Assessing clinical utility of machine learning and artificial intelligence approaches to analyze speech recordings in multiple sclerosis: A pilot study.

Autor: Svoboda E; Institute of Formal and Applied Linguistics, Faculty of Mathematics and Physics, Charles University, Prague, Czech Republic; Institute of Phonetics, Faculty of Arts, Charles University, Prague, Czech Republic., Bořil T; Institute of Phonetics, Faculty of Arts, Charles University, Prague, Czech Republic., Rusz J; Department of Circuit Theory, Faculty of Electrical Engineering, Czech Technical University in Prague, Prague, Czech Republic; Department of Neurology and Center of Clinical Neuroscience, First Faculty of Medicine, Charles University and General University Hospital, Prague, Czech Republic; Department of Neurology & ARTORG Center, Inselspital, Bern University Hospital, University of Bern, Switzerland., Tykalová T; Department of Circuit Theory, Faculty of Electrical Engineering, Czech Technical University in Prague, Prague, Czech Republic., Horáková D; Department of Neurology and Center of Clinical Neuroscience, First Faculty of Medicine, Charles University and General University Hospital, Prague, Czech Republic., Guttmann CRG; Center for Neurological Imaging, Brigham & Women's Hospital and Harvard Medical School, USA., Blagoev KB; Department of Biophysics, Johns Hopkins University, Baltimore, MD, 21218, USA., Hatabu H; Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA., Valtchinov VI; Center for Evidence-Based Imaging, USA; Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA; Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA. Electronic address: vvaltchinov@bwh.harvard.edu.
Jazyk: angličtina
Zdroj: Computers in biology and medicine [Comput Biol Med] 2022 Sep; Vol. 148, pp. 105853. Date of Electronic Publication: 2022 Jul 15.
DOI: 10.1016/j.compbiomed.2022.105853
Abstrakt: Background: An early diagnosis together with an accurate disease progression monitoring of multiple sclerosis is an important component of successful disease management. Prior studies have established that multiple sclerosis is correlated with speech discrepancies. Early research using objective acoustic measurements has discovered measurable dysarthria.
Method: The objective was to determine the potential clinical utility of machine learning and deep learning/AI approaches for the aiding of diagnosis, biomarker extraction and progression monitoring of multiple sclerosis using speech recordings. A corpus of 65 MS-positive and 66 healthy individuals reading the same text aloud was used for targeted acoustic feature extraction utilizing automatic phoneme segmentation. A series of binary classification models was trained, tuned, and evaluated regarding their Accuracy and area-under-the-curve.
Results: The Random Forest model performed best, achieving an Accuracy of 0.82 on the validation dataset and an area-under-the-curve of 0.76 across 5 k-fold cycles on the training dataset. 5 out of 7 acoustic features were statistically significant.
Conclusion: Machine learning and artificial intelligence in automatic analyses of voice recordings for aiding multiple sclerosis diagnosis and progression tracking seems promising. Further clinical validation of these methods and their mapping onto multiple sclerosis progression is needed, as well as a validating utility for English-speaking populations.
(Copyright © 2022 Elsevier Ltd. All rights reserved.)
Databáze: MEDLINE
Externí odkaz: