Sandra
    Univ Porto, Ctr Invest Biodiversidade & Recursos Genet CIBIO, InBIO, P-4485661 Vairao, Portugal..
    Blanco-Aguiar, Jose A.
    Univ Porto, Ctr Invest Biodiversidade & Recursos Genet CIBIO, InBIO, P-4485661 Vairao, Portugal.;CSIC, Inst Invest Recursos Cineget IREC, Ciudad Real 13005, Spain.;UCLM, CSIC, Ciudad Real 13005, Spain..
    Ferrand, Nuno
    Univ Porto, Ctr Invest Biodiversidade & Recursos Genet CIBIO, InBIO, P-4485661 Vairao, Portugal.;Univ Porto, Dept Biol, Fac Ciencias, P-4169007 Porto, Portugal.;Univ Johannesburg, Dept Zool, ZA-2006 Auckland Pk, South Africa..
    Rafati, Nima
    Uppsala Univ, Sci Life Lab Uppsala, Dept Med Biochem & Microbiol, S-75236 Uppsala, Sweden..
    Villafuerte, Rafael
    CSIC, IESA, Cordoba 14004, Spain..
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Damberg, Peter
    Karolinska Univ Hosp, Karolinska Expt Res & Imaging Ctr, S-17176 Solna, Sweden..
    Hallbook, Finn
    Uppsala Univ, Dept Neurosci, S-75236 Uppsala, Sweden..
    Fredrikson, Mats
    Uppsala Univ, Dept Psychol, S-75236 Uppsala, Sweden.;Karolinska Inst, Dept Clin Neurosci, S-17177 Stockholm, Sweden..
    Andersson, Leif
    Uppsala Univ, Sci Life Lab Uppsala, Dept Med Biochem & Microbiol, S-75236 Uppsala, Sweden.;Texas A&M Univ, Coll Vet Med & Biomed Sci, Dept Vet Integrat Biosci, College Stn, TX 77843 USA.;Swedish Univ Agr Sci, Dept Anim Breeding & Genet, S-75007 Uppsala, Sweden..
    Changes in brain architecture are consistent with altered fear processing in domestic rabbits2018In: Proceedings of the National Academy of Sciences of the United States of America, ISSN 0027-8424, E-ISSN 1091-6490, Vol. 115, no 28, p. 7380-7385Article in journal (Refereed)
    Abstract [en]

    The most characteristic feature of domestic animals is their change in behavior associated with selection for tameness. Here we show, using high-resolution brain magnetic resonance imaging in wild and domestic rabbits, that domestication reduced amygdala volume and enlarged medial prefrontal cortex volume, supporting that areas driving fear have lost volume while areas modulating negative affect have gained volume during domestication. In contrast to the localized gray matter alterations, white matter anisotropy was reduced in the corona radiata, corpus callosum, and the subcortical white matter. This suggests a compromised white matter structural integrity in projection and association fibers affecting both afferent and efferent neural flow, consistent with reduced neural processing. We propose that compared with their wild ancestors, domestic rabbits are less fearful and have an attenuated flight response because of these changes in brain architecture.

  • 7.
    Brusini, Irene
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Jörgens, Daniel
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Moreno, Rodrigo
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Dependency of neural tracts'€™ curvature estimations on tractography methods2017Conference paper (Refereed)
  • 8.
    Brusini, Irene
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Jörgens, Daniel
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Moreno, Rodrigo
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Influence of Tractography Algorithms and Settings on Local Curvature Estimations2017Conference paper (Refereed)
  • 9.
    Brusini, Irene
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Jörgens, Daniel
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Moreno, Rodrigo
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Voxel-Wise Clustering of Tractography Data for Building Atlases of Local Fiber Geometry2019Conference paper (Refereed)
    Abstract [en]

    This paper aims at proposing a method to generate atlases of white matter fibers’ geometry that consider local orientation and curvature of fibers extracted from tractography data. Tractography was performed on diffusion magnetic resonance images from a set of healthy subjects and each tract was characterized voxel-wise by its curvature and Frenet–Serret frame, based on which similar tracts could be clustered separately for each voxel and each subject. Finally, the centroids of the clusters identified in all subjects were clustered to create the final atlas. The proposed clustering technique showed promising results in identifying voxel-wise distributions of curvature and orientation. Two tractography algorithms (one deterministic and one probabilistic) were tested for the present work, obtaining two different atlases. A high agreement between the two atlases was found in several brain regions. This suggests that more advanced tractography methods might only be required for some specific regions in the brain. In addition, the probabilistic approach resulted in the identification of a higher number of fiber orientations in various white matter areas, suggesting it to be more adequate for investigating complex fiber configurations in the proposed framework as compared to deterministic tractography.

  • 10.
    Buizza, Giulia
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems. Politecn Milan, CartCasLab, Dept Elect Informat & Bioengn, Piazza Leonardo Da Vinci 42, I-20133 Milan, Italy..
    Toma-Dasu, Iuliana
    Karolinska Univ Sjukhuset, Karolinska Inst, Dept Oncol Pathol, Med Radiat Phys, S-17176 Solna, Sweden..
    Lazzeroni, Marta
    Karolinska Univ Sjukhuset, Karolinska Inst, Dept Oncol Pathol, Med Radiat Phys, S-17176 Solna, Sweden..
    Paganelli, Chiara
    Politecn Milan, CartCasLab, Dept Elect Informat & Bioengn, Piazza Leonardo Da Vinci 42, I-20133 Milan, Italy..
    Riboldi, Marco
    Politecn Milan, CartCasLab, Dept Elect Informat & Bioengn, Piazza Leonardo Da Vinci 42, I-20133 Milan, Italy.;Ludwig Maximilians Univ Munchen, Fac Phys, Coloumbwall 1, D-5748 Garching, Germany..
    Chang, Yong Jun
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems.
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Wang, Chunliang
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Early tumor response prediction for lung cancer patients using novel longitudinal pattern features from sequential PET/CT image scans2018In: Physica medica (Testo stampato), ISSN 1120-1797, E-ISSN 1724-191X, Vol. 54, p. 21-29Article in journal (Refereed)
    Abstract [en]

    Purpose: A new set of quantitative features that capture intensity changes in PET/CT images over time and space is proposed for assessing the tumor response early during chemoradiotherapy. The hypothesis whether the new features, combined with machine learning, improve outcome prediction is tested. Methods: The proposed method is based on dividing the tumor volume into successive zones depending on the distance to the tumor border. Mean intensity changes are computed within each zone, for CT and PET scans separately, and used as image features for tumor response assessment. Doing so, tumors are described by accounting for temporal and spatial changes at the same time. Using linear support vector machines, the new features were tested on 30 non-small cell lung cancer patients who underwent sequential or concurrent chemoradiotherapy. Prediction of 2-years overall survival was based on two PET-CT scans, acquired before the start and during the first 3 weeks of treatment. The predictive power of the newly proposed longitudinal pattern features was compared to that of previously proposed radiomics features and radiobiological parameters. Results: The highest areas under the receiver operating characteristic curves were 0.98 and 0.93 for patients treated with sequential and concurrent chemoradiotherapy, respectively. Results showed an overall comparable performance with respect to radiomics features and radiobiological parameters. Conclusions: A novel set of quantitative image features, based on underlying tumor physiology, was computed from PET/CT scans and successfully employed to distinguish between early responders and non-responders to chemoradiotherapy.

  • 11.
    Chen, Hongjian
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Evangelou, Dimitris
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Grishenkov, Dmitry
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Karolinska Institutet (KI), CLINTEC – Division of Medical Imaging and Technology.
    Sequence design for ultrasound imaging of polyvinyl alcohol microbubbles2019Conference paper (Refereed)
    Abstract [en]

    Nonlinear behavior of the ultrasound contrast agent (UCA) offers a unique feature to be distinguished from the surrounding tissue. In a recent years several methods were developed to enhance the nonlinear response of UCA. Crucial for efficient differentiation of the nonlinear response of UCA from the surrounding tissue is to design the contrast pulse sequence specific to the unique nonlinear properties that the particular UCA is offering.

    In the previous study, the nonlinear response from a novel polyvinyl alcohol (PVA) microbubbles (MB), in ultra-harmonic region was investigated over a pressure range from 50 kPa to 300 kPa. In this study, five contrast pulse sequences and reference B-mode sequence were designed to visualize PVA MB. The performance of those sequences were evaluated and compared.

  • 12.
    Chen, Hongjian
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Larsson, David
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Janerot-Sjöberg, Birgitta
    Colarieti-Tosti, Massimiliano
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Grishenkov, Dmitry
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Polymer Microbubbles as Dual Modal Contrast Agent for Ultrasound and Computed Tomography2018Conference paper (Refereed)
    Abstract [en]

    The hybrid imaging combines the anatomical information with the functional or metabolic information using different conventional single imaging modalities improving the overall diagnosis outcome of the clinical examination. Since the introduction of the first hybrid imaging device PET-CT in 1998 different combinations of hybrid imaging were developed such as PET-MRI, SPECT-CT.

    However, lack of multimodal contrast agent specifically aimed for hybrid imaging limits the diagnostic outcome of these novel techniques. Initial attempts in fabrication of hybrid contrast agents were made by combining previously existing single modal contrast agents into one. In this study, polyvinyl alcohol (PVA) microbubbles (MB) and gold nanoparticles - which by themselves are already established contrast agents used in preclinical studies for ultrasound and CT, respectively - were chosen as parent contrast agents to fabricate the dual modal Contrast Agent for UltraSound and CT (CACTUS).

