MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641057639 A) filed by Sai Krishna Prasad Munnaluri; and Abdul Nazeer K. A on May 06, 2026, for System And Method For Detecting Autism Spectrum Disorder From Facial Images Using Convolutional Neural Networks And Related Methods.
Inventors include Sai Krishna Prasad Munnaluri; and Abdul Nazeer K. A.
The application for the patent was published on June 26, 2026, under issue no. 26/2026.
Abstract: A method and system for detecting Autism Spectrum Disorder (ASD) from facial images using a convolutional neural network (CNN) trained on augmented image data are described. The process involves acquiring facial images, applying data augmentation, and training a CNN on these samples. Facial landmarks are detected to calculate head orientation parameters, which are used to geometrically transform images to a neutral pose. The images are then systematically rotated to create a pose- variant dataset for training the CNN to recognize facial features invariant to head orientation. The trained model can be converted to TensorFlow Lite for deployment on mobile applications, allowing on-device inference and user interaction through a mobile interface.
Disclaimer: Curated by HT Syndication.
