Deep Learning Enabled Segmentation, Classification and Risk Assessment of Cervical Cancer
Abdul Samad Shaik, Shashaank Mattur Aswatha, Rahul Jashvantbhai Pandya
TL;DR
This paper presents a lightweight, end-to-end framework for cervical cancer cell analysis that combines segmentation, classification, and risk assessment using the SIPaKMeD dataset. It introduces a Multi-Resolution Fusion Deep Convolutional Network (MRF-DCN) for efficient classification and a Multi-Task UNet for joint segmentation and classification, complemented by a probabilistic risk assessment (RA) pipeline based on 64‑D features. The results show competitive performance with far fewer parameters than large models (e.g., ~1.7M vs. VGG‑19) and demonstrate IoU scores around 0.83 and classification accuracy near 90%, with RA achieving >95% accuracy in feature-based predictions and high AUC values. Together, these components offer a practical, scalable approach for automated cervical cancer screening and prognosis, with potential applicability to other medical imaging tasks via dynamic multi-scale extensions.
Abstract
Cervical cancer, the fourth leading cause of cancer in women globally, requires early detection through Pap smear tests to identify precancerous changes and prevent disease progression. In this study, we performed a focused analysis by segmenting the cellular boundaries and drawing bounding boxes to isolate the cancer cells. A novel Deep Learning (DL) architecture, the ``Multi-Resolution Fusion Deep Convolutional Network", was proposed to effectively handle images with varying resolutions and aspect ratios, with its efficacy showcased using the SIPaKMeD dataset. The performance of this DL model was observed to be similar to the state-of-the-art models, with accuracy variations of a mere 2\% to 3\%, achieved using just 1.7 million learnable parameters, which is approximately 85 times less than the VGG-19 model. Furthermore, we introduced a multi-task learning technique that simultaneously performs segmentation and classification tasks and begets an Intersection over Union score of 0.83 and a classification accuracy of 90\%. The final stage of the workflow employs a probabilistic approach for risk assessment, extracting feature vectors to predict the likelihood of normal cells progressing to malignant states, which can be utilized for the prognosis of cervical cancer.