    Method

    The fabrication of MBs was adapted from Cavalieri et al.[1]. PVA powder (Sigma Aldrich, MO USA) was dissolved in the water at 80°C. The aqueous PVA-chains were cleaved by sodium metaperiodate (NaIO4, purity>99.0%, Sigma Aldrich, MO USA). Vigorous stirring force was applied to the resulting telechelic aldehydic PVA-chains for 2 hours to crosslink the telechelic aldehydic PVA-chains and form the PVA-coated MBs at the water-air interface.

    CACTUS MBs were synthesized in a similar fashion to the above, but adding gold nanoparticles (diameter 1.9nm, Nanoprobes, NY, USA) during formation of the MBs.

    The size distributions of MBs and CACATUS MBs were determined using an optical microscope (ECLIPSE Ci-S, Nikon, Tokyo, Japan) and a Neubauer counting chamber (Brand GmbH, Wertheim, Germany).

    The acoustic attenuation coefficients of the MBs suspension were acquired at peak negative pressure (PNP) from 10 - 300 kPa. Three MBs suspension samples with concentrations of (sample A),  (sample B) and  ml-1 (sample C) were prepared and loaded in a 1 cm thick two-cavity chamber. A flat single crystal ultrasound transducer with central frequency 3.5MHz was used to generate the ultrasound beam. The amplitude of received echoes through samples and water were compared at the fundamental frequency, as well as the 2nd and 3rd harmonic for each value of the concentration used.

    The mass attenuation of water, suspension of gold nanoparticles with concentration 160mg/L, plain MBs, and CACTUS MBs, was measured by quantum FX-CT micro-CT (PerkinElmer Inc, MA, USA). The micro-CT was operated at a current of 200mA with exposure time of 120s and varied voltage 50kV, 70kV and 90kV. Each 3D image has a size of 512*512*512 pixels or 75.8*75.8*75.8 mm. Contrast to noise ratios (CNR) between water and all samples were calculated following Eq. 1.Where S(x,y,z) and W(x,y,z) are the mass attenuation of the sample and water per voxel, respectively. ns(x,y,z) and nw(x,y,z) are the noise function with zero mean of sample and water respectively. Ms and Mw are the mean mass attenuation acquired for the sample and water in the volume of interest. The σs2 and σw2 are the variance of the mass attenuation read out of the sample and water in the volume of interested.

    In addition to the gas-core MBs for the CT tests, liquid-core gold loaded capsules were synthesized in two steps. In the first step, PVA shelled liquid-core capsules were obtained by exposing MBs to 66% v/v ethanol solution. In the second step, the resulting liquid-core capsules were mixed with high concentration gold nanoparticles suspension and homogenized by a shaker (MS 3 basic, IKA, Königswinter Germany) at 500rpm for 1 hour for goal loading. The resulting gold loaded capsules were washed with Milli-Q water using centrifuge (Galaxy 5D digital microcentrifuge, VWR, USA) at a speed of 1000 g for 5 min.

    Results and discussion

    The mean diameter of MBs is 3.6±1.1 μm. The mean diameter of CACTUS MBs is 3.2±0.7 μm. The size distribution of the gold loaded capsules was not investigated separately, but rather assumed identical to the plain MBs. The number and the volume distribution of MBs and CACTUS MBs are shown in figure 1. The results demonstrate that most of the CACTUS MBs and MBs have a diameter from 1 to 6 μm. Therefore, they are able pass through the capillaries and will resonate within typical clinical diagnostic ultrasound frequency below 15 MHz.

    Pressure dependent acoustic attenuation coefficients of the sample A, B, and C are shown in figure 2. The results show that attenuation coefficients of sample A and B at the fundamental frequency stay constant and slightly increase at the second harmonic at the PNP below 100kPa, indicating a linear oscillation of MBs. As the PNP reaches 200kPa, the attenuation coefficient of sample A at fundamental frequency decreases while at 2nd and 3rd harmonics increases, indicating that the energy of the echo shifts from the fundamental frequency to the 2nd and 3rd harmonics. As the PNP goes higher to 300kPa, the attenuation coefficient of sample A at the fundamental frequency, 2nd, and 3rd harmonics decreases, suggesting that the energy shifts to an even higher harmonic. At the same time, the attenuation coefficient of sample B stays constant at fundamental frequency, decreases at 2nd harmonics, and increases at the 3rd harmonic, suggesting the energy starts to shift to the 3rd harmonic. The attenuation coefficient of sample C at fundamental frequency, 2nd and 3rd harmonics keep constant and low due to low sample concentration. The test reveals the energy shifting of the echo to the higher harmonics at PNP higher than 100 kPa, indicating the nonlinear oscillation of MBs at PNP higher than 100 kPa. Moreover, the concentration of the MBs seems to influence the energy shifting: the higher the concentration the earlier the shift to the higher harmonics occurs, in the range of the concentration consider in this study.

    The pilot results of the micro-CT tests are presented in Table 1. The reference, gold nanoparticles solution, has the highest CNR per voxel at all CT operating voltages. The CNR per voxel of CACTUS MBs suspensions is below 0.1, virtually equaling the MBs at all operating voltages, suggesting that no gold or very little gold were loaded into the shell of the CACTUS MBs. The gold loaded capsules suspension has higher CNR per voxel than the capsule supernatant (the surrounding environment of capsules) and the MBs suspension, implying that the gold nanoparticles were loaded into the capsules. However, it is not clear whether the gold nanoparticles were loaded in the core of the MBs or in the MBs shell. The expected sharp increase of CNR per voxel at the k-edge of gold did not appear. We believe that is because even at our highest operating voltage of 90kV, the percentage of the photons with energy higher than 80.7 keV is still low. Introduction of a high-pass metal filter could increase the percentage of high energy photon. On the other hand, the metal filter will reduce the total number of the photons which would increase the noise of the images. Since same current was applied on every CT test, less X-ray photons reached the sensors when the CT was operated at low voltage. Therefore, it might be worth performing additional calibration tests to adjust the operating currents to make sure that the numbers of the photons that reach the sensor at every operating voltage are the same.

    Conclusion

    In this study, the CACTUS MBs and gold loaded capsules were fabricated as potential candidates for dual modal contrast agent. The characterization revealed that gold loaded capsule is a promising initial step. Nevertheless, the method to convert back liquid-core capsules to gas-core MBs needs to be established.

    [1] Cavalieri, F., El Hamassi, A., Chiessi, E., Paradossi, G., Villa, R., & Zaffaroni, N. (2006). Tethering functional ligands onto shell of ultrasound active polymeric microbubbles. Biomacromolecules, 7(2), 604-611.

  • 13.
    Dillon-Murphy, Desmond
    et al.
    Kings Coll London, Sch Biomed Engn & Imaging Sci, London, England..
    Marlevi, David
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Ruijsink, Bram
    Kings Coll London, Sch Biomed Engn & Imaging Sci, London, England..
    Qureshi, Ahmed
    Kings Coll London, Sch Biomed Engn & Imaging Sci, London, England..
    Chubb, Henry
    Stanford Univ, Dept Cardiothorac Surg, Palo Alto, CA 94304 USA..
    Kerfoot, Eric
    Kings Coll London, Sch Biomed Engn & Imaging Sci, London, England..
    O'Neill, Mark
    Kings Coll London, Sch Biomed Engn & Imaging Sci, London, England..
    Nordsleffen, David
    Kings Coll London, Sch Biomed Engn & Imaging Sci, London, England..
    Aslanidi, Oleg
    Kings Coll London, Sch Biomed Engn & Imaging Sci, London, England..
    de Vecchi, Adelaide
    Kings Coll London, Sch Biomed Engn & Imaging Sci, London, England..
    Modeling Left Atrial Flow, Energy, Blood Heating Distribution in Response to Catheter Ablation Therapy2018In: Frontiers in Physiology, ISSN 1664-042X, E-ISSN 1664-042X, Vol. 9, article id 1757Article in journal (Refereed)
    Abstract [en]

    Introduction: Atrial fibrillation (AF) is a widespread cardiac arrhythmia that commonly affects the left atrium (LA), causing it to quiver instead of contracting effectively. This behavior is triggered by abnormal electrical impulses at a specific site in the atrial wall. Catheter ablation (CA) treatment consists of isolating this driver site by burning the surrounding tissue to restore sinus rhythm (SR). However, evidence suggests that CA can concur to the formation of blood clots by promoting coagulation near the heat source and in regions with low flow velocity and blood stagnation. Methods: A patient-specific modeling workflow was created and applied to simulate thermal-fluid dynamics in two patients pre- and post-CA. Each model was personalized based on pre- and post-CA imaging datasets. The wall motion and anatomy were derived from SSFP Cine MRI data, while the trans-valvular flow was based on Doppler ultrasound data. The temperature distribution in the blood was modeled using a modified Pennes bioheat equation implemented in a finite-element based Navier-Stokes solver. Blood particles were also classified based on their residence time in the LA using a particle-tracking algorithm. Results: SR simulations showed multiple short-lived vortices with an average blood velocity of 0.2-0.22 m/s. In contrast, AF patients presented a slower vortex and stagnant flow in the LA appendage, with the average blood velocity reduced to 0.08-0.14 m/s. Restoration of SR also increased the blood kinetic energy and the viscous dissipation due to the presence of multiple vortices. Particle tracking showed a dramatic decrease in the percentage of blood remaining in the LA for longer than one cycle after CA (65.9 vs. 43.3% in patient A and 62.2 vs. 54.8% in patient B). Maximum temperatures of 76 degrees and 58 degrees C were observed when CA was performed near the appendage and in a pulmonary vein, respectively. Conclusion: This computational study presents novel models to elucidate relations between catheter temperature, patient-specific atrial anatomy and blood velocity, and predict how they change from SR to AF. The models can quantify blood flow in critical regions, including residence times and temperature distribution for different catheter positions, providing a basis for quantifying stroke risks.

