Personalizing Medicine and Technologies to Address the Experiences and Needs of People with Multiple Sclerosis

The role of machine learning in developing non-magnetic resonance imaging based biomarkers for multiple sclerosis: a systematic review

Background

Multiple sclerosis (MS) is a neurological condition whose symptoms, severity, and progression over time vary enormously among individuals. Ideally, each person living with MS should be provided with an accurate prognosis at the time of diagnosis, precision in initial and subsequent treatment decisions, and improved timeliness in detecting the need to reassess treatment regimens. To manage these three components, discovering an accurate, objective measure of overall disease severity is essential. Machine learning (ML) algorithms can contribute to finding such a clinically useful biomarker of MS through their ability to search and analyze datasets about potential biomarkers at scale. Our aim was to conduct a systematic review to determine how, and in what way, ML has been applied to the study of MS biomarkers on data from sources other than magnetic resonance imaging.

Methods

Systematic searches through eight databases were conducted for literature published in 2014–2020 on MS and specified ML algorithms.

Results

Of the 1, 052 returned papers, 66 met the inclusion criteria. All included papers addressed developing classifiers for MS identification or measuring its progression, typically, using hold-out evaluation on subsets of fewer than 200 participants with MS. These classifiers focused on biomarkers of MS, ranging from those derived from omics and phenotypical data (34.5% clinical, 33.3% biological, 23.0% physiological, and 9.2% drug response). Algorithmic choices were dependent on both the amount of data available for supervised ML (91.5%; 49.2% classification and 42.3% regression) and the requirement to be able to justify the resulting decision-making principles in healthcare settings. Therefore, algorithms based on decision trees and support vector machines were commonly used, and the maximum average performance of 89.9% AUC was found in random forests comparing with other ML algorithms.

Conclusions

ML is applicable to determining how candidate biomarkers perform in the assessment of disease severity. However, applying ML research to develop decision aids to help clinicians optimize treatment strategies and analyze treatment responses in individual patients calls for creating appropriate data resources and shared experimental protocols. They should target proceeding from segregated classification of signals or natural language to both holistic analyses across data modalities and clinically-meaningful differentiation of disease.

The current understanding of precision medicine and personalised medicine in selected research disciplines: study protocol of a systematic concept analysis

INTRODUCTION

The terms 'precision medicine' and 'personalised medicine' have become key terms in health-related research and in science-related public communication. However, the application of these two concepts and their interpretation in various disciplines are heterogeneous, which also affects research translation and public awareness. This leads to confusion regarding the use and distinction of the two concepts. Our aim is to provide a snapshot of the current understanding of these concepts.

METHODS AND ANALYSIS

Our study will use Rodgers' evolutionary concept analysis to systematically examine the current understanding of the concepts 'precision medicine' and 'personalised medicine' in clinical medicine, biomedicine (incorporating genomics and bioinformatics), health services research, physics, chemistry, engineering, machine learning and artificial intelligence, and to identify their respective attributes (clusters of characteristics) and surrogate and related terms. A systematic search of the literature will be conducted for 2016-2022 using databases relevant to each of these disciplines: ACM Digital Library, CINAHL, Cochrane Library, F1000Research, IEEE Xplore, PubMed/Medline, Science Direct, Scopus and Web of Science. These are among the most representative databases for the included disciplines. We will examine similarities and differences in definitions of 'precision medicine' and 'personalised medicine' in the respective disciplines and across (sub)disciplines, including attributes of each term. This will enable us to determine how these two concepts are distinguished.

ETHICS AND DISSEMINATION

Following ethical and research standards, we will comprehensively report the methodology for a systematic analysis following Rodgers' concept analysis method. Our systematic concept analysis will contribute to the clarification of the two concepts and distinction in their application in given settings and circumstances. Such a broad concept analysis will contribute to non-systematic syntheses of the concepts, or occasional systematic reviews on one of the concepts that have been published in specific disciplines, in order to facilitate interdisciplinary communication, translational medical research and implementation science.