MUMBAI, India, Aug. 10 -- Intellectual Property India has published a patent application (202641089489 A) filed by Mrs. Aishwarya. N. Kumar; and Mrs. Meenakshi Sundaram A on July 22, 2026, for A Hybrid Deep Learning And Fuzzy Logic Based System For Ovarian Cancer Subtype Classification And Clinical Recommendation Using Histopathological Images.
Inventors include Mrs. Aishwarya. N. Kumar; and Mrs. Meenakshi Sundaram A.
The application for the patent was published on July 31, 2026, under issue no. 31/2026.
Abstract: The invention provides a hybrid computational framework for automated classification of ovarian cancer subtypes using an integrated deep learning and fuzzy logic– based architecture. The system comprises an image pre-processing module, a multi-branch feature extraction unit including ResNet50, EfficientNetB0 and DenseNet121, and an auxiliary Vision Transformer module configured to model long-range spatial relationships in histopathological images. Extracted features are aggregated using an ensemble averaging mechanism and classified through a fully connected softmax layer. The framework is trained using sparse categorical cross- entropy loss with an Adam optimizer. A fuzzy logic–based decision support module generates clinical recommendations corresponding to the identified cancer subtype. The invention enables accurate identification of ovarian cancer subtypes including HGSC (High Grade Serous Carcinoma),LGSC (Low Grade Serous Carcinoma), CC (Cervical Cancer), EC(Endometrial Carcinoma) and MC (Metastatic Carcinoma), thereby supporting early diagnosis and personalized treatment planning.
Disclaimer: Curated by HT Syndication.
