Hybrids of Support Vector Regression with Bald Eagle Search Optimizer for Diagnosing Patients Suspected of Sarcoidosis

Searching for a reliable indicator of treatment response in sarcoidosis remains a challenge.The use of the soluble interleukin 2 receptor (sIL-2R) as a measure of disease activity has been proposed by researchers.The aim of this study was to develop a machine learning model to predict sIL-2R levels based on serum angiotensin-converting enzyme (ACE) levels of patients that can be used in evaluating lung function.This new sIL-2R prediction model combines Support Vector Regression (SVR) with the Bald Eagle Search (BES) algorithm.However, BES has only a few adjustable parameters, simple searching technique and produces consistent results all the time.In this way, the hyper-parameters of SVR are optimized by BES that eventually improved the accuracy of predicted sIL-2R values.The hybrid SVR-BES model was evaluated using a dataset consisting of medical records of patients suspected sarcoidosis focusing specifically on soluble interleukin 2 receptor (sIL-2R) levels and angiotensin converting enzyme (ACE) levels.These other methodologies include Hybrid SVR with Firefly Algorithm (SVR-FFA), Decision Tree (DT), SVR with Gray Wolf Optimization (SVR-GWO), Random Forest (RF).According to these results, Root Mean Square Error (RMSE) is 0.5615 while correlation coefficient (R) is 0.9457 thereby indicating higher accuracy compared to other methods as presented by hybrid SVR-BES model.