MUMBAI, India, Jan. 8 -- Intellectual Property India has published a patent application (202511104256 A) filed by Lovely Professional University, Phagwara, Punjab, on Oct. 29, 2025, for 'computer-implemented system for automated detection of autism spectrum disorder (asd) from facial images.'
Inventor(s) include Dr. Alok Jain; Meghna Gupta; Dr. Vishal Sarin; and Dr. Amit Dutt.
The application for the patent was published on Dec. 12, under issue no. 50/2025.
According to the abstract released by the Intellectual Property India: "A computer-implemented system for automated detection of autism spectrum disorder (ASD) from facial images comprising a preprocessing module 101, receives facial images, normalizes them using face detection, alignment, and illumination correction techniques, a data augmentation process, performs rotation, flipping, scaling, and brightness adjustment to enhance dataset diversity and model robustness, a feature extraction module 102, employs pre-trained convolutional neural network architectures including VGG16, VGG19, inceptionV3 103, VGGFace, and MobileNet 104, extracts discriminative facial features, a classification module 105, trained on labeled datasets of ASD and non-ASD subjects, analyzes extracted features and output 107 predictions indicating the presence of ASD characteristics, an explainability module, utilizes Local Interpretable Model-Agnostic Explanations, generates visual interpretations and heatmaps highlighting facial regions influencing the prediction, a clinician-assistance interface, provides interpretable results alongside prediction output 107, supports diagnostic decision-making, reduces reliance on manual assessments, and facilitates early detection."
Disclaimer: Curated by HT Syndication.
