Individualized lesion-symptom mapping using explainable artificial intelligence for the cognitive impact of white matter hyperintensities
Lesion-symptom mapping methods assess the relationship between lesions caused by cerebral small vessel disease and cognition, but current technology like support vector regression (SVR)) primarily provide group-level results. We propose a novel lesion-symptom mapping approach that can indicate how l...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | NeuroImage: Clinical |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2213158225000609 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849694868801060864 |
|---|---|
| author | Ryanne Offenberg Alberto De Luca Geert Jan Biessels Frederik Barkhof Wiesje M. van der Flier Argonde C. van Harten Ewoud van der Lelij Josien Pluim Hugo Kuijf |
| author_facet | Ryanne Offenberg Alberto De Luca Geert Jan Biessels Frederik Barkhof Wiesje M. van der Flier Argonde C. van Harten Ewoud van der Lelij Josien Pluim Hugo Kuijf |
| author_sort | Ryanne Offenberg |
| collection | DOAJ |
| description | Lesion-symptom mapping methods assess the relationship between lesions caused by cerebral small vessel disease and cognition, but current technology like support vector regression (SVR)) primarily provide group-level results. We propose a novel lesion-symptom mapping approach that can indicate how lesion patterns contribute to cognitive impairment on an individual level. A convolutional neural network (CNN) predicts cognitive scores and is combined with explainable artificial intelligence (XAI) to map the relation between cognition and vascular lesions.This method was evaluated primarily using real white matter hyperintensity maps of 821 memory clinic patients and simulated cognitive data, with weighted lesions and noise levels. Simulated data provided ground truth locations to assess predictive performance of the CNN and accuracy of strategic lesion identification by XAI, using an established lesion-symptom mapping method, SVR, and a simple fully connected neural network (FNN) as benchmarks. Real cognitive scores were used in a final proof-of-principle analysis.Predictive performance in simulation experiments was high for the CNN (R2 = 0.964), SVR (R2 = 0.875), and FNN (R2 = 0.863). CNN with XAI provided patient-specific attribution maps that highlighted the ground truth locations. All methods showed similar sensitivity to noise. Using real cognitive scores, SVR (R2 = 0.291) obtained a somewhat higher predictive performance than the CNN (R2 = 0.216), although both methods substantially exceeded the predictive performance of total WMH volume alone (R2 = 0.013). The FNN performed worse on real data (R2 = 0.020).To conclude, results show that CNNs combined with XAI can perform lesion-symptom mapping and generate individual attribution maps, which could be a valuable feature with further method development. |
| format | Article |
| id | doaj-art-d3c66d2f12ed4c77985da1c06b544643 |
| institution | DOAJ |
| issn | 2213-1582 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | NeuroImage: Clinical |
| spelling | doaj-art-d3c66d2f12ed4c77985da1c06b5446432025-08-20T03:19:56ZengElsevierNeuroImage: Clinical2213-15822025-01-014610379010.1016/j.nicl.2025.103790Individualized lesion-symptom mapping using explainable artificial intelligence for the cognitive impact of white matter hyperintensitiesRyanne Offenberg0Alberto De Luca1Geert Jan Biessels2Frederik Barkhof3Wiesje M. van der Flier4Argonde C. van Harten5Ewoud van der Lelij6Josien Pluim7Hugo Kuijf8Image Sciences Institute, UMC Utrecht, Utrecht, the Netherlands; Corresponding author at: Image Sciences Institute, UMC Utrecht, Heidelberglaan 100, 3584 CX Utrecht, the Netherlands.Image Sciences Institute, UMC Utrecht, Utrecht, the NetherlandsDepartment of Neurology, UMC Utrecht, Utrecht, the NetherlandsDepartment of Radiology & Nuclear Medicine, Amsterdam UMC, Vrije Universiteit, the Netherlands; Queen Square Institute of Neurology and Centre for Medical Image Computing, University College London, UKAlzheimer Center Amsterdam, Neurology, Vrije Universiteit Amsterdam, Amsterdam UMC Location VUmc, Amsterdam, the Netherlands; Amsterdam Neuroscience, Neurodegeneration, Amsterdam, the Netherlands; Epidemiology & Data Science, Vrije Universiteit Amsterdam, Amsterdam UMC Location VUmc, Amsterdam, the NetherlandsAlzheimer Center Amsterdam, Neurology, Vrije Universiteit Amsterdam, Amsterdam UMC Location VUmc, Amsterdam, the Netherlands; Amsterdam Neuroscience, Neurodegeneration, Amsterdam, the NetherlandsDepartment of Neurology, UMC Utrecht, Utrecht, the NetherlandsImage Sciences Institute, UMC Utrecht, Utrecht, the Netherlands; Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven, the NetherlandsImage Sciences Institute, UMC Utrecht, Utrecht, the NetherlandsLesion-symptom mapping methods assess