MUMBAI, India, Aug. 10 -- Intellectual Property India has published a patent application (202611030631 A) filed by Indian Institute Of Technology Banaras Hindu University, Varanasi on March 13, 2026, for System And Method For Ai-Based Bppv Maneuver Guidance Using Head-Pose Detection.
Inventors include Pratap, Ajay; Joshi, Deepika; Dhiman, Neetu Rani; Sharma, Nikita; and Singh, Himanshu.
The application for the patent was published on July 31, 2026, under issue no. 31/2026.
Abstract: The present disclosure provides an AI-based, vision-driven system (100) and method (300) for automated rehabilitation of patients suffering from posterior canal Benign Paroxysmal Positional Vertigo (BPPV). The system (100) employs a standard camera unit (108) to capture real-time video of a patient’s head and upper body during the Gans repositioning maneuver. Key body landmarks, including the nose, eyes, ears, and shoulders, are extracted using a computer vision framework to compute head orientation angles, yaw, pitch, and roll, which are stabilized through a Kalman filtering algorithm. The system (100) estimates the patient’s distance from the camera unit (108) to maintain optimal tracking, validates posture for each maneuver step, and provides synchronized bilingual audio and visual guidance, including step instructions, progress indicators, angle data, and countdown timers. Correct posture triggers automated step progression, and completion feedback confirms successful rehabilitation. The disclosure enables autonomous, sensor-free, real-time BPPV rehabilitation at home or in clinical settings, improving accuracy, safety, accessibility, and patient adherence.
Disclaimer: Curated by HT Syndication.