  • 14.
    Ghorbani, Morteza
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Sabanci Univ, Fac Engn & Nat Sci, Mechatron Engn Program, Istanbul, Turkey.
    Numerical study of cavitating flow in orifices and its effect on spray characteristics2018In: Journal of Hydrodynamics, ISSN 1001-6058, E-ISSN 1000-4874, Vol. 30, no 5, p. 908-919Article in journal (Refereed)
    Abstract [en]

    The bubbly flow regime inside orifices has significant effects on several applications, and studying its trend along an orifice could be helpful in identifying the flow mechanism in various situations. The flow regime inside an orifice depends on the situation which has been specified for the orifice. Orifice geometry has a considerable effect on bubbly flow in injectors. Meanwhile, spray characteristics are influenced by the fuel flow inside an orifice, which has strong effects on the mixture of fuel-air. In this study, spray characteristics are studied for different values of the orifice angle. The cavitation phenomenon which occurs inside an orifice varies in intensity and patterns at different angles of the orifice and consequently has diverse effects on spray characteristics. The governing equations are solved by the SIMPLE algorithm. The spray flow is modeled by the discrete droplet method (DDM), the droplet breakup is modeled by the WAVE model, and the primary breakup is modeled by the DIESEL BREAK UP model. In order to generate cavitation phenomenon inside orifices and investigate its effect on spray characteristics, the angle of orifice with respect to the injector body is varied and the problem is studied for different angles of orifice.

  • 15.
    Ghorbani, Morteza
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Mechatronics Engineering Program, Faculty of Engineering and Natural Science, Sabanci University, 34956 Tuzla, Istanbul, Turkey.
    Araz, Sheybani Aghdam
    Faculty of Engineering and Natural Sciences, Sabanci University, Tuzla, Istanbul 34956, Turkey.
    Talebian, Moein
    Mechatronics Engineering Program, Faculty of Engineering and Natural Science, Sabanci University, 34956 Tuzla, Istanbul, Turkey.
    Kosar, Ali
    Mechatronics Engineering Program, Faculty of Engineering and Natural Science, Sabanci University, 34956 Tuzla, Istanbul, Turkey ; Sabanci University Nanotechnology Research and Application Center, 34956 Tuzla, Istanbul, Turkey ; Center of Excellence for Functional Surfaces and Interfaces for Nano-Diagnostics (EFSUN), Sabanci University, Orhanli, 34956 Tuzla, Istanbul, Turkey.
    Cakmak Cebeci, Fevzi
    Faculty of Engineering and Natural Sciences, Sabanci University, Tuzla, Istanbul 34956, Turkey ; Sabanci University Nanotechnology Research and Application Center, 34956 Tuzla, Istanbul, Turkey.
    Grishenkov, Dmitry
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Svagan, Anna Justina
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Fibre- and Polymer Technology.
    Facile Hydrodynamic Cavitation ON CHIP via Cellulose Nanofibers Stabilized Perfluorodroplets inside Layer-by-Layer Assembled SLIPS Surfaces2019In: Chemical Engineering Journal, ISSN 1385-8947, E-ISSN 1873-3212Article in journal (Refereed)
    Abstract [en]

    The tremendous potential of “hydrodynamic cavitation on microchips” has been highlighted during recent years in various applications. Cavitating flow patterns, substantially depending upon thermophysical and geometrical characteristics, promote diverse industrial and engineering applications, including food and biomedical treatment. Highly vaporous and fully developed patterns in microfluidic devices are of particular interest. In this study, the potential of a new approach, which includes cellulose nanofiber (CNF)- stabilized perfluorodroplets (PFC5s), was assessed inside microfluidic devices. The surfaces of these devices were modified by assembling various sizes of silica nanoparticles, which facilitated in the generation of cavitation bubbles. To examine the pressure effects on the stabilized droplets in the microfluidic devices, the upstream pressure was varied, and the cavitation phenomenon was characterized under different experimental conditions. The results illustrate generation of interesting, fully developed, cavitating flows at low pressures for the stabilized droplets, which has not been previously observed in the literature. Supercavitation flow pattern, filling the entire microchannel, were recorded at the upstream pressure of 1.7 MPa for the case of CNF-stabilized PFC5s, which hardly corresponds to cavitation inception for pure water in the same microfluidic device.

    The full text will be freely available from 2021-09-13 13:48
  • 16. Ghorbani, Morteza
    et al.
    Chen, Hongjian
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems.
    Villanueva, Luis Guillermo
    Grishenkov, Dmitry
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Kocsar, Ali
    Intensifying cavitating flows in microfluidic devices with poly(vinyl alcohol) (PVA) microbubbles2018In: Physics of Fluids, Vol. 30, no 10Article in journal (Refereed)
    Abstract [en]

    Cavitation and the energy associated with the collapse of resulting cavitation bubbles constitute an important research subject. The collapse of the hydrodynamic cavitation bubbles at the outlet of the flow elements leads to a high energy release and generates localized shock waves and a large temperature rise on exposed surfaces. The concept of “hydrodynamic cavitation on chip” is an emerging topic which emphasizes phase change phenomena in microscale and their utilizations in energy and biomedical applications. This study is aimed to investigate the potential of poly(vinyl alcohol) (PVA) Microbubbles (MBs) to generate cavitation bubbles and to evaluate their effects on flow regimes and energy dissipation. For this, three different microchannel configurations with different roughness elements were considered. The structural side wall and surface roughened channels were fabricated along with the smooth channel according to the techniques adopted from semiconductor based microfabrication. The upstream pressure varied from 1 to 7 MPa, and the flow patterns were recorded and analyzed using a high-speed camera. The pressure was locally measured at three locations along the microfluidic devices to determine the conditions for fully developed cavitating flows. The results were compared to the pure water case, and different trends for the cavitating flow pattern transitions were obtained for the water-PVA MB solution case. Accordingly, the twin cavity clouds extended to the end of the side wall roughened channel at a lower upstream pressure for the case of PVA MBs, while the smooth and surface roughened channels do not demonstrate this flow pattern. In addition, the cavitation number has the lowest values under the same working conditions for the case of PVA MBs. Moreover, the impact pressure generated by the bubble collapse inside the side wall roughened channel for the case of PVA MBs was notably higher than that for pure water.

  • 17.
    Ghorbani, Morteza
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Deprem, G.
    Ozdemir, E.
    Motezakker, A. R.
    Villanueva, L. G.
    Kosar, A.
    On ``Cavitation on Chip'' in Microfluidic Devices With Surface and Sidewall Roughness Elements2019In: Journal of microelectromechanical systems, ISSN 1057-7157, E-ISSN 1941-0158, Vol. 28, no 5, p. 890-899Article in journal (Refereed)
    Abstract [en]

    In this paper, cavitating flows are characterized in 29 microfluidic devices in order to achieve a comprehensive perspective regarding flow patterns in microscale, which is crucial in the applications, such as energy harvesting and biomedical treatment. While the assessment of size effects is vital for the design and development of microfluidic devices involving phase change, surface/sidewall roughness and pressure pulses as a result of nanomechanical oscillations increase the performance with respect to cavitation by providing more cavitation bubbles. A typical device generates cavitating flows under different conditions (from inception to choked flow). In this device, a restrictive element and a big channel downstream of the restrictive element--where the cavitation is formed and developed--are included. The cavitating flows are obtained inside 24 sidewall roughened and 5 surface roughened/plain microfluidic devices at different pressure drops. The length and height of the sidewall roughness elements are varied to achieve the most optimum performance in terms of cavitation generation. Moreover, different surface roughened and plain devices are considered to provide a comprehensive overview of cavitation generation in microscale. The results show that sidewall roughness elements have a remarkable effect on the cavitation inception and flow patterns. [2019-0025] IEEE

  • 18.
    Ghorbani, Morteza
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Sabanci Univ, Fac Engn & Nat Sci, Mechatron Engn Program, TR-34956 Istanbul, Turkey.
    Olofsson, Karl
    KTH, School of Engineering Sciences (SCI), Applied Physics.
    Benjamins, Jan-Willem
    Research Institute of Sweden (RISE), Chemistry, Materials and Surfaces, Box 5607, SE-114 86 Stockholm, Sweden.
    Loskutova, Ksenia
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Paulraj, Thomas
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Fibre- and Polymer Technology, Polymeric Materials.
    Wiklund, Martin
    KTH, School of Engineering Sciences (SCI), Applied Physics, Biomedical and X-ray Physics.
    Grishenkov, Dmitry
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Svagan, Anna Justina
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Fibre- and Polymer Technology.
    Unravelling the Acoustic and Thermal Responses of Perfluorocarbon Liquid Droplets Stabilized with Cellulose Nanofibers2019In: Langmuir, ISSN 0743-7463, E-ISSN 1520-5827, Vol. 35, no 40, p. 13090-13099Article in journal (Refereed)
    Abstract [en]

    The attractive colloidal and physicochemical properties of cellulose nanofibers (CNFs) at interfaces have recently been exploited in the facile production of a number of environmentally benign materials, e.g. foams, emulsions, and capsules. Herein, these unique properties are exploited in a new type of CNF-stabilized perfluoropentane droplets produced via a straightforward and simple mixing protocol. Droplets with a comparatively narrow size distribution (ca. 1–5 μm in diameter) were fabricated, and their potential in the acoustic droplet vaporization process was evaluated. For this, the particle-stabilized droplets were assessed in three independent experimental examinations, namely temperature, acoustic, and ultrasonic standing wave tests. During the acoustic droplet vaporization (ADV) process, droplets were converted to gas-filled microbubbles, offering enhanced visualization by ultrasound. The acoustic pressure threshold of about 0.62 MPa was identified for the cellulose-stabilized droplets. A phase transition temperature of about 22 °C was observed, at which a significant fraction of larger droplets (above ca. 3 μm in diameter) were converted into bubbles, whereas a large part of the population of smaller droplets were stable up to higher temperatures (temperatures up to 45 °C tested). Moreover, under ultrasound standing wave conditions, droplets were relocated to antinodes demonstrating the behavior associated with the negative contrast particles. The combined results make the CNF-stabilized droplets interesting in cell-droplet interaction experiments and ultrasound imaging.

