Automated Diagnosis of "Rheumatoid arthritis" based on CNN
Mehveen Mehdi Khatoon, Bhaludra R Nadh Singh, Maddala Sree Harshita, K. Sreeja, Syamala Sreeya Reddy, Jajula Sri Latha · 2023
Diseases like "rheumatoid arthritis" are chronic inflammatory conditions that wreak havoc on the body's tissue, most noticeably the joints. An efficient system analysis is necessary for manual "Rheumatoid arthritis" identification and diagnosis, especially in the pre-diagnostic stages. Using image processing and a convolutional neural network, this study aims to create an intelligent system capable of identifying hand "Rheumatoid arthritis" cases. There are two primary parts to the system. "Image processing" refers to the procedure that begins with the processing of images. Pre-processing, image segmentation, and Gabor filter-based feature extraction are all examples of such methods. In the second stage, the retrieved characteristics are sent into a neural convolution network, which determines whether or not the shown hands are healthy (arthritic). Using normal and abnormal hand pictures, the "CNN" algorithm does classification. The identification rate in the experiments was 83.5% when the same number of photos was used as in the test set.