Change search
CiteExportLink to record
Permanent link

Direct 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
Machine Learning Techniques with Specific Application to the Early Olfactory System
KTH, School of Computer Science and Communication (CSC), Computational Biology, CB. (Anders Lansner)
2012 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

This thesis deals with machine learning techniques for the extraction of structure and the analysis of the vertebrate olfactory pathway based on related methods. Some of its main contributions are summarized below.

We have performed a systematic investigation for classification in biomedical images with the goal of recognizing a material in these images by its texture. This investigation included (i) different measures for evaluating the importance of image descriptors (features), (ii) methods to select a feature set based on these evaluations, and (iii) classification algorithms. Image features were evaluated according to their estimated relevance for the classification task and their redundancy with other features. For this purpose, we proposed a framework for relevance and redundancy measures and, within this framework, we proposed two new measures. These were the value difference metric and the fit criterion. Both measures performed well in comparison with other previously used ones for evaluating features. We also proposed a Hopfield network as a method for feature selection, which in experiments gave one of the best results relative to other previously used approaches.

We proposed a genetic algorithm for clustering and tested it on several realworld datasets. This genetic algorithm was novel in several ways, including (i) the use of intra-cluster distance as additional optimization criterion, (ii) an annealing procedure, and (iii) adaptation of mutation rates. As opposed to many conventional clustering algorithms, our optimization framework allowed us to use different cluster validation measures including those which do not rely on cluster centroids. We demonstrated the use of the clustering algorithm experimentally with several cluster validity measures as optimization criteria. We compared the performance of our clustering algorithm to that of the often-used fuzzy c-means algorithm on several standard machine learning datasets from the University of California/Urvine (UCI) and obtained good results.

The organization of representations in the brain has been observed at several stages of processing to spatially decompose input from the environment into features that are somehow relevant from a behavioral or perceptual standpoint. For the perception of smells, the analysis of such an organization, however, is not as straightforward because of the missing metric. Some studies report spatial clusters for several combinations of physico-chemical properties in the olfactory bulb at the level of the glomeruli. We performed a systematic study of representations based on a dataset of activity-related images comprising more than 350 odorants and covering the whole spatial array of the first synaptic level in the olfactory system. We found clustered representations for several physico-chemical properties. We compared the relevance of these properties to activations and estimated the size of the coding zones. The results confirmed and extended previous studies on olfactory coding for physico-chemical properties. Particularly of interest was the spatial progression by carbon chain that we found. We discussed our estimates of relevance and coding size in the context of processing strategies. We think that the results obtained in this study could guide the search into olfactory coding primitives and the understanding of the stimulus space.

In a second study on representations in the olfactory bulb, we grouped odorants together by perceptual categories, such as floral and fruity. By the application of the same statistical methods as in the previous study, we found clustered zones for these categories. Furthermore, we found that distances between spatial representations were related to perceptual differences in humans as reported in the literature. This was possibly the first time that such an analysis had been done. Apart from pointing towards a spatial decomposition by perceptual dimensions, results indicate that distance relationships between representations could be perceptually meaningful.

In a third study, we modeled axon convergence from olfactory receptor neurons to the olfactory bulb. Sensory neurons were stimulated by a set of biologically-relevant odors, which were described by a set of physico-chemical properties that covaried with the neural and glomerular population activity in the olfactory bulb. Convergence was mediated by the covariance between olfactory neurons. In our model, we could replicate the formation of glomeruli and concentration coding as reported in the literature, and further, we found that the spatial relationships between representational zones resulting from our model correlated with reported perceptual differences between odor categories. This shows that natural statistics, including similarity of physico-chemical structure of odorants, can give rise to an ordered arrangement of representations at the olfactory bulb level where the distances between representations are perceptually relevant.

