Spårning av skelettala strukturer på 4D-CT
Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
The technical development of computed tomography (CT) with faster image collection frequencies and lower radiation dose while maintaining or increasing spatial and tissue-specific resolution, has enabled timeresoluted dynamic CT for the study of body parts in motion.
The development of this technology has come to the attention of the supervisors at the Karolinska Institute and the Royal Institute of Technology. They were interested to investigate the possibility of using today´s technology to calculate the kinematics of skeletal foot segments, based on a dynamic CT collection of a human foot. Thus a method to detect, extract and track landmarks in sequences of 3-dimensional (3D) volumes, and the description of foot kinematics during plantarflexion and dorsiflexion was developed in this thesis.
The developed method intended to automatically track skeletal foot segments and calculate their relative movement and rotation in the frontal, sagittal and transverse plane. A program was developed using commercially available software that isolated a selected skeletal segment and detected its landmarks. The automatic landmark detection useful for tracking needs further development and therefore complementary manual detection was performed. These landmarks were used to calculate the kinematics by tracking individually created local coordinate system for each structure. As an example the maximum rotation stated between calcaneus and talus was 51 degrees within the sagittal plane. The thesis illustrates methods used and presents the developed program and all the results of the individual and relative motion of the calcaneus and talus. The method used in this thesis has a good future potential with possibilities to visualize and calculate normal and pathological musculoskeletal motion.
Place, publisher, year, edition, pages
2013. , 22 p.
dynamisk CT, 4D, kinematik, landmärke, detektion, tracking, Matlab
IdentifiersURN: urn:nbn:se:kth:diva-150460OAI: oai:DiVA.org:kth-150460DiVA: diva2:743504
Subject / course
Master of Science in Engineering - Medical Engineering
Janerot Sjöberg, Birgitta, Professor
Lars-Gösta Hellström, Lars-Gösta, PhD