MUMBAI, India, Aug. 24 -- Intellectual Property India has published a patent application (202641095178 A) filed by Aleena M A; and Dr. Sangeetharadhakrishnan on August 06, 2026, for System And Method For Deep Learning-Based Automated Surgical Skill Assessment In Robotic- Assisted Urologic Oncologic Surgery.
Inventors include Aleena M A; and Dr. Sangeetharadhakrishnan.
The application for the patent was published on August 21, 2026, under issue no. 34/2026.
Abstract: The present invention discloses a system and method for deep learning-based automated surgical skill assessment in robotic-assisted urologic oncologic surgery. The system comprises a multi-modal data acquisition and fusion module receiving robotic surgical video, kinematic, force, and sensor data; a computer vision module employing a U-Net CNN for pixel-level robotic instrument segmentation and optical flow for trajectory tracking; a gesture recognition engine comprising a bidirectional LSTM and Temporal Convolutional Network for suturing gesture detection and classification; a feature extraction module; a Transformer-based skill classification engine classifying surgeon skill into Expert, Intermediate, and Novice categories; a Monte Carlo Dropout-based result validation and confidence scoring module; and a real-time feedback and skill assessment reporting module. The method acquires and synchronizes multi-modal data, segments instruments, recognizes suturing gestures, classifies skill level, validates confidence, and generates real-time corrective feedback and longitudinal performance analytics. The invention objectively replaces subjective manual assessment, eliminates inter-rater variability, and enables continuous intraoperative skill monitoring.
Disclaimer: Curated by HT Syndication.
