Covid-19 tespiti için değiştirilmiş geliştirilmiş yapay ekosistem kullanılan ağırlık optimizasyonu CNN+MLP

Machine learning has been at the forefront of medical research efforts. It serves as a tool to assist researchers and practitioners in making an informed decision in the face of COVID-19 to forecast the spread of such diseases and identify the infected with much higher confidence. This study outlines the process of designing and refining an AI framework to classify COVID-19 from Pneumonia using X-ray scans combined with textual clinical data. The primary objective of this study is to merge multiple types of neural networks and study the effect of using metaheuristic algorithms in optimizing the weights of the neural network. The proposed framework also utilizes a lung segmentation process using a pre-trained ResNet34 model to generate a mask for each lung to eliminate the unnecessary features that might affect the result. The training data consists of 579 segmented X-ray (AP, PA views) images of COVID-19 and Pneumonia with each patient's textual medical data that include age and gender. The proposed framework achieved an accuracy of 97.85% compared to 94.32% without weight optimization. Furthermore, an extensive comparison with several other architectures from the literature was used to evaluate the model viability in detecting COVID-19......