MUMBAI, India, Feb. 13 -- Intellectual Property India has published a patent application (202641008785 A) filed by Kanneboina Rakshitha; Muntha Raju; Kondragunta Siva Rama Krishna; Miryala Manoj; and Satla Saicharan, Hyderabad, Telangana, on Jan. 28, for 'automated lung cancer detection from ct images using hybrid glcm-cnn features and svm classification with roi localization.'
Inventor(s) include Kanneboina Rakshitha; Muntha Raju; Kondragunta Siva Rama Krishna; Miryala Manoj; and Satla Saicharan.
The application for the patent was published on Feb. 13, under issue no. 07/2026.
According to the abstract released by the Intellectual Property India: "The invention relates to an automated system for lung cancer detection from computed tomography (CT) images using hybrid GLCM CNN features and support vector machine (SVM) classification with region of interest (ROI) localization. CT images are pre processed and lung regions are segmented to identify candidate nodule ROIs, which are extracted as fixed size patches. For each ROI, Gray Level Co occurrence Matrix (GLCM) based texture features and deep features from a convolutional neural network (CNN) backbone are computed and fused into a hybrid feature vector, optionally after dimensionality reduction. An SVM classifier, trained on labeled nodule data, operates on the hybrid features to assign benign or malignant labels and malignancy scores. The system visualizes suspicious ROIs by overlaying bounding boxes and scores on CT slices, supporting radiologist decision making. By combining ROI localization, complementary handcrafted and deep features, and SVM classification, the invention improves accuracy, robustness, and efficiency of lung cancer detection in CT based screening and diagnostic workflows."
Disclaimer: Curated by HT Syndication.
