Publication

Abstract : Medical imaging technology in modern healthcare is essential to ensure prompt and accurate diagnosis. This research addresses the urgent need for better diagnostic tools by proposing a unique method for identifying images from computed tomography (CT) scans. By combining the advantages of VGG16 and EfficientNet designs, the proposed hybrid model achieves 99.24% accuracy in distinguishing normal, cystic, tumor and abnormal conditions. This study shows the success of each architectural interaction while examining its contribution. In addition to the accuracy achieved, the proposed model shows the potential to improve the diagnostic skills, medical investigation and validation in the field of medical image processing. Furthermore, the proposed hybrid model plays a significant role in accelerating the diagnostic process by reducing the time consumed by the traditional medical diagnosis process, further to achieved higher accuracy in differentiating various medical conditions in CT scans. Finally, this study illustrates the proposed model’s enhanced overall performance by analyzing the architectural relationships and also states the future research directions in medical image processing.