MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202641126663 A) filed by Dr M Elavarasi; Subbaiahgari Ramachandraiah Ajitha; Swatantra Agrawal; Amshavalli M; Er K. Peermohamed; Pradeepkumar G; B. Soujania; Jenisha J; Pragathi N; Dr Karthikeyan M V; Mr. Arpan Debnath; and S. Nageswari on October 23, 2026, for Machine Learning-Based Early Detection And Classification Of Cervical Cancer Using Pap Smear Images.
Inventors include Dr M Elavarasi; Subbaiahgari Ramachandraiah Ajitha; Swatantra Agrawal; Amshavalli M; Er K. Peermohamed; Pradeepkumar G; B. Soujania; Jenisha J; Pragathi N; Dr Karthikeyan M V; Mr. Arpan Debnath; and S. Nageswari.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: The present invention relates to a machine learning-based system for early detection and classification of cervical cellular abnormalities using digitized Pap smear images. The system includes a module for capturing images, a module for evaluating image quality, a module for identifying cellular regions, a module for extracting features, an engine for fusing cytological evidence, a classification engine, and a module for aggregating specimen-level data. Cellular feature representations and image-quality indicators are integrated to aid in multiclass cytological classification. A module for assessing uncertainty evaluates confidence in predictions, quality of images, uncertainty in segmentation, and sufficiency of evidence to identify results that need reprocessing or expert review. A module for generating reports offers anticipated cytological classifications along with related image-based proof. A managed model maintenance component aids in performance assessment and enhancement through validated annotations and quality-assured results. The system aims to support qualified healthcare practitioners in analyzing Pap smear images and prioritizing specimens that need additional evaluation. FIG.1
Disclaimer: Curated by HT Syndication.
