Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Investigating the Effectiveness of Explainability Methods in Parkinson's Detection from Speech

Eleonora Mancini, Francesco Paissan, Paolo Torroni, Mirco Ravanelli, Cem Subakan

TL;DR

This study systematically evaluates several explainability methods to identify PD-specific speech features, aiming to support the development of accurate, interpretable models for clinical decision-making in PD diagnosis and monitoring.

Abstract

Speech impairments in Parkinson's disease (PD) provide significant early indicators for diagnosis. While models for speech-based PD detection have shown strong performance, their interpretability remains underexplored. This study systematically evaluates several explainability methods to identify PD-specific speech features, aiming to support the development of accurate, interpretable models for clinical decision-making in PD diagnosis and monitoring. Our methodology involves (i) obtaining attributions and saliency maps using mainstream interpretability techniques, (ii) quantitatively evaluating the faithfulness of these maps and their combinations obtained via union and intersection through a range of established metrics, and (iii) assessing the information conveyed by the saliency maps for PD detection from an auxiliary classifier. Our results reveal that, while explanations are aligned with the classifier, they often fail to provide valuable information for domain experts.

Investigating the Effectiveness of Explainability Methods in Parkinson's Detection from Speech

TL;DR

This study systematically evaluates several explainability methods to identify PD-specific speech features, aiming to support the development of accurate, interpretable models for clinical decision-making in PD diagnosis and monitoring.

Abstract

Speech impairments in Parkinson's disease (PD) provide significant early indicators for diagnosis. While models for speech-based PD detection have shown strong performance, their interpretability remains underexplored. This study systematically evaluates several explainability methods to identify PD-specific speech features, aiming to support the development of accurate, interpretable models for clinical decision-making in PD diagnosis and monitoring. Our methodology involves (i) obtaining attributions and saliency maps using mainstream interpretability techniques, (ii) quantitatively evaluating the faithfulness of these maps and their combinations obtained via union and intersection through a range of established metrics, and (iii) assessing the information conveyed by the saliency maps for PD detection from an auxiliary classifier. Our results reveal that, while explanations are aligned with the classifier, they often fail to provide valuable information for domain experts.

Paper Structure

This paper contains 17 sections, 1 equation, 2 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Methodology
  4. Transforming Waveforms for Gradient-Based Attributions
  5. Quantitative Analysis through Explainability Metrics
  6. Exploring Overlap in Explainability Methods
  7. Classification Using Saliency Maps and Selective Metrics
  8. Experiments
  9. Dataset
  10. PD Detector
  11. Saliency Maps Classifier
  12. Results
  13. Faithfulness Metrics
  14. Overlap between Explainability Methods
  15. Saliency Maps Classifier
  16. ...and 2 more sections

Figures (2)

  • Figure 1: Explanations generated for a PD sample correctly classified by HuBERT. The explanations highlight different portions of the spectrogram, suggesting that understandability is difficult to achieve in this setting. From left to right: original sample, Integrated Gradients, Guided Backprop, and Guided GradCAM.
  • Figure 2: Analysis of the correlation between explanations overlap among interpretability techniques and faithfulness metrics. Combining interpretability techniques is most effective when there is already significant overlap between attribution masks.