  • 19.
    Ghorbani, Morteza
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Svagan, Anna Justina
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Fibre- and Polymer Technology.
    Grishenkov, Dmitry
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. KTH, School of Engineering Sciences (SCI), Aeronautical and Vehicle Engineering, Marcus Wallenberg Laboratory MWL. Karolinska Institutet (KI), CLINTEC – Division of Medical Imaging and Technology.
    Acoustic Response of a Novel Class of Pickering Stabilized Perfluorodroplets2019Conference paper (Refereed)
    Abstract [en]

    Introduction

    Acoustic Droplet Vaporization (ADV) is a phase change phenomenon in which the liquid state, in the form of droplets, is converted to gas as a result of bursts in the excited ultrasound field. Having a wide range of medical applications, ADV has drawn considerable attention in imaging [1], diagnosis and critical medical treatment [2]. Therefore, benefitting from its broad potentials, with the consideration of its capability in localized noninvasive energy exposure, it is possible to utilize its effect in different medical applications from targeted drug delivery [3] to embolotherapy [4].

    Apart from the droplet characterization and ADV effectiveness on the applied region, the physics of ADV and particularly the ultrasound analysis is an essential parameter in the initiation of the vaporization. This part, which is related to acoustic wave physics, implies that ADV is mostly dependent on ultrasound pressure, frequency and temperature. In this sense, Miles et al. [5] tried to find incident negative pressure - called as ADV threshold- which is necessary for the induction of nucleation. It was successfully shown that the negative pressure required for the nucleation prior to collapse can be determined via perturbation analysis of a compressible inviscid flow around a droplet for various frequencies and diameters. In addition, the fluid medium which constitutes the droplet emulsion and the surrounding fluid constructs a significant field within ADV. In this regard, there are many studies which illustrated that the diameter of the droplets subjected to the acoustic waves undergoes a significant expansion of 5 to 6 times of their regular sizes [6-8].

    In this study, a new type of pickering stabilized perfluorodroplets (PFC) was examined under the effect of the different acoustic parameters to evaluate its potential in the acoustic droplet vaporization process. To assess the pressure effects on the stabilized droplets, the acoustic power within the ultrasound tests was varied and the phase trasnition was characterized according to the experimental conditions. Opticell® was utilized as the transparent device to visualize the droplets, which were exposed to the acoustic waves with the aid of the microscope and multi-well microplate.

    Methods

    Materials and emulsion preparation

    Perfluoropentane (PFC5) was purchased from Apollo Scientific (City, U.K.). Bleached sulfite pulp (from Nordic Paper Seffle AB, Sweden) was used in the production of the cationic cellulose nanofibers (CNFs). The CNF suspension (1.32 wt%) were prepared as described previously [9]. The CNFs had a dimension of 3.9 ± 0.8 nm in width and a length in the micrometer range. The amount of cationic groups was 0.13 mmol per g fiber, obtained from conductometric titration [9]. A suspension of CNF (0.28 wt%) was prepared by diluting the stock CNF with MilliQ-water (pH of diluted CNF suspension was 9.5). The suspension was treated with ultra-sonication at amplitude of 90% for 180 s (Sonics, Vibracell W750). The suspension was brought to room temperature. An amount of 36 g of the 0.28 wt% CNF suspension was mixed with 1 g of PFC5. The mixture was sonicated for 60s at an amplitude of 80% (under ice-cooling) to obtain the CNF-stabilized PFC5 droplets.

    The protocol for the acoustic tests

    100 μL of CNF-stabilized PFC5 droplets were added to 1900 μL of deionized water in order to prepare the solution which were exposed to the ultrasound waves. The droplet sample, diluted 1:19 in distilled water was introduced to the Opticell® and the acoustic waves at a fixed frequency and different powers were applied to the trageted area inside the Opticell® which is located inside a water bath. The ultrasound triggered sample then was placed under a 20X magnification objective of upright transmitted light microscope (ECLIPSE Ci-S, Nikon, Tokyo, Japan). 

    The acoustic tests were performed using high-power tone burst pulser-receiver (SNAP Mark IV,  Ritec, Inc., Warwick, RI, USA) equipped with a transducer (V382-SU Olympus NDT, Waltham, MA ) operating at the frequency of 3.5 MHz. The emulsion of CNF-stabilized PFC5 droplets were exposed to the power range which has the acsending trend from -30 to 0 dB at the given frequency. To investigate the droplet size variations at each power between, the droplets were collected inside the Opticell® and the droplet diameter was measured with the aid of the ImageJ software (version 1.50b, National institutes of health, USA) to determine the concentration and size distribution. The Gaussian distribution is ploted with mean value and standad deviation recover from the experimental data. An in-house image edge detection MATLAB™ script (MathWorks Inc., Natick, MA) were applied to analyze the images obtained from the microscope and provides the size and volume distributions.

    Results

    The size of PFP droplets is an important parameter to controll in the therapeutic applications. Here, a new type of Pickering stabilized perfluorodroplets were prepared where the PFP/water interface was stabilized with cellulose nanofibers (CNF) and the size of the droplets could easily be controlled by varying the amount of CNF added.  The resulting droplets were investigated using a single crystal transducer. Apart from the medical applications, controlling the droplet size is important from droplet dynamics point of view, becausethe interfacial energy is crucial in the assumption of the critical nucleus radius. Therefore, it is possible to estimate the negative peak pressure required for the phase transition once the droplet is controlled and interfacial energy deposited inside and on the surface of the droplet are balanced.

    According to the results in Figure 1, there is an appreciable rise of the size of the droplets after ultrasound waves exposure, particularly at -8 dB power. The experiments were performed for 30 seconds at different powers ranging from -30 to 0 dB, while the frequency was kept constant at 3.5 MHz, burst width in cycles was selected as 12 and repetition rate was set to 100. Images included in Figure 1 demonstrate major transitions in the intervals at -16, -8 and 0 dB. As shown in this figure, the droplet size increased with the power rise and more bubbles with bigger sizes appears at higher powers. This outcome implies the significant role of the applied frequency and power on the phase shift and subsequent mechanisms as a result of the acoustic wave exposure on the new nontoxic and incompatible droplet type.

    Figure 2 shows the average number of droplets and volume distribution at the corresponding powers to the Figure 1. It is shown that while the average diameter of the droplets is around 3.5 µm, the generated bubbles, as a result of the ADV, reaches up to 15 µm at the highest possible power. For each set of experiment (corresponding to a given power) 32 images were taken, thus, to reduce the errors and obtain the standard deviation (approximately 0.8 for all the cases), the presented diagrams for the droplet distributions exhibits the mean value for all of the acquired images. Therefore, it is shown that the droplet emulsion exhibited in NO US in Figure 2, which shows the regular view and distribution range of the CNF-stabilized PFC5 droplets at the room temperature, experiences ADV process with the diameter rise of about 5 times at the highest power when the frequency is fixed at 3.5 MHz.

    Conclusions

    The results show that there is appreciable rise on the size of the droplets after ultrasound waves exposure at a fixed frequency. Acoustic droplet vaporization (ADV) was illustrated at different powers for CNF-stabilized PFC5 droplets as a new class of pickering stabilized perfluorodroplets with the increase in the size of the droplets and following phase trasition to bubbles. Diameter increase of 5 times were obtained after the ultrasound exposure indicating the efficiency of the suggested droplets for the ADV process and therapeutic applications.   

    References

    [1] Arena CB, Novell A, Sheeran PS, Puett C, Moyer LC, Dayton PA, Dual-Frequency Acoustic Droplet Vaporization Detection for Medical Imaging 2015, IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control, 62: 9.

    [2] Kripfgans OD, Fowlkes JB, Miller DL, Eldevik OP, Carson PL, Acoustic droplet vaporization for therapeutic and diagnostic applications 2000, Ultrasound Med. Biol, 26:1177–1189.

    [3] Kang ST, Yeh CK, Intracellular Acoustic Droplet Vaporization in a Single Peritoneal Macrophage for Drug Delivery Applications 2011, Langmuir, 27:13183–13188.

    [4] Zhu M, Jiang L, Fabiilli ML, Zhang A, Fowlkes JB, Xu LX, Treatment of murine tumors using acoustic droplet vaporization-enhanced high intensity focused 2013, Ultrasound Phys. Med. Biol, 58:6179–6191.

    [5] Miles CJ, Doering CR, Kripfgans OD, Nucleation pressure threshold in acoustic droplet vaporization 2016, Journal of Applied Physics, 120:034903.

    [6] Sheeran PS, Wong VP, Luois S, McFarland RJ, Ross WD, Feingold S, Matsunaga TO, Dayton PA, Decafluorobutane as a phase-change contrast agent for low-energy extravascular ultrasonic imaging 2011, Ultrasound Med. Biol, 37:1518–1530.

    [7] Kripfgans OD, Fowlkes JB, Miller DL, Eldevik OP, Carson PL, Acoustic droplet vaporization for therapeutic and diagnostic applications 2000, Ultrasound Med. Biol, 26:1177–1189.

    [8] Kang S, Huang Y, Yeh C, Characterization of acoustic droplet vaporization for control of bubble generation under flow conditions 2014, Ultrasound Med. Biol, 40:551–561.

