MUMBAI, India, Aug. 10 -- Intellectual Property India has published a patent application (202641090832 A) filed by Udaya Sri Vangipuram; Vangipuram Radhakrishna; and V. Sravan Kiran on July 27, 2026, for “system And Method For Automated Disease Diagnosis From Medical Images Using Spectrogram- Based Matrix Structural Feature Extraction”.
Inventors include Udaya Sri Vangipuram; Vangipuram Radhakrishna; and V. Sravan Kiran.
The application for the patent was published on July 31, 2026, under issue no. 31/2026.
Abstract: A system and method for automated disease diagnosis from medical images are disclosed. One or more medical images acquired from a diagnostic imaging device (500) are received and preprocessed to generate standardized medical images (420). The standardized medical images are transformed into an Andrews curve representation and a spectrogram representation (450) using a time-frequency transformation. Image descriptors comprising pixel intensity descriptors and matrix structural descriptors are extracted from the standardized medical images and the spectrogram representation. The extracted descriptors are combined to generate an augmented feature vector (470), which is processed to determine a disease classification result. Diagnostic information is generated through an output interface (280) and may be communicated through a communication network (510) between a server (520) and one or more user devices (530). The invention provides an integrated framework for automated analysis and diagnosis of medical images.
Disclaimer: Curated by HT Syndication.
