Identification of Systemic Sclerosis through Machine Learning Algorithms and Gene Expression

Systemic sclerosis (SSc) is an autoimmune, chronic disease that remains not well understood. It believed the cause of illness a combination genetic and environmental factors. The evolution also greatly varies from patient to patient. A common complication illness, with associated higher mortality, interstitial lung (ILD). We present in this paper algorithm (using machine learning techniques) it able identify, 92.2% accuracy, patients suffering ILD-SSc using gene expression data obtained peripheral blood. were public sources (GEO accession GSE181228) contains for 134 at initial stage as follow up date (12 months later) 98 these patients. Additionally, there are 45 control (healthy) cases. identified 172 genes might be involved illness. These appeared all 20 most accurate classification models among total half million estimated. Their frequency suggest they related some degree. proposed algorithm, besides differentiating between patients, was distinguish different variants (diffuse variants). This can have significance treatment point view. type prognosis.