Machine Learning Smart System for Parkinson Disease Classification Using the Voice as a Biomarker
Objectives This study presents PD Predict, a machine learning system for Parkinson disease classification using voice as a biomarker. Methods We first created an original set of recordings from the mPower study, and then extracted several audio features, such as mel-frequency cepstral coefficient (M...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
The Korean Society of Medical Informatics
2022-07-01
|
| Series: | Healthcare Informatics Research |
| Subjects: | |
| Online Access: | http://www.e-hir.org/upload/pdf/hir-2022-28-3-210.pdf |
| _version_ | 1811186825638707200 |
|---|---|
| author | Ilias Tougui Abdelilah Jilbab Jamal El Mhamdi |
| author_facet | Ilias Tougui Abdelilah Jilbab Jamal El Mhamdi |
| author_sort | Ilias Tougui |
| collection | DOAJ |
| description | Objectives This study presents PD Predict, a machine learning system for Parkinson disease classification using voice as a biomarker. Methods We first created an original set of recordings from the mPower study, and then extracted several audio features, such as mel-frequency cepstral coefficient (MFCC) components and other classical speech features, using a windowing procedure. The generated dataset was then divided into training and holdout sets. The training set was used to train two machine learning pipelines, and their performance was estimated using a nested subject-wise cross-validation approach. The holdout set was used to assess the generalizability of the pipelines for unseen data. The final pipelines were implemented in PD Predict and accessed through a prediction endpoint developed using the Django REST Framework. PD Predict is a two-component system: a desktop application that records audio recordings, extracts audio features, and makes predictions; and a server-side web application that implements the machine learning pipelines and processes incoming requests with the extracted audio features to make predictions. Our system is deployed and accessible via the following link: https://pdpredict.herokuapp.com/. Results Both machine learning pipelines showed moderate performance, between 65% and 75% using the nested subject-wise cross-validation approach. Furthermore, they generalized well to unseen data and they did not overfit the training set. Conclusions The architecture of PD Predict is clear, and the performance of the implemented machine learning pipelines is promising and confirms the usability of smartphone microphones for capturing digital biomarkers of disease. |
| first_indexed | 2024-04-11T13:51:40Z |
| format | Article |
| id | doaj.art-906f684ba15549d7a99a9b50e4ef3061 |
| institution | Directory Open Access Journal |
| issn | 2093-3681 2093-369X |
| language | English |
| last_indexed | 2024-04-11T13:51:40Z |
| publishDate | 2022-07-01 |
| publisher | The Korean Society of Medical Informatics |
| record_format | Article |
| series | Healthcare Informatics Research |
| spelling | doaj.art-906f684ba15549d7a99a9b50e4ef30612022-12-22T04:20:33ZengThe Korean Society of Medical InformaticsHealthcare Informatics Research2093-36812093-369X2022-07-0128321022110.4258/hir.2022.28.3.2101122Machine Learning Smart System for Parkinson Disease Classification Using the Voice as a BiomarkerIlias Tougui0Abdelilah Jilbab1Jamal El Mhamdi2E2SN, ENSIAS, Mohammed V University in Rabat, Rabat, MoroccoE2SN, ENSIAS, Mohammed V University in Rabat, Rabat, MoroccoE2SN, ENSIAS, Mohammed V University in Rabat, Rabat, MoroccoObjectives This study presents PD Predict, a machine learning system for Parkinson disease classification using voice as a biomarker. Methods We first created an original set of recordings from the mPower study, and then extracted several audio features, such as mel-frequency cepstral coefficient (MFCC) components and other classical speech features, using a windowing procedure. The generated dataset was then divided into training and holdout sets. The training set was used to train two machine learning pipelines, and their performance was estimated using a nested subject-wise cross-validation approach. The holdout set was used to assess the generalizability of the pipelines for unseen data. The final pipelines were implemented in PD Predict and accessed through a prediction endpoint developed using the Django REST Framework. PD Predict is a two-component system: a desktop application that records audio recordings, extracts audio features, and makes predictions; and a server-side web application that implements the machine learning pipelines and processes incoming requests with the extracted audio features to make predictions. Our system is deployed and accessible via the following link: https://pdpredict.herokuapp.com/. Results Both machine learning pipelines showed moderate performance, between 65% and 75% using the nested subject-wise cross-validation approach. Furthermore, they generalized well to unseen data and they did not overfit the training set. Conclusions The architecture of PD Predict is clear, and the performance of the implemented machine learning pipelines is promising and confirms the usability of smartphone microphones for capturing digital biomarkers of disease.http://www.e-hir.org/upload/pdf/hir-2022-28-3-210.pdfparkinson diseasevoice disordersmachine learningdiagnosiscomputer-assistedmedical informatics applications |
| spellingShingle | Ilias Tougui Abdelilah Jilbab Jamal El Mhamdi Machine Learning Smart System for Parkinson Disease Classification Using the Voice as a Biomarker Healthcare Informatics Research parkinson disease voice disorders machine learning diagnosis computer-assisted medical informatics applications |
| title | Machine Learning Smart System for Parkinson Disease Classification Using the Voice as a Biomarker |
| title_full | Machine Learning Smart System for Parkinson Disease Classification Using the Voice as a Biomarker |
| title_fullStr | Machine Learning Smart System for Parkinson Disease Classification Using the Voice as a Biomarker |
| title_full_unstemmed | Machine Learning Smart System for Parkinson Disease Classification Using the Voice as a Biomarker |
| title_short | Machine Learning Smart System for Parkinson Disease Classification Using the Voice as a Biomarker |
| title_sort | machine learning smart system for parkinson disease classification using the voice as a biomarker |
| topic | parkinson disease voice disorders machine learning diagnosis computer-assisted medical informatics applications |
| url | http://www.e-hir.org/upload/pdf/hir-2022-28-3-210.pdf |
| work_keys_str_mv | AT iliastougui machinelearningsmartsystemforparkinsondiseaseclassificationusingthevoiceasabiomarker AT abdelilahjilbab machinelearningsmartsystemforparkinsondiseaseclassificationusingthevoiceasabiomarker AT jamalelmhamdi machinelearningsmartsystemforparkinsondiseaseclassificationusingthevoiceasabiomarker |