    [9] Svagan AJ, Benjamins JW, Al-Ansari Z, Shalom DB, Müllertz A, Wågberg L, Löbmann K, Solid cellulose nanofiber based foams–towards facile design of sustained drug delivery systems 2016, J. Control Release, 244:74–82 (Part A).

     

  • 20. Hadimeri, Ursula
    et al.
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Fransson, Sven-Göran
    Stegmayr, Bernd
    Hadimeri, Henrik
    Fistula diameter correlates with echocardiographic characteristics in stable hemodialysis patients2015In: Nephrology@ Point of Care, ISSN 2059-3007, Vol. 1, no 1Article in journal (Refereed)
    Abstract [en]

    Aims and background: Left ventricular hypertrophy (LVH) is a common finding in hemodialysis patients. The aim of the present study was to investigate if the diameter of the distal radiocephalic fistula could influence left ventricular variables in stable hemodialysis patients. Methods: Nineteen patients were investigated. Measurements of the diameter of the arteriovenous (AV) fistula were performed in 4 different locations. The patients were investigated using M-mode recordings and measurements in the 2D image. Doppler ultrasound was also performed. Transonic measurements were performed after ultrasound investigation. Results: Fistula mean and maximal diameter correlated with left ventricular characteristics. Fistula flow correlated neither with the left ventricular characteristics nor with fistula diameters. Conclusions: The maximal diameter of the distal AV fistula seems to be a sensitive marker of LVH in stable hemodialysis patients.

  • 21. Holstensson, Maria
    et al.
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Poludniowski, Gavin
    Sanches Crespo, Alejandro
    Savitcheva, Irina
    Öberg, Michael
    Grybäck, Per
    Gabrielson, Stefan
    Sandqvist, Patricia
    Bartholdson, Erika
    Axelsson, Rimma
    Comparison of acquisition protocols for ventilation/perfusion SPECT - a Monte Carlo study.2019In: Physics in Medicine and Biology, ISSN 0031-9155, E-ISSN 1361-6560Article in journal (Refereed)
    Abstract [en]

    One of the most commonly used imaging techniques for diagnosing pulmonary embolism (PE) is ventilation/perfusion (V/P) scintigraphy. The aim of this study was to evaluate the performance of the currently used imaging protocols for V/P single photon emission computed tomography (V/P SPECT) at two nuclear medicine department sites and to investigate the effect of altering important protocol parameters. 
 
 The Monte Carlo technique was used to simulate 4D digital phantoms with perfusion defects. Six imaging protocols were included in the study and a total of 72 digital patients were simulated. Six dually trained radiologists/nuclear medicine physicians reviewed the images and reported all perfusion mismatch findings. The radiologists also visually graded the image quality. 
 
 No statistically significant differences in diagnostic performance were found between the studied protocols, but visual grading analysis pointed out one protocol as significantly superior to four of the other protocols. Considering the study results, we have decided to harmonize our clinical protocols for imaging patients with suspected PE. The administered Technegas and macro aggregated albumin activities have been altered, a low energy all purpose collimator is used instead of a low energy high resolution collimator and the acquisition times have been lowered.

  • 22.
    Jensen, Kristin
    et al.
    Oslo Univ Hosp, Dept Diagnost Phys, N-0454 Oslo, Norway.;Univ Oslo, Dept Phys, POB 1048 Blindern, N-0316 Oslo, Norway.;Oslo & Akershus Univ Coll Appl Sci, Dept Life Sci & Hlth, POB 4 St Olavs Plass, N-0130 Oslo, Norway..
    Andersen, Hilde Kjernlie
    Oslo Univ Hosp, Dept Diagnost Phys, N-0454 Oslo, Norway..
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Osteras, Bjorn Helge
    Oslo Univ Hosp, Dept Diagnost Phys, N-0454 Oslo, Norway.;Univ Oslo, Inst Clin Med, Oslo, Norway..
    Aarsnes, Anette
    Oslo Univ Hosp, Dept Diagnost Phys, N-0454 Oslo, Norway..
    Tingberg, Anders
    Lund Univ, Dept Med Radiat Phys, Skane Univ Hosp, Malmo, Sweden..
    Fosse, Erik
    Univ Oslo, Inst Clin Med, Oslo, Norway.;Natl Hosp Norway, Intervent Ctr, Oslo, Norway..
    Martinsen, Anne Catrine
    Oslo Univ Hosp, Dept Diagnost Phys, N-0454 Oslo, Norway.;Univ Oslo, Dept Phys, POB 1048 Blindern, N-0316 Oslo, Norway..
    Quantitative Measurements Versus Receiver Operating Characteristics and Visual Grading Regression in CT Images Reconstructed with Iterative Reconstruction: A Phantom Study2018In: Academic Radiology, ISSN 1076-6332, E-ISSN 1878-4046, Vol. 25, no 4, p. 509-518Article in journal (Refereed)
    Abstract [en]

    Rationale and Objectives: This study aimed to evaluate the correlation of quantitative measurements with visual grading regression (VGR) and receiver operating characteristics (ROC) analysis in computed tomography (CT) images reconstructed with iterative reconstruction. Materials and Methods: CT scans on a liver phantom were performed on CT scanners from GE, Philips, and Toshiba at three dose levels. Images were reconstructed with filtered back projection (FBP) and hybrid iterative techniques (ASiR, iDose, and AIDR 3D of different strengths). Images were visually assessed by five readers using a four- and five-grade ordinal scale for liver low contrast lesions and for 10 image quality criteria. The results were analyzed with ROC and VGR. Standard deviation, signal-to-noise ratios, and contrast to-noise ratios were measured in the images. Results: All data were compared to FBP. The results of the quantitative measurements were improved for all algorithms. ROC analysis showed improved lesion detection with ASiR and AIDR and decreased lesion detection with iDose. VGR found improved noise properties for all algorithms, increased sharpness with iDose and AIDR, and decreased artifacts from the spine with AIDR, whereas iDose increased the artifacts from the spine. The contrast in the spine decreased with ASiR and iDose. Conclusions: Improved quantitative measurements in images reconstructed with iterative reconstruction compared to FBP are not equivalent to improved diagnostic image accuracy.

  • 23.
    Jörgens, Daniel
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Poulin, Philippe
    Moreno, Rodrigo
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Jodoin, Pierre-Marc
    Descoteaux, Maxime
    Towards a deep learning model for diffusion-aware tractogram filtering2019Conference paper (Refereed)
  • 24.
    Jörgens, Daniel
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Moreno, Rodrigo
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Learning a single step of streamline tractography based on neural networks2018In: Computational Diffusion MRI, Springer, Cham , 2018, p. 103-116Chapter in book (Other academic)
  • 25.
    Klintström, Eva
    et al.
    Linköping Univ, Dept Med & Hlth Sci, Campus US, S-58185 Linköping, Sweden.;Linkoping Univ, Ctr Med Image Sci & Visualizat CMIV, Campus US, S-58185 Linkoping, Sweden..
    Klintström, Benjamin
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH).
    Pahr, Dieter
    Vienna Univ Technol, Inst Lightweight Design & Struct Biomech, Vienna, Austria..
    Brismar, Torkel B.
    Karolinska Univ Hosp, Karolinska Inst, Dept Clin Sci Intervent & Technol, Stockholm, Sweden.;Karolinska Univ Hosp, Dept Radiol, Stockholm, Sweden..
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Linköping Univ, Dept Med & Hlth Sci, Linköping, Sweden..
    Moreno, Rodrigo
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Direct estimation of human trabecular bone stiffness using cone beam computed tomography2018In: Oral surgery, oral medicine, oral pathology and oral radiology, ISSN 2212-4403, E-ISSN 2212-4411, Vol. 126, no 1, p. 72-82Article in journal (Refereed)
    Abstract [en]

    Objectives. The aim of this study was to evaluate the possibility of estimating the biomechanical properties of trabecular bone through finite element simulations by using dental cone beam computed tomography data. Study Design. Fourteen human radius specimens were scanned in 3 cone beam computed tomography devices: 3-D Accuitomo 80 (J. Morita MFG., Kyoto, Japan), NewTom 5 G (QR Verona, Verona, Italy), and Verity (Planmed, Helsinki, Finland). The imaging data were segmented by using 2 different methods. Stiffness (Young modulus), shear moduli, and the size and shape of the stiffness tensor were studied. Corresponding evaluations by using micro-CT were regarded as the reference standard. Results. The 3-D Accuitomo 80 (J. Morita MFG., Kyoto, Japan) showed good performance in estimating stiffness and shear moduli but was sensitive to the choice of segmentation method. Newtom 5 G (QR Verona, Verona, Italy) and Verity (Planmed, Helsinki, Finland) yielded good correlations, but they were not as strong as Accuitomo 80 U. Morita MFG., Kyoto, Japan). The cone beam computed tomography devices overestimated both stiffness and shear compared with the micro-CT estimations. Conclusions. Finite element-based calculations of biomechanics from cone beam computed tomography data are feasible, with strong correlations for the Accuitomo 80 scanner a. Morita MFG., Kyoto, Japan) combined with an appropriate segmentation method. Such measurements might be useful for predicting implant survival by in vivo estimations of bone properties.

