# Individualized lesion-symptom mapping using explainable artificial intelligence for the cognitive impact of white matter hyperintensities

## TL;DR

This paper introduces a new method using AI to map how brain lesions affect cognition on an individual level, showing promising results.

## Contribution

A novel combination of convolutional neural networks and explainable AI for individualized lesion-symptom mapping.

## Key findings

- The CNN with XAI produced accurate patient-specific maps matching simulated ground truth.
- The CNN performed well in simulations (R2 = 0.964) but slightly underperformed SVR on real data.
- All methods outperformed using total WMH volume alone for predicting cognitive scores.

## Abstract

•First step towards novel lesion-symptom mapping method.•Combination of a neural network and eXplainable AI.•Patient-specific maps that indicate lesion location importance.•Novel method achieves comparable performance to benchmarks.•Application on simulated and real cognitive data.

First step towards novel lesion-symptom mapping method.

Combination of a neural network and eXplainable AI.

Patient-specific maps that indicate lesion location importance.

Novel method achieves comparable performance to benchmarks.

Application on simulated and real cognitive data.

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.