Place, publisher, year, edition, pages
Stockholm: KTH Royal Institute of Technology, 2012. , xiv, 216 p.
Series
Trita-CSC-A, ISSN 1653-5723 ; 2012:01
Keyword [en]
feature selection, image features, pattern classification, relevance, redundancy, distributional similarity, divergence measure, genetic algorithms, clustering algorithms, annealing, olfactory coding, olfactory bulb, odorants, glomeruli, property-activity relationship, olfaction, plasticity, axonal guidance, odor category, perception, spatial coding, population coding, memory organization, odor quality
National Category
Biological Sciences
Research subject
SRA - ICT
Identifiers
URN: urn:nbn:se:kth:diva-90474ISBN: 978-91-7501-273-5 (print)OAI: oai:DiVA.org:kth-90474DiVA: diva2:505615
Public defence
2012-03-16, D3, Lindstedtsvägen 5, KTH, Stockholm, 10:00 (English)
Opponent
Supervisors
Funder
Swedish e‐Science Research Center
Note

QC 20120224

Available from: 2012-02-24 Created: 2012-02-24 Last updated: 2013-04-09Bibliographically approved
List of papers
1. Hopfield Networks in Relevance and Redundancy Feature Selection Applied to Classification of Biomedical High-Resolution Micro-CT Images
Open this publication in new window or tab >>Hopfield Networks in Relevance and Redundancy Feature Selection Applied to Classification of Biomedical High-Resolution Micro-CT Images
2008 (English)In: Advances in Data Mining: Medical Applications, E-Commerce, Marketing, and Theoretical Aspects / [ed] Petra Perner, Heidelberg: Springer , 2008, 16-31 p.Chapter in book (Refereed)
Abstract [en]

We study filter-based feature selection methods for classification of biomedical images. For feature selection, we use two filters - a relevance filter which measures usefulness of individual features for target prediction, and a redundancy filter, which measures similarity between features. As selection method that combines relevance and redundancy we try out a Hopfield network. We experimentally compare selection methods, running unitary redundancy and relevance filters, against a greedy algorithm with redundancy thresholds [9], the min-redundancy max-relevance integration [8,23,36], and our Hopfield network selection. We conclude that on the whole, Hopfield selection was one of the most successful methods, outperforming min-redundancy max-relevance when more features are selected.

Place, publisher, year, edition, pages
Heidelberg: Springer, 2008
Series
Lecture notes in artificial intelligence, ISSN 0302-9743 ; 5077
Keyword
feature selection, image features, pattern classification
National Category
Engineering and Technology
Research subject
SRA - ICT
Identifiers
urn:nbn:se:kth:diva-48062 (URN)10.1007/978-3-540-70720-2_2 (DOI)000258494700002 ()
Conference
8th Industrial Conference on Data Mining. Leipzig, GERMANY. JUL 16-18, 2008
Note

QC 20111116

Available from: 2011-11-15 Created: 2011-11-15 Last updated: 2017-03-30Bibliographically approved
2. Comparison of Redundancy and Relevance Measures for Feature Selection in Tissue Classification of CT images
Open this publication in new window or tab >>Comparison of Redundancy and Relevance Measures for Feature Selection in Tissue Classification of CT images
2010 (English)In: Advances in Data Mining: Applications in Medicine, Web Mining, Marketing, Image and Signal Mining / [ed] Petra Perner, Heidelberg: Springer Berlin/Heidelberg, 2010, 248-262 p.Chapter in book (Refereed)
Abstract [en]

In this paper we report on a study on feature selection within the minimum-redundancy maximum-relevance framework. Features are ranked by their correlations to the target vector. These relevance scores are then integrated with correlations between features in order to ob- tain a set of relevant and least-redundant features. Applied measures of correlation or distributional similarity for redundancy and relevance include Kolmogorov-Smirnov (KS) test, Spearman correlations, Jensen-Shannon divergence, and the sign-test. We introduce a metric called “value difference metric“ (VDM) and present a simple measure, which we call “fit criterion“ (FC). We draw conclusions about the usefulness of different measures. While KS-test and sign-test provided useful information, Spearman correlations are not fit for comparison of data of different measurement intervals. VDM was very good in our experiments as both redundancy and relevance measure. Jensen-Shannon and the sign-test are good redundancy measure alternatives and FC is a good relevance measure alternative.

