Application of Machine Learning Models in Systemic Lupus Erythematosus

Abstract

Systemic Lupus Erythematosus (SLE) is a systemic autoimmune disease and is extremely heterogeneous in terms of immunological features and clinical manifestations. This complexity could result in a delay in the diagnosis and treatment introduction, with impacts on long-term outcomes. In this view, the application of innovative tools, such as machine learning models (MLMs), could be useful. Thus, the purpose of the present review is to provide the reader with information about the possible application of artificial intelligence in SLE patients from a medical perspective. To summarize, several studies have applied MLMs in large cohorts in different disease-related fields. In particular, the majority of studies focused on diagnosis and pathogenesis, disease-related manifestations, in particular Lupus Nephritis, outcomes and treatment. Nonetheless, some studies focused on peculiar features, such as pregnancy and quality of life. The review of published data demonstrated the proposal of several models with good performance, suggesting the possible application of MLMs in the SLE scenario.

Machine learning approaches to identify systemic lupus erythematosus in anti-nuclear antibody-positive patients using genomic data and electronic health records

“…ML models have been extensively explored for diagnosing SLE, defining clinical phenotypes, determining outcomes, and informing therapeutic decisions [ 23 ]. Previous studies utilizing ML to facilitate SLE diagnosis have employed diverse input data, including EHRs, genetic biomarkers, proteomics, lipidomes, or a combination of these data types [ 23 , 24 , 27 – 32 ]. This study is the first to incorporate genome-wide SNPs, PRS, and EHRs in an ML analysis.…”

“…This study is the first to incorporate genome-wide SNPs, PRS, and EHRs in an ML analysis. Moreover, ML algorithms for diagnostic purposes in previous studies included RF, LASSO, SVM, LR, XGB, and Partial Least Square [ 23 , 24 , 27 – 32 ]. This study is the first to attempt to compare the diagnostic accuracy among six ML models.…”

“…To forecast the likelihood of patients developing SLE within one year from the index date, six ML algorithms, LR, RF, SVM, LGBM, GTB, and XGB, were employed [ 11 , 23 ]. The entire dataset ( n = 2,838) was randomly divided into a training set (80%) and a testing set (20%), maintaining proportional representation across both sets.…”

Machine learning: Predicting hospital length of stay in patients admitted for lupus flares

“…ML has been applied across various aspects of SLE, from diagnostics to treatment outcomes. 16,17 In diagnostics, ML has been applied towards redefining SLE criteria. For example, Adamichou et al developed a new diagnostic criterion and a severity scoring system using ML, achieving 94.8% accuracy.…”

Helios as a Potential Biomarker in Systemic Lupus Erythematosus and New Therapies Based on Immunosuppressive Cells

“…Nevertheless, this only represents a small piece in the complete immune system puzzle and many questions remain due to the difficulties in fitting all the pieces together. Resolving these questions could open new opportunities for the treatment of these heterogeneous diseases, among which abundant subclinical cases exist, and new findings are driving constant changes in diagnosis criteria (e.g., in SLE) [ 143 ]. Hence, “the search for a specific SLE biomarker” [ 144 ] is still one of the most important challenges in relation to this disease.…”