TY - JOUR

T1 - Discovering predictive features of multiple sclerosis from clinically isolated syndrome with machine learning

AU - Luu, Minh Sao Khue

AU - Tuchinov, Bair N.

AU - Prokaeva, Anna I.

AU - Коробко, Денис Сергеевич

AU - Малкова, Надежда Алексеевна

AU - Tulupov, Andrey A.

N1 - Discovering predictive features of multiple sclerosis from clinically isolated syndrome with machine learning / M. S. Kh. Luu, B. N. Tuchinov, A. I. Prokaeva [et al.] // Artificial Intelligence in Health. – 2024. – Vol. 1. - No. 4. – Pp. 107-122. – DOI 10.36922/aih.4255. – EDN FZCWCK. This work was supported by a grant from the Russian Science Foundation (RSF 23-15-00377).

PY - 2024/9/24

Y1 - 2024/9/24

N2 - Accurately predicting the progression of clinically isolated syndrome (CIS) to multiple sclerosis (MS) is crucial for early intervention and management. This study employs a range of machine learning models, including categorical boosting, extreme gradient boosting, light gradient boosting machine, random forest, support vector machine, and logistic regression, to classify CIS patients based on their likelihood of developing MS. Our best model achieves and demonstrates superior predictive accuracy of 0.9312, measured using the area under the curve metric. In addition, we apply explainability techniques to determine the most influential features driving the predictions, identifying which CISs are most indicative of MS progression. Furthermore, we explore feature interactions to detect relationships between features, providing a deeper understanding of the underlying mechanisms. The study utilizes public data from 273 CISs patients, offering significant contributions to the clinical management and early diagnosis of MS.

AB - Accurately predicting the progression of clinically isolated syndrome (CIS) to multiple sclerosis (MS) is crucial for early intervention and management. This study employs a range of machine learning models, including categorical boosting, extreme gradient boosting, light gradient boosting machine, random forest, support vector machine, and logistic regression, to classify CIS patients based on their likelihood of developing MS. Our best model achieves and demonstrates superior predictive accuracy of 0.9312, measured using the area under the curve metric. In addition, we apply explainability techniques to determine the most influential features driving the predictions, identifying which CISs are most indicative of MS progression. Furthermore, we explore feature interactions to detect relationships between features, providing a deeper understanding of the underlying mechanisms. The study utilizes public data from 273 CISs patients, offering significant contributions to the clinical management and early diagnosis of MS.

KW - Binary classification

KW - Clinically isolated syndromes

KW - Machine learning

KW - Model explainability

KW - Multiple sclerosis

KW - Predictive features

KW - Clinically isolated syndromes

KW - Multiple sclerosis

KW - Machine learning

KW - Binary classification

KW - Predictive features

KW - Model explainability

UR - https://www.scopus.com/pages/publications/105021437632

UR - https://www.elibrary.ru/item.asp?id=74143876

UR - https://www.mendeley.com/catalogue/80a0db0f-4f3b-3532-bd2a-0a568ef13bef/

U2 - 10.36922/aih.4255

DO - 10.36922/aih.4255

M3 - Article

VL - 1

SP - 107

EP - 122

JO - Artificial Intelligence in Health

JF - Artificial Intelligence in Health

SN - 3041-0894

IS - 4

ER -