  • 26.
    Loskutova, Ksenia
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Grishenkov, Dmitry
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Ghorbani, Morteza
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Mechatronics Engineering Program, Faculty of Engineering and Natural Science, Sabanci University, Istanbul 34956, Turkey.
    Review on Acoustic Droplet Vaporization in Ultrasound Diagnostics and Therapeutics2019In: BioMed Research International, ISSN 2314-6133, E-ISSN 2314-6141, article id 9480193Article, review/survey (Refereed)
    Abstract [en]

    Acoustic droplet vaporization (ADV) is the physical process in which liquid undergoes phase transition to gas after exposure to a pressure amplitude above a certain threshold. In recent years, new techniques in ultrasound diagnostics and therapeutics have been developed which utilize microformulations with various physical and chemical properties. The purpose of this review is to give the reader a general idea on how ADV can be implemented for the existing biomedical applications of droplet vaporization. In this regard, the recent developments in ultrasound therapy which shed light on the ADV are considered. Modern designs of capsules and nanodroplets (NDs) are shown, and the material choices and their implications for function are discussed. The influence of the physical properties of the induced acoustic field, the surrounding medium, and thermophysical effects on the vaporization are presented. Lastly, current challenges and potential future applications towards the implementation of the therapeutic droplets are discussed.

  • 27.
    Mahbod, A.
    et al.
    Romania.
    Ellinger, I.
    Romania.
    Ecker, R.
    Romania.
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Wang, Chunliang
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Breast Cancer Histological Image Classification Using Fine-Tuned Deep Network Fusion2018In: 15th International Conference on Image Analysis and Recognition, ICIAR 2018, Springer, 2018, p. 754-762Conference paper (Refereed)
    Abstract [en]

    Breast cancer is the most common cancer type in women worldwide. Histological evaluation of the breast biopsies is a challenging task even for experienced pathologists. In this paper, we propose a fully automatic method to classify breast cancer histological images to four classes, namely normal, benign, in situ carcinoma and invasive carcinoma. The proposed method takes normalized hematoxylin and eosin stained images as input and gives the final prediction by fusing the output of two residual neural networks (ResNet) of different depth. These ResNets were first pre-trained on ImageNet images, and then fine-tuned on breast histological images. We found that our approach outperformed a previous published method by a large margin when applied on the BioImaging 2015 challenge dataset yielding an accuracy of 97.22%. Moreover, the same approach provided an excellent classification performance with an accuracy of 88.50% when applied on the ICIAR 2018 grand challenge dataset using 5-fold cross validation.

  • 28. Mahbod, A.
    et al.
    Schaefer, G.
    Ellinger, I.
    Ecker, R.
    Pitiot, A.
    Wang, Chunliang
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Fusing fine-tuned deep features for skin lesion classification2019In: Computerized Medical Imaging and Graphics, ISSN 0895-6111, E-ISSN 1879-0771, Vol. 71, p. 19-29Article in journal (Refereed)
    Abstract [en]

    Malignant melanoma is one of the most aggressive forms of skin cancer. Early detection is important as it significantly improves survival rates. Consequently, accurate discrimination of malignant skin lesions from benign lesions such as seborrheic keratoses or benign nevi is crucial, while accurate computerised classification of skin lesion images is of great interest to support diagnosis. In this paper, we propose a fully automatic computerised method to classify skin lesions from dermoscopic images. Our approach is based on a novel ensemble scheme for convolutional neural networks (CNNs) that combines intra-architecture and inter-architecture network fusion. The proposed method consists of multiple sets of CNNs of different architecture that represent different feature abstraction levels. Each set of CNNs consists of a number of pre-trained networks that have identical architecture but are fine-tuned on dermoscopic skin lesion images with different settings. The deep features of each network were used to train different support vector machine classifiers. Finally, the average prediction probability classification vectors from different sets are fused to provide the final prediction. Evaluated on the 600 test images of the ISIC 2017 skin lesion classification challenge, the proposed algorithm yields an area under receiver operating characteristic curve of 87.3% for melanoma classification and an area under receiver operating characteristic curve of 95.5% for seborrheic keratosis classification, outperforming the top-ranked methods of the challenge while being simpler compared to them. The obtained results convincingly demonstrate our proposed approach to represent a reliable and robust method for feature extraction, model fusion and classification of dermoscopic skin lesion images.

  • 29. Mahbod, A.
    et al.
    Schaefer, G.
    Ellinger, I.
    Ecker, R.
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Wang, Chunliang
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    A Two-Stage U-Net Algorithm for Segmentation of Nuclei in H&E-Stained Tissues2019In: Digital Pathology: 15th European Congress, ECDP 2019, Warwick, UK, April 10–13, 2019, Proceedings / [ed] Constantino Carlos Reyes-Aldasoro, Andrew Janowczyk, Mitko Veta, Peter Bankhead, Korsuk Sirinukunwattana, Springer Verlag , 2019, p. 75-82Conference paper (Refereed)
    Abstract [en]

    Nuclei segmentation is an important but challenging task in the analysis of hematoxylin and eosin (H&E)-stained tissue sections. While various segmentation methods have been proposed, machine learning-based algorithms and in particular deep learning-based models have been shown to deliver better segmentation performance. In this work, we propose a novel approach to segment touching nuclei in H&E-stained microscopic images using U-Net-based models in two sequential stages. In the first stage, we perform semantic segmentation using a classification U-Net that separates nuclei from the background. In the second stage, the distance map of each nucleus is created using a regression U-Net. The final instance segmentation masks are then created using a watershed algorithm based on the distance maps. Evaluated on a publicly available dataset containing images from various human organs, the proposed algorithm achieves an average aggregate Jaccard index of 56.87%, outperforming several state-of-the-art algorithms applied on the same dataset.

  • 30.
    Mahbod, Amirreza
    et al.
    Med Univ Vienna, Inst Pathophysiol & Allergy Res, Vienna, Austria.;TissueGnost GmbH, Dept Res & Dev, Vienna, Austria..
    Schaefer, Gerald
    Loughborough Univ, Dept Comp Sci, Loughborough, Leics, England..
    Wang, Chunliang
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Ecker, Rupert
    TissueGnost GmbH, Dept Res & Dev, Vienna, Austria..
    Ellinger, Isabella
    Med Univ Vienna, Inst Pathophysiol & Allergy Res, Vienna, Austria..
    SKIN LESION CLASSIFICATION USING HYBRID DEEP NEURAL NETWORKS2019In: 2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), IEEE , 2019, p. 1229-1233Conference paper (Refereed)
    Abstract [en]

    Skin cancer is one of the major types of cancers with an increasing incidence over the past decades. Accurately diagnosing skin lesions to discriminate between benign and malignant skin lesions is crucial to ensure appropriate patient treatment. While there are many computerised methods for skin lesion classification, convolutional neural networks (CNNs) have been shown to be superior over classical methods. In this work, we propose a fully automatic computerised method for skin lesion classification which employs optimised deep features from a number of well-established CNNs and from different abstraction levels. We use three pre-trained deep models, namely AlexNet, VGG16 and ResNet-18, as deep feature generators. The extracted features then are used to train support vector machine classifiers. In a final stage, the classifier outputs are fused to obtain a classification. Evaluated on the 150 validation images from the ISIC 2017 classification challenge, the proposed method is shown to achieve very good classification performance, yielding an area under receiver operating characteristic curve of 83.83% for melanoma classification and of 97.55% for seborrheic keratosis classification.

  • 31. Mahbod, Amirreza
    et al.
    Wang, Chunliang
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Automatic multiple sclerosis lesion segmentation using hybrid artificial neural networks2016In: MSSEG Challenge Proceedings: Multiple Sclerosis Lesions Segmentation Challenge Using a Data Management and Processing Infrastructure, p. 29-36Article in journal (Refereed)
    Abstract [en]

    Multiple sclerosis (MS) is a demyelinating disease which could cause severe motor and cognitive deterioration. Segmenting MS lesions could be highly beneficial for diagnosing, analyzing and monitoring treatment efficacy. To do so, manual segmentation, performed by experts, is the conventional method in hospitals and clinical environments. Although manual segmentation is accurate, it is time consuming, expensive and might not be reliable. The aim of this work was to propose an automatic method for MS lesion segmentation and evaluate it using brain images available within the MICCAI MS segmentation challenge. The proposed method employs supervised artificial neural network based algorithm, exploiting intensity-based and spatial-based features as the input of the network. This method achieved relatively accurate results with acceptable training and testing time for training datasets.

  • 32.
    Marlevi, David
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Non-invasive imaging for improved cardiovascular diagnostics: Shear wave elastography, relative pressure estimation, and tomographic reconstruction2019Doctoral thesis, comprehensive summary (Other academic)
    Abstract [en]

    Throughout the last century, medical imaging has come to revolutionise the way we diagnose disease, and is today an indispensable part of virtually any clinical practice. In cardiovascular care imaging is extensively utilised, and the development of novel techniques promises refined diagnostic abilities: ultrasound elastography allows for constitutive tissue assessment, 4D flow magnetic resonance imaging (MRI) enables full-field flow mapping, and micro-Computed Tomography (CT) permits high-resolution imaging at pre-clinical level. However, following the complex nature of cardiovascular disease, refined methods are still very much needed to accurately utilise these techniques and to effectively isolate disease developments.

    The aim of this thesis has been to develop such methods for refined cardiovascular image diagnostics. In total eight studies conducted over three separate focus areas have been included: four on vascular shear wave elastography (SWE), three on non-invasive cardiovascular relative pressure estimations, and one on tomographic reconstruction for pre-clinical imaging.

    In Study I-IV, the accuracy and feasibility of vascular SWE was evaluated, with particular focus on refined carotid plaque characterisation. With confined arterial or plaque tissue restricting acoustic wave propagation, analysis of group and phase velocity was performed with SWE output validated against reference mechanical testing and imaging. The results indicate that geometrical confinement has a significant impact on SWE accuracy, however that a combined group and phase velocity approach can be utilised to identify vulnerable carotid plaque lesions in-vivo.

    In Study V-VII, a non-invasive method for the interrogation of relative pressure from imaged cardiovascular flow was developed. Using the concept of virtual work-energy, the method was applied to accurately assess relative pressures throughout complex, turbulence-inducing, branching vasculatures. The method was also applied on a dilated cardiomyopathy cohort, indicating arterial hemodynamic changes in cardiac disease.

    Lastly, in Study VIII a method for multigrid image reconstruction of tomographic data was developed, utilising domain splitting and operator masking to accurately reconstruct high-resolution regions-of-interests at a fraction of the computational cost of conventional full-resolution methods.

    Together, the eight studies have incorporated a range of different imaging modalities, developed methods for both constitutive and hemodynamic cardiovascular assessment, and utilised refined pre-clinical imaging, all with the same purpose: to refine current state cardiovascular imaging and to improve our ability to non-invasively assess cardiovascular disease. With promising results reached, the studies lay the foundation for continued clinical investigations, advancing the presented methods and maturing their usage for an improved future cardiovascular care.

  • 33.
    Marlevi, David
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Ha, Hojin
    Dillon-Murphy, Desmond
    Fernandes, Joao F
    Fovargue, Daniel
    Colarieti-Tosti, Massimiliano
    Larsson, Matilda
    Lamata, Pablo
    Figueroa, C Alberto
    Ebbers, Tino
    Nordsletten, David
    Non-invasive estimation of relative pressure in turbulent flow using virtual work-energyManuscript (preprint) (Other academic)
  • 34.
    Marlevi, David
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Kohr, Holger
    Buurlage, Jan-Willem
    Gao, Bo
    Batenburg, Joost
    Colarieti-Tosti, Massimiliano
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Multigrid reconstruction in tomographic imaging2019In: IEEE Transactions on Radiation and Plasma Medical Sciences, ISSN 2469-7311Article in journal (Refereed)
    Abstract [en]

    In this work, we present an efficient methodology for multigrid tomographic image reconstruction from non-truncated projection data. By partitioning the reconstruction domain and adapting the forward and backward operators, an image can be reconstructed accurately within multiple domains of varying discretisation or regularisation. We demonstrate the efficacy of the multigrid reconstruction principle using simulated data for quantitative assessment and experimental measurements from a μ-CT scanner for a clinically relevant use case scenario. A major advantage of using multiple reconstruction grids is the possibility to drastically reduce the number of unknowns in the inverse problem, and thereby the associated computational cost. This cost reduction helps to enlarge the class of available algorithms in applications with strict limitations on computation time or resources, and it enables full system resolution reconstruction of subregions that would otherwise be infeasible for the full field of view. The numerical experiments, along with a brief error analysis, show that the expected artefacts from coarse discretisation outside the region of interest become noticeable only for large differences in discretisation between subregions.

  • 35.
    Marlevi, David
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Kohr, Holger
    Jan-Willem, Buurlage
    Gao, Bo
    Batenburg, K Joost
    Colarieti-Tosti, Massimiliano
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Multigrid reconstruction in tomographic imagingManuscript (preprint) (Other academic)
  • 36.
    Marlevi, David
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Mariscal-Harana, Jorge
    Sotelo, Julio
    Ruijsink, Bram
    Hadjicharalambous, Myrianthi
    Asner, Liya
    Sammut, Eva
    Chabiniok, Radomir
    Uribe, Sergio
    Winter, Reidar
    Lamata, Pablo
    Alastruey, Jordi
    Nordsletten, David
    Altered aortic hemodynamics and relative pressure in patients with dilated cardiomyopathyManuscript (preprint) (Other academic)
  • 37.
    Marlevi, David
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Mulvagh, Sharon
    Huang, Runqing
    DeMarco, J Kevin
    Ota, Hideki
    Huston, John
    Winter, Reidar
    Macedo, Thanila
    Abdelmoneim, Sahar
    Larsson, Matilda
    Pellikka, Patricia
    Urban, Matthew W
    Shear wave elastography enables detection of vulnerable carotid plaques – MRI-validation of combined spatiotemporal and frequency-dependent wave analysisManuscript (preprint) (Other academic)
  • 38.
    Marlevi, David
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Karolinska Institutet, Stockholm, Sweden.
    Ruijsink, B.
    Balmus, M.
    Dillon-Murphy, D.
    Fovargue, D.
    Pushparajah, K.
    Bertoglio, C.
    Colarieti-Tosti, Massimiliano
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Karolinska Institutet, Stockholm, Sweden.
    Larsson, Matilda
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Lamata, P.
    Figueroa, C. A.
    Razavi, R.
    Nordsletten, D. A.
    Estimation of Cardiovascular Relative Pressure Using Virtual Work-Energy2019In: Scientific Reports, ISSN 2045-2322, E-ISSN 2045-2322, Vol. 9, no 1, article id 1375Article in journal (Refereed)
    Abstract [en]

    Many cardiovascular diseases lead to local increases in relative pressure, reflecting the higher costs of driving blood flow. The utility of this biomarker for stratifying the severity of disease has thus driven the development of methods to measure these relative pressures. While intravascular catheterisation remains the most direct measure, its invasiveness limits clinical application in many instances. Non-invasive Doppler ultrasound estimates have partially addressed this gap; however only provide relative pressure estimates for a range of constricted cardiovascular conditions. Here we introduce a non-invasive method that enables arbitrary interrogation of relative pressures throughout an imaged vascular structure, leveraging modern phase contrast magnetic resonance imaging, the virtual work-energy equations, and a virtual field to provide robust and accurate estimates. The versatility and accuracy of the method is verified in a set of complex patient-specific cardiovascular models, where relative pressures into previously inaccessible flow regions are assessed. The method is further validated within a cohort of congenital heart disease patients, providing a novel tool for probing relative pressures in-vivo.

  • 39. Medrano-Gracia, Pau
    et al.
    Ormiston, John
    Webster, Mark
    Beier, Susann
    Ellis, Chris
    Wang, Chunliang
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Young, Alistair
    Cowan, Brett
    A Statistical Model of the Main Bifurcation of the Left Coronary Artery using Coherent Point Drift2015Conference paper (Refereed)
  • 40.
    Mårtensson, Gustav
    et al.
    Karolinska Inst, Dept Neurobiol Care Sci & Soc, Div Clin Geriatr, Stockholm, Sweden..
    Ferreira, Daniel
    Karolinska Inst, Dept Neurobiol Care Sci & Soc, Div Clin Geriatr, Stockholm, Sweden..
    Cavallin, Lena
    Karolinska Inst, Dept Clin Neurosci, Stockholm, Sweden.;Karolinska Univ Hosp, Dept Radiol, Stockholm, Sweden..
    Muehlboeck, J-Sebastian
    Karolinska Inst, Dept Neurobiol Care Sci & Soc, Div Clin Geriatr, Stockholm, Sweden..
    Wahlund, Lars-Olof
    Karolinska Inst, Dept Neurobiol Care Sci & Soc, Div Clin Geriatr, Stockholm, Sweden..
    Wang, Chunliang
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Westman, Eric
    Karolinska Inst, Dept Neurobiol Care Sci & Soc, Div Clin Geriatr, Stockholm, Sweden.;Kings Coll London, Ctr Neuroimaging Sci, Inst Psychiat Psychol & Neurosci, Dept Neuroimaging, London, England..
    AVRA: Automatic visual ratings of atrophy from MRI images using recurrent convolutional neural networks2019In: NeuroImage: Clinical, ISSN 0353-8842, E-ISSN 2213-1582, Vol. 23, article id UNSP 101872Article in journal (Refereed)
    Abstract [en]

    Quantifying the degree of atrophy is done clinically by neuroradiologists following established visual rating scales. For these assessments to be reliable the rater requires substantial training and experience, and even then the rating agreement between two radiologists is not perfect. We have developed a model we call AVRA (Automatic Visual Ratings of Atrophy) based on machine learning methods and trained on 2350 visual ratings made by an experienced neuroradiologist. It provides fast and automatic ratings for Scheltens' scale of medial temporal atrophy (MTA), the frontal subscale of Pasquier's Global Cortical Atrophy (GCA-F) scale, and Koedam's scale of Posterior Atrophy (PA). We demonstrate substantial inter-rater agreement between AVRA's and a neuroradiologist ratings with Cohen's weighted kappa values of kappa(w) = 0.74/0.72 (MTA left/right), kappa(w) = 0.62 (GCA-F) and kappa(w) = 0.74 (PA). We conclude that automatic visual ratings of atrophy can potentially have great scientific value, and aim to present AVRA as a freely available toolbox.

  • 41.
    Nordenfur, Tim
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Karolinska Institute, Sweden.
    Babic, A.
    Bulatovic, I.
    Giesecke, A.
    Günyeli, E.
    Ripsweden, J.
    Samset, E.
    Winter, R.
    Larsson, Matilda
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Method comparison for cardiac image registration of coronary computed tomography angiography and 3-D echocardiography2018In: Journal of Medical Imaging, ISSN 2329-4302, E-ISSN 2329-4310, Vol. 5, no 1, article id 014001Article in journal (Refereed)
    Abstract [en]

    Treatment decision for coronary artery disease (CAD) is based on both morphological and functional information. Image fusion of coronary computed tomography angiography (CCTA) and three-dimensional echocardiography (3DE) could combine morphology and function into a single image to facilitate diagnosis. Three semiautomatic feature-based methods for CCTA/3DE registration were implemented and applied on CAD patients. Methods were verified and compared using landmarks manually identified by a cardiologist. All methods were found feasible for CCTA/3DE fusion.

