Machine learning identifies an immunological pattern associated with multiple juvenile idiopathic arthritis subtypes

Abstract

ObjectivesJuvenile idiopathic arthritis (JIA) is the most common class of childhood rheumatic diseases, with distinct disease subsets that may have diverging pathophysiological origins. Both adaptive and innate immune processes have been proposed as primary drivers, which may account for the observed clinical heterogeneity, but few high-depth studies have been performed.MethodsHere we profiled the adaptive immune system of 85 patients with JIA and 43 age-matched controls with indepth flow cytometry and machine learning approaches.ResultsImmune profiling identified immunological changes in patients with JIA. This immune signature was shared across a broad spectrum of childhood inflammatory diseases. The immune signature was identified in clinically distinct subsets of JIA, but was accentuated in patients with systemic JIA and those patients with active disease. Despite the extensive overlap in the immunological spectrum exhibited by healthy children and patients with JIA, machine learning analysis of the data set proved capable of discriminating patients with JIA from healthy controls with ~90% accuracy.ConclusionsThese results pave the way for large-scale immune phenotyping longitudinal studies of JIA. The ability to discriminate between patients with JIA and healthy individuals provides proof of principle for the use of machine learning to identify immune signatures that are predictive to treatment response group.

Disease-associated and patient-specific immune cell signatures in juvenile-onset systemic lupus erythematosus: patient stratification using a machine-learning approach

“…In the past 4 years, indepth computational analysis of multiomic datasets has accelerated the understanding of complex heterogeneous diseases such as SLE and juvenileonset SLE. [10][11][12][13] A retrospective study of previous longitudinal gene expression data from paediatric and adult SLE populations identified three stratified groups within each cohort with unique disease activity and trajectories, supporting strategies to identify clinically informative groups using immune profiling. 13 Another study applied three different machinelearning approaches, including knearest neighbours, generalised logistic models, and random forest models to predict disease activity in patients with SLE using wholegenome gene expression profiles.…”

“…To assess whether the juvenileonset SLE signature could be used to stratify patients with juvenileonset SLE further, kmeans clustering, an unsupervised machine learning algorithm was used. After screening, eight of 28 immune cell subsets were selected: total CD4, total CD8, CD8 effector memory (EM), CD8 naive, and invariant natural killer T cells; Bm1 and unswitched memory B cells; and total CD14 monocytes (figure 2B, C; appendix pp [11][12]. Based on these variables, kmeans clustering was done to stratify patients with juvenile onset SLE into four groups (group 1, n=10; group 2, n=21; group 3, n=21; group 4, n=15; figure 5A).…”

Early Prediction of Clinical Response to Etanercept Treatment in Juvenile Idiopathic Arthritis Using Machine Learning

“…Many advances have been made in medical prediction, such as assistant diagnoses, prognosis evaluation, and new drug development. For example, Nieuwenhove et al identified an immunological pattern associated with JIA subtypes using machine learning (Van Nieuwenhove et al, 2019); Motwani et al (2017) used machine learning to predict 5-year mortality in coronary artery disease patients. On the other hand, the increasing number of electronic medical record (EMR) data containing rich comprehensive information of patients such as examination and diagnosis, coupled with the development of machine learning, provides new opportunities for highperformance efficacy prediction model generation (Rahimian et al, 2018).…”

“…For example, Nieuwenhove et al. identified an immunological pattern associated with JIA subtypes using machine learning ( Van Nieuwenhove et al., 2019 ); Motwani et al. (2017) used machine learning to predict 5-year mortality in coronary artery disease patients.…”

A machine learning approach for precision diagnosis of juvenile-onset SLE

“…Recently, in depth computational analysis of large multi-omic datasets has accelerated understanding of complex heterogeneous diseases such as SLE/JSLE [12][13][14][15]. Machine learning (ML) is a subdivision of artificial intelligence that builds analytical models by learning by example and has been used in a wide range of clinical areas, including medical image classification and prediction [16], drug discovery by predicting the optimal pharmaceutical target [17], and building predictive models for disease diagnosis and prognosis [18].…”