MULTICLASS_RESP_DNN: A Novel Deep Learning Architecture for COVID, Pneumonia and Tuberculosis Detection from Radiology Images

AI, particularly machine learning, is increasingly utilized in healthcare for disease diagnosis. This paper presents a novel deep learning architecture, MULTICLASS_RESP_DNN, for the detection of respiratory diseases such as COVID-19, pneumonia, and tuberculosis from chest X-ray images. The model aims to address the scarcity of competent radiologists by providing automated disease classification, aiding in timely and accurate diagnosis. We utilize a comprehensive dataset sourced from Kaggle, comprising chest X-ray images of normal cases, COVID-19, pneumonia, and tuberculosis. The proposed model undergoes rigorous evaluation, considering various performance metrics, interpretability, and ethical implications. Experimental results demonstrate the effectiveness and potential of the MULTICLASS_RESP_DNN model in respiratory disease detection. Our method can accurately detect 95.52% of abnormal X-rays, including COVID-19, pneumonia and tuberculosis.