Strategies for feature extraction from structural brain imaging in lesion‐deficit modelling
| Autor: | Hans-Otto Karnath, Vanessa Kasties, Christoph Sperber |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
Computer science
Feature extraction Models Neurological Feature selection Neuroimaging Cognitive neuroscience Machine Learning Atlases as Topic features Humans Radiology Nuclear Medicine and imaging Research Articles imaging biomarkers Radiological and Ultrasound Technology VLSM business.industry Representation (systemics) Pattern recognition prediction stroke Support vector machine Neurology Feature (computer vision) parcellation Principal component analysis Neurology (clinical) Artificial intelligence Anatomy business Biomarkers Research Article |
| Zdroj: | Human Brain Mapping |
| ISSN: | 1097-0193 1065-9471 |
| Popis: | High‐dimensional modelling of post‐stroke deficits from structural brain imaging is highly relevant to basic cognitive neuroscience and bears the potential to be translationally used to guide individual rehabilitation measures. One strategy to optimise model performance is well‐informed feature selection and representation. However, different feature representation strategies were so far used, and it is not known what strategy is best for modelling purposes. The present study compared the three common main strategies: voxel‐wise representation, lesion‐anatomical componential feature reduction and region‐wise atlas‐based feature representation. We used multivariate, machine‐learning‐based lesion‐deficit models to predict post‐stroke deficits based on structural lesion data. Support vector regression was tuned by nested cross‐validation techniques and tested on held‐out validation data to estimate model performance. While we consistently found the numerically best models for lower‐dimensional, featurised data and almost always for principal components extracted from lesion maps, our results indicate only minor, non‐significant differences between different feature representation styles. Hence, our findings demonstrate the general suitability of all three commonly applied feature representations in lesion‐deficit modelling. Likewise, model performance between qualitatively different popular brain atlases was not significantly different. Our findings also highlight potential minor benefits in individual fine‐tuning of feature representations and the challenge posed by the high, multifaceted complexity of lesion data, where lesion‐anatomical and functional criteria might suggest opposing solutions to feature reduction. Kasties et al. compared the suitability of different featurisation strategies in lesion‐behaviour modelling. They found no significant differences in model performance between voxel‐wise, atlas‐based region‐wise and componential featurisation. This demonstrates the general suitability of all three commonly applied feature representations in lesion‐behaviour modelling. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Plný text