A Hybrid Deep Fused Learning Approach to Segregate Infectious Diseases

Humankind is facing another deadliest pandemic of all times in history, caused by COVID-19. Apart from this challenging pandemic, World Health Organization (WHO) considers tuberculosis (TB) as a preeminent infectious disease due to its high infection rate. Generally, both TB and COVID-19 severely affect the lungs, thus hardening job medical practitioners who can often misidentify these diseases current situation. Therefore, time need calls for an immediate meticulous automatic diagnostic tool that accurately discriminate diseases. As one preliminary smart health systems examine three clinical states (COVID-19, TB, normal cases), study proposes amalgam image filtering, data-augmentation technique, transfer learning-based approach, advanced deep-learning classifiers effectively segregate It first employed generative adversarial network (GAN) Crimmins speckle removal filter on X-ray images overcome issue limited data noise. Each pre-processed then converted into red, green, blue (RGB) Commission Internationale de l’Elcairage (CIE) color spaces which deep fused features are formed extracting relevant using DenseNet121 ResNet50. feature extractor extracts 1000 most useful finally fed two variants recurrent neural (RNN) precise discrimination three-clinical states. Comparative analysis showed proposed Bi-directional long-short-term-memory (Bi-LSTM) model dominated (LSTM) attaining overall accuracy 98.22% three-class classification task, whereas LSTM hardly achieved 94.22% test dataset.