Detecting Parkinson’s Disease Using Voice Recordings From Mobile Devices
(2024) 32nd European Signal Processing Conference, EUSIPCO 2024 p.1516-1520- Abstract
This study examines how one may detect Parkinson’s disease (PD) by analyzing voice recordings made with a mobile phone. The key objectives include creating a model that ensures fair and unbiased predictions while maintaining interpretability that is consistent with the existing literature on PD. We introduce a novel group-wise scaling method to address typical age and biological sex biases in the data sets, demonstrating an average 14% improvement as compared to conventional scaling. We also show the importance of evaluating the developed model on unseen individuals to achieve reliable classification results. The developed model offers an accuracy of 80.3% with a recall of 89.4% for unseen individuals, surpassing current... (More)
This study examines how one may detect Parkinson’s disease (PD) by analyzing voice recordings made with a mobile phone. The key objectives include creating a model that ensures fair and unbiased predictions while maintaining interpretability that is consistent with the existing literature on PD. We introduce a novel group-wise scaling method to address typical age and biological sex biases in the data sets, demonstrating an average 14% improvement as compared to conventional scaling. We also show the importance of evaluating the developed model on unseen individuals to achieve reliable classification results. The developed model offers an accuracy of 80.3% with a recall of 89.4% for unseen individuals, surpassing current state-of-the-art approaches. Furthermore, we offer insights into the decision-making of the model using SHAP values. Our analysis reveals that a low standard deviation in the slope of the fundamental frequency (indicative of monotone voice quality) increases the likelihood of classifying a PD voice. Additionally, longer unvoiced segments, increased loudness standard deviation, and jitter correlate with the presence of PD. The significance of our proposed model lies in its generalizability and reliability for early PD detection, potentially decelerating disease progression, reducing healthcare costs, and improving the quality of life for patients.
(Less)
- author
- Momeni, Niloofar
LU
; Whitling, Susanna LU and Jakobsson, Andreas LU
- organization
-
- Mathematical Statistics
- LTH Profile Area: AI and Digitalization
- Logopedics, Phoniatrics and Audiology
- The voice group (research group)
- Communication and Cognition (research group)
- Statistical Signal Processing Group (research group)
- Biomedical Modelling and Computation (research group)
- eSSENCE: The e-Science Collaboration
- LTH Profile Area: Engineering Health
- LU Profile Area: Natural and Artificial Cognition
- ELLIIT: the Linköping-Lund initiative on IT and mobile communication
- publishing date
- 2024
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- bias, Parkinson’s disease detection, unfairness, voice features interpretability
- host publication
- 32nd European Signal Processing Conference, EUSIPCO 2024 - Proceedings
- pages
- 5 pages
- publisher
- European Signal Processing Conference, EUSIPCO
- conference name
- 32nd European Signal Processing Conference, EUSIPCO 2024
- conference location
- Lyon, France
- conference dates
- 2024-08-26 - 2024-08-30
- external identifiers
-
- scopus:85208436867
- ISBN
- 9789464593617
- DOI
- 10.23919/eusipco63174.2024.10715471
- language
- English
- LU publication?
- yes
- id
- 6802d1ce-3ebc-447b-ab19-b79bcb828c61
- date added to LUP
- 2025-02-18 10:15:22
- date last changed
- 2025-04-04 14:24:49
@inproceedings{6802d1ce-3ebc-447b-ab19-b79bcb828c61, abstract = {{<p>This study examines how one may detect Parkinson’s disease (PD) by analyzing voice recordings made with a mobile phone. The key objectives include creating a model that ensures fair and unbiased predictions while maintaining interpretability that is consistent with the existing literature on PD. We introduce a novel group-wise scaling method to address typical age and biological sex biases in the data sets, demonstrating an average 14% improvement as compared to conventional scaling. We also show the importance of evaluating the developed model on unseen individuals to achieve reliable classification results. The developed model offers an accuracy of 80.3% with a recall of 89.4% for unseen individuals, surpassing current state-of-the-art approaches. Furthermore, we offer insights into the decision-making of the model using SHAP values. Our analysis reveals that a low standard deviation in the slope of the fundamental frequency (indicative of monotone voice quality) increases the likelihood of classifying a PD voice. Additionally, longer unvoiced segments, increased loudness standard deviation, and jitter correlate with the presence of PD. The significance of our proposed model lies in its generalizability and reliability for early PD detection, potentially decelerating disease progression, reducing healthcare costs, and improving the quality of life for patients.</p>}}, author = {{Momeni, Niloofar and Whitling, Susanna and Jakobsson, Andreas}}, booktitle = {{32nd European Signal Processing Conference, EUSIPCO 2024 - Proceedings}}, isbn = {{9789464593617}}, keywords = {{bias; Parkinson’s disease detection; unfairness; voice features interpretability}}, language = {{eng}}, pages = {{1516--1520}}, publisher = {{European Signal Processing Conference, EUSIPCO}}, title = {{Detecting Parkinson’s Disease Using Voice Recordings From Mobile Devices}}, url = {{http://dx.doi.org/10.23919/eusipco63174.2024.10715471}}, doi = {{10.23919/eusipco63174.2024.10715471}}, year = {{2024}}, }