Multiple Sclerosis Diagnosis Using Machine Learning and Deep Learning: Challenges and Opportunities

Abstract

Multiple Sclerosis (MS) is a disease that impacts the central nervous system (CNS), which can lead to brain, spinal cord, and optic nerve problems. A total of 2.8 million are estimated to suffer from MS. Globally, a new case of MS is reported every five minutes. In this review, we discuss the proposed approaches to diagnosing MS using machine learning (ML) published between 2011 and 2022. Numerous models have been developed using different types of data, including magnetic resonance imaging (MRI) and clinical data. We identified the methods that achieved the best results in diagnosing MS. The most implemented approaches are SVM, RF, and CNN. Moreover, we discussed the challenges and opportunities in MS diagnosis to improve AI systems to enable researchers and practitioners to enhance their approaches and improve the automated diagnosis of MS. The challenges faced by automated MS diagnosis include difficulty distinguishing the disease from other diseases showing similar symptoms, protecting the confidentiality of the patients’ data, achieving reliable ML models that are also easily understood by non-experts, and the difficulty of collecting a large reliable dataset. Moreover, we discussed several opportunities in the field such as the implementation of secure platforms, employing better AI solutions, developing better disease prognosis systems, combining more than one data type for better MS prediction and using OCT data for diagnosis, utilizing larger, multi-center datasets to improve the reliability of the developed models, and commercialization.

Explainable artificial intelligence toward usable and trustworthy computer-aided early diagnosis of multiple sclerosis from Optical Coherence Tomography

“…The different ways of quantifying the layer thicknesses have been used as feature sets for building machine learning systems toward the computer-aided diagnosis of MS. The comprehensive review recently provided by Aslam et al in [2] describes the advances in the performance of these methods and discusses some limitations.…”

“…The major criticism from Aslam et al in [2] is that these studies have been conducted with a very small number of samples. Consequently, the models may not be robust presenting a high variance.…”

Künstliche Intelligenz in der Neurologie

“…In der Neuroimmunologie wird die intelligente Analyse der (MRT-)Bildgebung, analog zur Neuroonkologie zur Diagnosefindung sowie Therapieansprechen vor allem bei Multipler Sklerose (MS) genutzt 54 . Seltenere Erkrankungen aus dem Bereich der Neuroimmunologie spielen in der KI-Forschung kaum eine Rolle.…”

“…Zum möglichst großen Informationsgewinn und zur Optimierung der Therapie werden nicht nur MRT-Daten analysiert, sondern auch nuklearmedizinische bildgebende Verfahren, molekulare Maker oder DNA-Methylisierungsmethoden[51,52]. All diese Ansätze dienen der Verbesserung der Standarisierung und Klassifizierung von ZNS-Tumoren sowie der Personalisierung der Therapie[53].NeuroimmunologieIn der Neuroimmunologie wird die intelligente Analyse der (MRT-) Bildgebung, analog zur Neuroonkologie zur Diagnosefindung sowie Therapieansprechen vor allem bei Multipler Sklerose (MS) genutzt[54]. Seltenere Erkrankungen aus dem Bereich der Neuroimmunologie spielen in der KI-Forschung kaum eine Rolle.…”

Peculiarities of the Course, Diagnosis, and Treatment of Multiple Sclerosis

“…The first signs in patients with late-onset MS are usually movement disorders with predominating lower-limb paraparesis (in more than 50% of patients), often with a significant increase in muscle tone. Some publications indicate a higher frequency of the primary-progressive course in the case of a late debut [8,9].…”