Innovative Speech-Based Deep Learning Approaches for Parkinson's Disease Classification: A Systematic Review

Lisanne van Gelderen,Cristian Tejedor-García

DOI: https://doi.org/10.3390/app14177873

2024-09-24

Abstract:Parkinson's disease (PD), the second most prevalent neurodegenerative disorder worldwide, frequently presents with early-stage speech impairments. Recent advancements in Artificial Intelligence (AI), particularly deep learning (DL), have significantly enhanced PD diagnosis through the analysis of speech data. Nevertheless, the progress of research is restricted by the limited availability of publicly accessible speech-based PD datasets, primarily due to privacy concerns. The goal of this systematic review is to explore the current landscape of speech-based DL approaches for PD classification, based on 33 scientific works published between January 2020 and March 2024. We discuss their available resources, capabilities, and potential limitations, and issues related to bias, explainability, and privacy. Furthermore, this review provides an overview of publicly accessible speech-based datasets and open-source material for PD. The DL approaches identified are categorized into end-to-end (E2E) learning, transfer learning (TL), and deep acoustic feature extraction (DAFE). Among E2E approaches, Convolutional Neural Networks (CNNs) are prevalent, though Transformers are increasingly popular. E2E approaches face challenges such as limited data and computational resources, especially with Transformers. TL addresses these issues by providing more robust PD diagnosis and better generalizability across languages. DAFE aims to improve the explainability and interpretability of results by examining the specific effects of deep features on both other DL approaches and more traditional machine learning (ML) methods. However, it often underperforms compared to E2E and TL approaches.

Sound,Artificial Intelligence,Computation and Language,Machine Learning,Audio and Speech Processing

What problem does this paper attempt to address?

### Problems Addressed by the Paper

This paper primarily explores the latest advancements in speech-based deep learning methods for Parkinson's Disease (PD) classification. Specifically, this systematic review covers 33 relevant scientific publications from 2020 to March 2024 and focuses on the following points:

1. **Current Research Status**: The paper provides a detailed overview of the application of speech-based deep learning methods in PD classification, including existing resources, capabilities, and potential limitations.

2. **Bias, Interpretability, and Privacy Issues**: It discusses the potential bias, interpretability, and privacy-related issues that these methods may encounter in the process of PD classification.

3. **Public Datasets and Open-Source Resources**: An overview of publicly available speech datasets and open-source code for PD research as of March 2024 is provided.

Through these aspects, the paper aims to answer the following research questions:
- **What are the latest speech-based deep learning methods used for PD classification?**
- **How do these speech-based deep learning methods perform in PD classification?**
- **What are the issues related to bias, interpretability, and privacy in these methods for PD classification?**

Overall, the goal of this review paper is to evaluate and summarize the latest advancements and challenges in speech-based deep learning methods for Parkinson's Disease diagnosis, ensuring privacy and fairness.