Place, publisher, year, edition, pages
Heidelberg: Springer Berlin/Heidelberg, 2010
Series
Lecture Notes in Artificial Intelligence, ISSN 0302-9743 ; 6171
Keyword
distributional similarity; divergence measure; feature selection; relevance and redundancy
National Category
Signal Processing
Research subject
SRA - ICT
Identifiers
urn:nbn:se:kth:diva-48064 (URN)10.1007/978-3-642-14400-4_20 (DOI)000286902300020 ()
Conference
10th Industrial Conference on Data Mining. Berlin, GERMANY. JUL 12-14, 2010
Note

QC 20111116

Available from: 2011-11-15 Created: 2011-11-15 Last updated: 2016-08-09Bibliographically approved
3. Clustering by a genetic algorithm with biased mutation operator
Open this publication in new window or tab >>Clustering by a genetic algorithm with biased mutation operator
2010 (English)In: 2010 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC), IEEE , 2010, 1-8 p.Conference paper, Published paper (Refereed)
Abstract [en]

In this paper we propose a genetic al- gorithm that partitions data into a given number of clusters. The algorithm can use any cluster validity function as fitness function. Cluster validity is used as a criterion for cross-over operations. The cluster assignment for each point is accompanied by a tem- perature and points with low confidence are pref- erentially mutated. We present results applying this genetic algorithm to several UCI machine learning data sets and using several objective cluster validity functions for optimization. It is shown that given an appropriate criterion function, the algorithm is able to converge on good cluster partitions within few generations. Our main contributions are: 1. to present a genetic algorithm that is fast and able to converge on meaningful clusters for real-world data sets, 2. to define and compare several cluster validity criteria. 

Place, publisher, year, edition, pages
IEEE, 2010
Keyword
learning (artificial intelligence), pattern clustering, UCI machine learning, cluster validity function, criterion function, crossover operation, fitness function, genetic algorithm, mutation operator, optimization
National Category
Signal Processing
Research subject
SRA - ICT
Identifiers
urn:nbn:se:kth:diva-48065 (URN)10.1109/CEC.2010.5586090 (DOI)000287375801062 ()
Conference
2010 IEEE World Congress on Computational Intelligence. Barcelona, SPAIN. JUL 18-23, 2010
Note
© 2010 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. QC 20111115Available from: 2012-02-20 Created: 2011-11-15 Last updated: 2012-03-12Bibliographically approved
4. Continuous Spatial Representations in the Olfactory Bulb may Reflect Perceptual Categories
Open this publication in new window or tab >>Continuous Spatial Representations in the Olfactory Bulb may Reflect Perceptual Categories
2011 (English)In: Frontiers in Neuroscience, ISSN 1662-4548, E-ISSN 1662-453X, Vol. 5, no 82Article in journal (Refereed) Published
Abstract [en]

In sensory processing of odors, the olfactory bulb is an important relay station, where odor representations are noise-filtered, sharpened, and possibly re-organized. An organization by perceptual qualities has been found previously in the piriform cortex, however several recent studies indicate that the olfactory bulb code reflects behaviorally relevant dimensions spatially as well as at the population level. We apply a statistical analysis on 2-deoxyglucose images, taken over the entire bulb of glomerular layer of the rat, in order to see how the recognition of odors in the nose is translated into a map of odor quality in the brain. We first confirm previous studies that the first principal component could be related to pleasantness, however the next higher principal components are not directly clear. We then find mostly continuous spatial representations for perceptual categories. We compare the space spanned by spatial and population codes to human reports of perceptual similarity between odors and our results suggest that perceptual categories could be already embedded in glomerular activations and that spatial representations give a better match than population codes. This suggests that human and rat perceptual dimensions of odorant coding are related and indicates that perceptual qualities could be represented as continuous spatial codes of the olfactory bulb glomerulus population.

Place, publisher, year, edition, pages
Frontiers Research Foundation, 2011
Keyword
olfaction, olfactory bulb, glomeruli, spatial coding, population coding, memory organization, odor quality, perception
National Category
Natural Sciences
Research subject
SRA - ICT
Identifiers
urn:nbn:se:kth:diva-48052 (URN)10.3389/fnsys.2011.00082 (DOI)2-s2.0-84856173464 (Scopus ID)
Projects
Neurochem
Note
This Document is Protected by copyright and was first published by Frontiers. All rights reserved. It is reproduced with permission QC 20111116Available from: 2011-11-16 Created: 2011-11-15 Last updated: 2017-12-08Bibliographically approved
5. Statistical analysis of coding for molecular properties in the olfactory bulb
Open this publication in new window or tab >>Statistical analysis of coding for molecular properties in the olfactory bulb
2011 (English)In: Frontiers in Neuroscience, ISSN 1662-4548, E-ISSN 1662-453X, Vol. 5, no 62Article in journal (Refereed) Published
Abstract [en]