the relationship between lesions caused by cerebral small vessel disease and cognition, but current technology like support vector regression (SVR)) primarily provide group-level results. We propose a novel lesion-symptom mapping approach that can indicate how lesion patterns contribute to cognitive impairment on an individual level. A convolutional neural network (CNN) predicts cognitive scores and is combined with explainable artificial intelligence (XAI) to map the relation between cognition and vascular lesions.This method was evaluated primarily using real white matter hyperintensity maps of 821 memory clinic patients and simulated cognitive data, with weighted lesions and noise levels. Simulated data provided ground truth locations to assess predictive performance of the CNN and accuracy of strategic lesion identification by XAI, using an established lesion-symptom mapping method, SVR, and a simple fully connected neural network (FNN) as benchmarks. Real cognitive scores were used in a final proof-of-principle analysis.Predictive performance in simulation experiments was high for the CNN (R2 = 0.964), SVR (R2 = 0.875), and FNN (R2 = 0.863). CNN with XAI provided patient-specific attribution maps that highlighted the ground truth locations. All methods showed similar sensitivity to noise. Using real cognitive scores, SVR (R2 = 0.291) obtained a somewhat higher predictive performance than the CNN (R2 = 0.216), although both methods substantially exceeded the predictive performance of total WMH volume alone (R2 = 0.013). The FNN performed worse on real data (R2 = 0.020).To conclude, results show that CNNs combined with XAI can perform lesion-symptom mapping and generate individual attribution maps, which could be a valuable feature with further method development.http://www.sciencedirect.com/science/article/pii/S2213158225000609Lesion-symptom mappingNeural networkExplainable artificial intelligenceVascular cognitive impairmentMachine learning |
| spellingShingle | Ryanne Offenberg Alberto De Luca Geert Jan Biessels Frederik Barkhof Wiesje M. van der Flier Argonde C. van Harten Ewoud van der Lelij Josien Pluim Hugo Kuijf Individualized lesion-symptom mapping using explainable artificial intelligence for the cognitive impact of white matter hyperintensities NeuroImage: Clinical Lesion-symptom mapping Neural network Explainable artificial intelligence Vascular cognitive impairment Machine learning |
| title | Individualized lesion-symptom mapping using explainable artificial intelligence for the cognitive impact of white matter hyperintensities |
| title_full | Individualized lesion-symptom mapping using explainable artificial intelligence for the cognitive impact of white matter hyperintensities |
| title_fullStr | Individualized lesion-symptom mapping using explainable artificial intelligence for the cognitive impact of white matter hyperintensities |
| title_full_unstemmed | Individualized lesion-symptom mapping using explainable artificial intelligence for the cognitive impact of white matter hyperintensities |
| title_short | Individualized lesion-symptom mapping using explainable artificial intelligence for the cognitive impact of white matter hyperintensities |
| title_sort | individualized lesion symptom mapping using explainable artificial intelligence for the cognitive impact of white matter hyperintensities |
| topic | Lesion-symptom mapping Neural network Explainable artificial intelligence Vascular cognitive impairment Machine learning |
| url | http://www.sciencedirect.com/science/article/pii/S2213158225000609 |
| work_keys_str_mv | AT ryanneoffenberg individualizedlesionsymptommappingusingexplainableartificialintelligenceforthecognitiveimpactofwhitematterhyperintensities AT albertodeluca individualizedlesionsymptommappingusingexplainableartificialintelligenceforthecognitiveimpactofwhitematterhyperintensities AT geertjanbiessels individualizedlesionsymptommappingusingexplainableartificialintelligenceforthecognitiveimpactofwhitematterhyperintensities AT frederikbarkhof individualizedlesionsymptommappingusingexplainableartificialintelligenceforthecognitiveimpactofwhitematterhyperintensities AT wiesjemvanderflier individualizedlesionsymptommappingusingexplainableartificialintelligenceforthecognitiveimpactofwhitematterhyperintensities AT argondecvanharten individualizedlesionsymptommappingusingexplainableartificialintelligenceforthecognitiveimpactofwhitematterhyperintensities AT ewoudvanderlelij individualizedlesionsymptommappingusingexplainableartificialintelligenceforthecognitiveimpactofwhitematterhyperintensities AT josienpluim individualizedlesionsymptommappingusingexplainableartificialintelligenceforthecognitiveimpactofwhitematterhyperintensities AT hugokuijf individualizedlesionsymptommappingusingexplainableartificialintelligenceforthecognitiveimpactofwhitematterhyperintensities |