Cerviscan: Advancing Cervical Cancer Detection through Deep Learning Innovations
Hariharan Ramesh, Abdulla Al Serkal, Houssein Kanso, Noora Aljallaf, Jinane Mounsef · 2024
Cervical cancer ranks as a leading cause of mortality among women, with the conventional pap smear-based detection posing significant challenges due to its labor-intensive and time-consuming nature. This study explores how Artificial Intelligence (AI) technology can be integrated into cervical cancer diagnostics to alleviate the workload of pathologists by streamlining and improving the detection process. We used deep learning (DL) models, specifically VGG-19, VGG-16, Inception, and Resnet-50, to facilitate early identification of precancerous cervical cells from pathological slide images. Our research includes a detailed comparison of these models’ performances, highlighting the Inception model’s exceptional accuracy of 98% in early detection efforts. Finally, we propose Cerviscan, a comprehensive system for early cancer detection.