Inflammatory disease classification using deep learning and machine learning features

One type of chronic illness is inflammation-related illnesses. Our immune system can be impacted in a variety of ways, which can lead to inflammation. Studies have shown this to be widespread among pathogens and have highlighted the serious impact these problems can have on a person's ability to carry out daily activities. This necessitates the identification of the various disease types and their effects. This study considers 150 MR images from the AANLIB database, including patients with AIDS dementia, herpes encephalitis, lymphoblastic encephalopathy, multiple sclerosis, cerebral toxoplasmosis, and Creutzfeldt-Jakob disease. The Orange data mining tool utilizes SqueezeNet and Inception V3 deep learning models to extract features at the image level. To determine the SVM's classification accuracy for anomalies, the regularizer is varied. Findings indicate that the extracted features successfully differentiate abnormal brain tissue from healthy tissue.