MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641068305 A) filed by Sri Sairam Engineering Collegeautonomous on June 01, 2026, for Nutri-Net:cross Anatomical Contrastive Alignment For Non Invasive Micronutrient Deficiency Screening.
Inventors include Haripriya V; Ishvarya K; Yaashica Sakthivel; Dr. G. Tony Santhosh; and Dr. S. Brlndha.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: The present invention discloses a non-invasive,. artificial intelligence-based multi-modal diagnostic system and method for early detection and screening of micronutrient deficiencies using physiological images obtained from anatomical regions including the tongue, eyes, skin, and nails. The invention employs a deep learning architecture with parallel feature extraction branches and a Cross- Anatomical Contrastive Alignment (CACA) mechanism for integrating heterogeneous anatomical features into a unified symptomatic feature space to improve diagnostic accuracy and reliability. The system further utilizes a two-phase training strategy involving unimodal and clinically curated paired datasets for robust feature learning and . classification of deficiencies including Vitamin A, Vitamin Bl2, and Vitamin C deficiencies. Explainable Artificial Intelligence (XAI) techniques and Bayesian uncertainty estimation are integrated to provide interpretable and reliable predictions. The proposed framework is lightweight, scalable, and mobile- compatible, enabling accessible nutritional deficiency screening without dependence on invasive biochemical testing or specialized laboratory infrastructure. .
Disclaimer: Curated by HT Syndication.
