Interpretable Early Detection of Parkinson's Disease through Speech Analysis

2025-04-24

AI Summary

To address the clinical need for non-invasive, interpretable early screening of Parkinson’s disease (PD), this study proposes a speech-driven deep learning framework. First, multidimensional acoustic features—including MFCCs, jitter, and shimmer—are extracted from sustained phonations. Second, an end-to-end classification model is built, integrated with an attention-guided attribution mechanism to localize discriminative phonatory segments at the utterance level. Third, attribution outputs are mapped onto clinically meaningful dysarthria phenotypes—such as delayed voice onset and tremulous vocal fold oscillation—to infer underlying neuromuscular impairments. To our knowledge, this is the first approach that simultaneously achieves state-of-the-art classification accuracy (831 samples from the Italian PD Speech Database) and fine-grained, clinically interpretable, segment-level explainability. The framework enhances model transparency and diagnostic utility for clinical decision support.

Technology Category

• Natural Language Processing: Speech
• Machine Learning: Transparent, Interpretable, Explainable ML
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search

Abstract

Parkinson's disease is a progressive neurodegenerative disorder affecting motor and non-motor functions, with speech impairments among its earliest symptoms. Speech impairments offer a valuable diagnostic opportunity, with machine learning advances providing promising tools for timely detection. In this research, we propose a deep learning approach for early Parkinson's disease detection from speech recordings, which also highlights the vocal segments driving predictions to enhance interpretability. This approach seeks to associate predictive speech patterns with articulatory features, providing a basis for interpreting underlying neuromuscular impairments. We evaluated our approach using the Italian Parkinson's Voice and Speech Database, containing 831 audio recordings from 65 participants, including both healthy individuals and patients. Our approach showed competitive classification performance compared to state-of-the-art methods, while providing enhanced interpretability by identifying key speech features influencing predictions.

Problem

• Early detection of Parkinson's disease using speech analysis
• Interpretable deep learning for identifying predictive speech patterns
• Linking speech impairments to neuromuscular dysfunction in Parkinson's

Innovation

• Deep learning for early Parkinson's detection
• Interpretable vocal segment analysis
• Articulatory feature-based prediction patterns