Novel computational model enhances early detection of cervical cancer

New Delhi: For the advancement of women’s health, scientists at the Institute of Advanced Study in Science and Technology (IASST), an autonomous institute under the Department of Science and Technology (DST), have developed a novel computational model aimed at improving the early detection of cervical cancer.

The new model focuses on accurately diagnosing cervical dysplasia, an abnormal growth of cells on the cervix surface that can lead to cancer. Accurate identification and classification of these cellular patterns are crucial for effective diagnosis and management of the condition.

Dr. Lipi B. Mahanta and her research team at IASST have pioneered this machine learning (ML) framework, which excels in real-world applications by achieving unmatched accuracy with minimal computation time. The team’s extensive experimentation with various color models, transformation techniques, feature representation schemes, and classification methods culminated in this advanced ML model. The comprehensive analysis and experimentation aimed to identify the optimal combination for detecting cervical dysplasia.

The model’s efficacy was validated using two datasets: one from healthcare centers in India and another publicly available dataset. By employing Non-subsampled Contourlet Transform (NSCT) and the YCbCr color model for image processing, the model achieved an impressive average accuracy of 98.02%. These findings, published in the journal “Mathematics” by MDPI, underscore the model’s potential to revolutionize cervical dysplasia detection.