Federated learning for secure development of AI models for Parkinson's disease detection using speech from different languages

TL;DR

This paper demonstrates that federated learning can effectively develop AI models for Parkinson's disease detection from speech data across multiple languages, ensuring privacy and improving diagnostic accuracy without data sharing.

Contribution

It introduces a federated learning approach for multilingual PD detection, enabling collaborative model training while preserving patient privacy.

Findings

FL model outperforms local models in accuracy

FL model approaches centralized training performance

No data sharing required among institutions

Abstract

Parkinson's disease (PD) is a neurological disorder impacting a person's speech. Among automatic PD assessment methods, deep learning models have gained particular interest. Recently, the community has explored cross-pathology and cross-language models which can improve diagnostic accuracy even further. However, strict patient data privacy regulations largely prevent institutions from sharing patient speech data with each other. In this paper, we employ federated learning (FL) for PD detection using speech signals from 3 real-world language corpora of German, Spanish, and Czech, each from a separate institution. Our results indicate that the FL model outperforms all the local models in terms of diagnostic accuracy, while not performing very differently from the model based on centrally combined training sets, with the advantage of not requiring any data sharing among collaborators. This will simplify inter-institutional collaborations, resulting in enhancement of patient outcomes.