The relationship between molecular properties of odorants and neural activities is arguably one of the most important issues in olfaction and the rules governing this relationship are still not clear. In the olfactory bulb (OB), glomeruli relay olfactory information to second-order neurons which in turn project to cortical areas. We investigate relevance of odorant properties, spatial localization of glomerular coding sites, and size of coding zones in a dataset of [14C] 2-deoxyglucose images of glomeruli over the entire OB of the rat. We relate molecular properties to activation of glomeruli in the OB using a non-parametric statistical test and a support-vector machine classification study. Our method permits to systematically map the topographic representation of various classes of odorants in the OB. Our results suggest many localized coding sites for particular molecular properties and some molecular properties that could form the basis for a spatial map of olfactory information. We found that alkynes, alkanes, alkenes, and amines affect activation maps very strongly as compared to other properties and that amines, sulfur-containing compounds, and alkynes have small zones and high relevance to activation changes, while aromatics, alkanes, and carboxylics acid recruit very big zones in the dataset. Results suggest a local spatial encoding for molecular properties.

Place, publisher, year, edition, pages
Frontiers Research Foundation, 2011
Keyword
olfactory coding, olfactory bulb, odorants, glomeruli, property–activity relationship
National Category
Natural Sciences
Research subject
SRA - ICT
Identifiers
urn:nbn:se:kth:diva-48060 (URN)10.3389/fnsys.2011.00062 (DOI)2-s2.0-82455200567 (Scopus ID)
Note
This Document is Protected by copyright and was first published by Frontiers. All rights reserved. It is reproduced with permission. QC 20111116Available from: 2011-11-16 Created: 2011-11-15 Last updated: 2017-12-08Bibliographically approved
6. Map formation in the olfactory bulb by axon guidance of olfactory neurons
Open this publication in new window or tab >>Map formation in the olfactory bulb by axon guidance of olfactory neurons
2011 (English)In: Frontiers in Systems Neuroscience, ISSN 1662-5137, E-ISSN 1662-5137, Vol. 5, no 0Article in journal (Refereed) Published
Abstract [en]

The organization of representations in the brain has been observed to locally reflect subspaces of inputs that are relevant to behavioral or perceptual feature combinations, such as in areas receptive to lower and higher-order features in the visual system. The early olfactory system developed highly plastic mechanisms and convergent evidence indicates that projections from primary neurons converge onto the glomerular level of the olfactory bulb (OB) to form a code composed of continuous spatial zones that are differentially active for particular physico?-chemical feature combinations, some of which are known to trigger behavioral responses. In a model study of the early human olfactory system, we derive a glomerular organization based on a set of real-world,biologically-relevant stimuli, a distribution of receptors that respond each to a set of odorants of similar ranges of molecular properties, and a mechanism of axon guidance based on activity. Apart from demonstrating activity-dependent glomeruli formation and reproducing the relationship of glomerular recruitment with concentration, it is shown that glomerular responses reflect similarities of human odor category perceptions and that further, a spatial code provides a better correlation than a distributed population code. These results are consistent with evidence of functional compartmentalization in the OB and could suggest a function for the bulb in encoding of perceptual dimensions.

Place, publisher, year, edition, pages
Ch. de la Pécholettaz 6 CH – 1066 Epalinges Switzerland: Frontiers Media SA, 2011
Keyword
olfaction, plasticity, axonal guidance, olfactory coding, olfactory bulb, glomeruli, odor category, perception
National Category
Bioinformatics (Computational Biology) Neurosciences
Identifiers
urn:nbn:se:kth:diva-45308 (URN)10.3389/fnsys.2011.00084 (DOI)2-s2.0-84856170519 (Scopus ID)
Projects
NEUROCHEM
Note
This Document is Protected by copyright and was first published by Frontiers. All rights reserved. It is reproduced with permission.Available from: 2011-10-31 Created: 2011-10-28 Last updated: 2017-12-08Bibliographically approved

Open Access in DiVA

machine_learning_olfactory_system(4489 kB)1216 downloads
File information
File name FULLTEXT02.pdfFile size 4489 kBChecksum SHA-512
2d63f0419ebe9881b378d03809ad262604c34030d3cdc491e575de66decdb0556c052fa5e828012fa30d261bdf7a78f569d30fd3be5f7e6e99a96441ad46347e
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Auffarth, Benjamin
By organisation
Computational Biology, CB
Biological Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 1257 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 698 hits
CiteExportLink to record
Permanent link

Direct 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