Deep ensemble learning model for cervical cancer disease classification on image dataset
Sonam Juneja, Shikha Atwal, Reema Goyal, Bhoopesh Singh Bhati · Journal of Information and Optimization Sciences · 2025
Early detection is the hallmark of ensuring better patient outcomes and successful treatment for most forms of cancer. In our proposed model, we classify the cervical cancer disorders through the developed deep ensemble learning model that was trained on a large dataset of colposcopy photos. It is evident that combining deep learning with an ensemble methodology would make colposcopy data-driven cervical cancer screening more accurate and robust. The proposed model was experimented with various known performance measures such as sensitivity, specificity, accuracy, PPV, and NPV. The experimental outcome results find this ensemble deep learning model to perform outstandingly well with remarkable robustness and outstanding accuracy in ‘detecting’ cervical cancer over individual models. Integrating the Colposcopy Ensemble Network architecture developed to address the problem of cervical cancer detection, this model increases the overall sensitivity and accuracy. Thus, it is highly efficient given other performance standards and fostering the development of ensemble learning models in classifying various diseases based on medical images.