Published Paper Details:

As a result, AI methods are rapidly being applied to healthcare issues. In recent years, scientists have employed deep learning methods to categorize the severity of Chikungunya infection. Overfitting and hyper-parameter tweaking are problems that arise when using these methods, though. Methods: In this study, we propose a cyber-physical system (CPS) powered by artificial intelligence for determining how severe cases of Chikungunya are. Computational algorithms are combined with the CPS system's physical components to improve performance. Random forest is used to create a model for determining how severe cases of Chikungunya sickness are (RF). Overfitting and slow computing are problems plaguing RF, however, because of its complicated design and huge number of connection weights. Thus, we propose a genetic method with adaptive crossovers to be used in a developing RF model (ACGA). ACGA has the ability to effectively optimize RF design, leading to improved outcomes and increased computing speed. Multiple tests are run on the Chikungunya illness dataset. To sum up, the examination of performance reveals that ACGA-RF is superior to competing models in terms of F-measure, accuracy, sensitivity, and specificity. Patients who reside distant from hospitals will still have access to medical care according to the CPS proposal.