Hybrid Approach for Medical Decision-Making: Integrating ResNet-Darknet19 Based Transfer Learning with Radiomics Features for COVID-19 Classification

This study presents a novel hybrid approach for the classification of COVID-19 cases employing a combination between radiomics features and Res-Net-Darknet19 architecture based transfer-learning model to enhance the accuracy of COVID-19 diagnosis. The primary phase of the model demonstrated excellent performance, with 97.33% accuracy, 97.99% precision, 97.40% recall, and a F1-score of 97.69%. The model correctly identified 352 COVID-19 cases and 487 normal cases, predicting 10 false- approving and 13 false-negative outcomes. The retrained model showed remarkable improvement, achieving 98.02% accuracy, 98.21% precision, 98.21% recall, and a F1-score of 98.21%, demonstrating its ability to categorize radiomics features. Experimental results demonstrate that our hybrid approach outperforms traditional methods, achieving higher classification accuracy and robustness. This method shows tons of promise for increasing the effectiveness and reliability of COVID-19 diagnosis in clinical settings.