Machine learning approaches to improve disease management of patients with rheumatoid arthritis: review and future directions

Abstract

Introduction: Although the management of rheumatoid arthritis (RA) has improved in major way over the last decades, this disease still leads to an important burden for patients and society, and there is a need to develop more personalized approaches. Machine learning (ML) methods are more and more used in health-related studies and can be applied to different sorts of data (clinical, radiological, or "omics" data). Such approaches may improve the management of patients with RA. Areas covered: In this paper, we propose a review regarding ML approaches applied to RA. A scoping literature search was performed in PubMed, in September 2021 using the following MeSH terms: "arthritis, rheumatoid" and "machine learning". Based on this search, the usefulness of ML methods for RA diagnosis, monitoring and prediction of response to treatment and RA outcomes, is discussed. Expert opinion: ML methods have the potential to revolutionize RA-related research and improve disease management and patient care. Nevertheless, these models are not yet ready to contribute fully to rheumatologists' daily practice. Indeed, these methods raise technical, methodological, and ethical issues, which should be addressed properly to allow their implementation. Collaboration between data scientists, clinical researchers and physicians is therefore required to move this field forward.

Understanding the role and adoption of artificial intelligence techniques in rheumatology research: an in-depth review of the literature

“…Different categories have been proposed to classify research articles. For instance, the scoping review presented in [34] suggested the following categories and subcategories to describe the ML studies applied in RA research: As shown, there is no standard form to classify the different clinical research articles in which data science is involved. The proposal followed in this review article tries to achieve a balance between resolution and significance.…”

Big data analyses and individual health profiling in the arena of rheumatic and musculoskeletal diseases (RMDs)

“…Growing number of studies recognize the great promise and rising potential of ML methods in revolutionizing research, disease management, and patient care for patients with RMDs. 50 , 77 , 78 However, ML methods are still rudimentary and not ready for prime time. Indeed, implementation of the novel ML prediction models in routine rheumatology practice meets technical, methodological and ethical limitations, 77 including small sample size (particularly in rare RMDs), lack of external validation, difficulty in operationalizing and implementing the models in independent clinical data set, and thus uncertain clinical utility.…”

“… 50 , 77 , 78 However, ML methods are still rudimentary and not ready for prime time. Indeed, implementation of the novel ML prediction models in routine rheumatology practice meets technical, methodological and ethical limitations, 77 including small sample size (particularly in rare RMDs), lack of external validation, difficulty in operationalizing and implementing the models in independent clinical data set, and thus uncertain clinical utility. 50 For example, ML model predicting the diagnosis of AS had accuracy of 0.81 in the training dataset, but had inferior performance in the data set not used in the original model development, yielding a PPV of only 6.24%, although this value was still higher than that of clinical model using Assessment of SpondyloArthritis international Society classification criteria (PPV = 1.29%) and higher than PPV of the logistic regression model (PPV = 2.55%).…”

A multicentre validation study of a smartphone application to screen hand arthritis

“…ML techniques have been applied to several aspects of rheumatology, including electronic health records, imaging, disease classification, disease outcome and treatment response prediction [ 15 , 16 ]. Previous ML applications to diagnose rheumatological conditions have used biological samples, such as serum biomarkers and genomic data from synovial tissue [ 17 , 18 ].…”