  • 42. Pavoni, Marco
    et al.
    Chang, Yongjun
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH).
    Park, Sang-Ho
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Convolutional neural network-based image enhancement for x-ray percutaneous coronary intervention2018In: Journal of Medical Imaging, ISSN 2329-4302, E-ISSN 2329-4310, Vol. 5, no 2, article id 024006Article in journal (Refereed)
    Abstract [en]

    Percutaneous coronary intervention (PCI) uses x-ray images, which may give high radiation dose and high concentrations of contrast media, leading to the risk of radiation-induced injury and nephropathy. These drawbacks can be reduced by using lower doses of x-rays and contrast media, with the disadvantage of noisier PCI images with less contrast. Vessel-edge-preserving convolutional neural networks (CNN) were designed to denoise simulated low x-ray dose PCI images, created by adding artificial noise to high-dose images. Objective functions of the designed CNNs have been optimized to achieve an edge-preserving effect of vessel walls, and the results of the proposed objective functions were evaluated qualitatively and quantitatively. Finally, the proposed CNN-based method was compared with two state-of-the-art denoising methods: K-SVD and block-matching and 3D filtering. The results showed promising performance of the proposed CNN-based method for PCI image enhancement with interesting capabilities of CNNs for real-time denoising and contrast enhancement tasks.

  • 43.
    Qin, Chunxia
    et al.
    Shanghai Jiao Tong Univ, Sch Biomed Engn, Shanghai, Peoples R China.;Shanghai Jiao Tong Univ, Sch Mech Engn, Room 805,Dongchuan Rd 800, Shanghai 200240, Peoples R China..
    Cao, Zhenggang
    Shanghai Jiao Tong Univ, Sch Mech Engn, Room 805,Dongchuan Rd 800, Shanghai 200240, Peoples R China..
    Fan, Shengchi
    Shanghai Jiao Tong Univ, Shanghai Peoples Hosp 9, Sch Med, Shanghai, Peoples R China..
    Wu, Yiqun
    Shanghai Jiao Tong Univ, Shanghai Peoples Hosp 9, Sch Med, Shanghai, Peoples R China..
    Sun, Yi
    Katholieke Univ Leuven, Fac Med, Dept Imaging & Pathol, OMFS IMPATH Res Grp, Louvain, Belgium.;Univ Hosp Leuven, Dept Oral & Maxillofacial Surg, Louvain, Belgium..
    Politis, Constantinus
    Katholieke Univ Leuven, Fac Med, Dept Imaging & Pathol, OMFS IMPATH Res Grp, Louvain, Belgium.;Univ Hosp Leuven, Dept Oral & Maxillofacial Surg, Louvain, Belgium..
    Wang, Chunliang
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Chen, Xiaojun
    Shanghai Jiao Tong Univ, Sch Mech Engn, Room 805,Dongchuan Rd 800, Shanghai 200240, Peoples R China..
    An oral and maxillofacial navigation system for implant placement with automatic identification of fiducial points2019In: International Journal of Computer Assisted Radiology and Surgery, ISSN 1861-6410, E-ISSN 1861-6429, Vol. 14, no 2, p. 281-289Article in journal (Refereed)
    Abstract [en]

    PurposeSurgical navigation system (SNS) has been an important tool in surgery. However, the complicated and tedious manual selection of fiducial points on preoperative images for registration affects operational efficiency to large extent. In this study, an oral and maxillofacial navigation system named BeiDou-SNS with automatic identification of fiducial points was developed and demonstrated.MethodsTo solve the fiducial selection problem, a novel method of automatic localization for titanium screw markers in preoperative images is proposed on the basis of a sequence of two local mean-shift segmentation including removal of metal artifacts. The operation of the BeiDou-SNS consists of the following key steps: The selection of fiducial points, the calibration of surgical instruments, and the registration of patient space and image space. Eight cases of patients with titanium screws as fiducial markers were carried out to analyze the accuracy of the automatic fiducial point localization algorithm. Finally, a complete phantom experiment of zygomatic implant placement surgery was performed to evaluate the whole performance of BeiDou-SNS. Results and conclusionThe coverage of Euclidean distances between fiducial marker positions selected automatically and those selected manually by an experienced dentist for all eight cases ranged from 0.373 to 0.847mm. Four implants were inserted into the 3D-printed model under the guide of BeiDou-SNS. And the maximal deviations between the actual and planned implant were 1.328mm and 2.326mm, respectively, for the entry and end point while the angular deviation ranged from 1.094 degrees to 2.395 degrees. The results demonstrate that the oral surgical navigation system with automatic identification of fiducial points can meet the requirements of the clinical surgeries.

  • 44.
    Sinzinger, Fabian
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Pahr, Dieter
    Moreno, Rodrigo
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Predicting The Trabecular Bone Stiffness Tensor with Spherical Convolutional Neural Networks2019In: Book of Abstracts of the 25th Congress of the European Society of Biomechanics, 2019Conference paper (Refereed)
  • 45. Talebian Gevari, Moein
    et al.
    Ghorbani, Morteza
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    J. Svagan, Anna
    Grishenkov, Dmitry
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. KTH, School of Engineering Sciences (SCI), Aeronautical and Vehicle Engineering, Marcus Wallenberg Laboratory MWL.
    Kosar, Ali
    Energy harvesting with micro scale hydrodynamic cavitation-thermoelectric generation coupling2019In: AIP Advances, ISSN 2158-3226, E-ISSN 2158-3226Article in journal (Refereed)
  • 46. Talebian Gevari, Moein
    et al.
    Ghorbani, Morteza
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    J. Svagan, Anna
    Grishenkov, Dmitry
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. KTH, School of Engineering Sciences (SCI), Aeronautical and Vehicle Engineering, Marcus Wallenberg Laboratory MWL.
    Kosar, Ali
    Energy harvesting with micro scale hydrodynamic cavitation-thermoelectric generation coupling2019In: AIP Advances, ISSN 2158-3226, E-ISSN 2158-3226Article in journal (Refereed)
  • 47.
    Wan, Fengkai
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems. Novamia AB, Uppsala, Sweden.
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Novamia AB, Uppsala, Sweden.
    Wang, Chunliang
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Novamia AB, Uppsala, Sweden.
    Simultaneous MR knee image segmentation and bias field correction using deep learning and partial convolution2019In: Medical Imaging 2019: Image Processing, SPIE - International Society for Optical Engineering, 2019, Vol. 10949, article id 1094909Conference paper (Refereed)
    Abstract [en]

    Intensity inhomogeneity is a great challenge for automated organ segmentation in magnetic resonance (MR) images. Many segmentation methods fail to deliver satisfactory results when the images are corrupted by a bias field. Although inhomogeneity correction methods exist, they often fail to remove the bias field completely in knee MR images. We present a new iterative approach that simultaneously predicts the segmentation mask of knee structures using a 3D U-net and estimates the bias field in 3D MR knee images using partial convolution operations. First, the test images run through a trained 3D U-net to generate a preliminary segmentation result, which is then fed to the partial convolution filter to create a preliminary estimation of the bias field using the segmented bone mask. Finally, the estimated bias field is then used to produce bias field corrected images as the new inputs to the 3D U-net. Through this loop, the segmentation results and bias field correction are iteratively improved. The proposed method was evaluated on 20 proton-density (PD)-weighted knee MRI scans with manually created segmentation ground truth using 10 fold cross-validation. In our preliminary experiments, the proposed methods outperformed conventional inhomogeneity-correction-plus-segmentation setup in terms of both segmentation accuracy and speed.

  • 48.
    Wang, Chunliang
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Connolly, Bryan
    de Oliveira Lopes, Pedro Filipe
    Frangi, Alejandro F.
    Smedby, Örjan
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Pelvis segmentation using multi-pass U-Net and iterative shape estimation2018In: Computational Methods and Clinical Applications in Musculoskeletal Imaging, Springer, 2018, Vol. 11404, p. 49-57Conference paper (Refereed)
    Abstract [en]

    In this report, an automatic method for segmentation of the pelvis in three-dimensional (3D) computed tomography (CT) images is proposed. The method is based on a 3D U-net which has as input the 3D CT image and estimated volumetric shape models of the targeted structures and which returns the probability maps of each structure. During training, the 3D U-net is initially trained using blank shape context inputs to generate the segmentation masks, i.e. relying only on the image channel of the input. The preliminary segmentation results are used to estimate a new shape model, which is then fed to the same network again, with the input images. With the additional shape context information, the U-net is trained again to generate better segmentation results. During the testing phase, the input image is fed through the same 3D U-net multiple times, first with blank shape context channels and then with iteratively re-estimated shape models. Preliminary results show that the proposed multi-pass U-net with iterative shape estimation outperforms both 2D and 3D conventional U-nets without the shape model.

  • 49.
    Wang, Chunliang
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging. Linköping University, Sweden.
    Forsberg, Daniel
    Segmentation of intervertebral discs in 3D MRI data using multi-atlas based registration2015In: Computational Methods and Clinical Applications for Spine Imaging, Springer, 2015, Vol. 9402, p. 107-116Conference paper (Refereed)
    Abstract [en]

    This paper presents one of the participating methods to the intervertebral disc segmentation challenge organized in conjunction with the 3rd MICCAI Workshop & Challenge on Computational Methods and Clinical Applications for Spine Imaging - MICCAI-CSI2015. The presented method consist of three steps. In the first step, vertebral bodies are detected and labeled using integral channel features and a graphical parts model. The second step consists of image registration, where a set of image volumes with corresponding intervertebral disc atlases are registered to the target volume using the output from the first step as initialization. In the final step, the registered atlases are combined using label fusion to derive the final segmentation. The pipeline was evaluated using a set of 15 + 10 T2-weighted image volumes provided as training and test data respectively for the segmentation challenge. For the training data, a mean disc centroid distance of 0.86 mm and an average DICE score of 91% was achieved, and for the test data the corresponding results were 0.90 mm and 90%.

  • 50.
    Wang, Chunliang
    et al.
    KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Medical Imaging.
    Smedby, Örjan
    Model-based left ventricle segmentation in 3D ultrasound using phase image2014Conference paper (Refereed)
12 1 - 50 of 52
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf