MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641084443 A) filed by Vallurupalli Nageswara Rao Vignana Jyothi Institute Of Engineering And on July 09, 2026, for Deep Learning-Based System For Automated Detection And Localization Of Cervical Spine Fractures From Computed Tomography Images.
Inventors include Mrs. Soujanya Ambala; Naini Sudeepthi; Anuhya Srestha; Ruthvij Reddy; and Neha.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: ABSTRACT [0023] The present invention discloses an advanced deep learning-based system and method for the automated detection and localization of cervical spine fractures from CT scan images in DICOM format. The proposed Cervical Spine Fracture Detection System (CSFDS) integrates sophisticated image preprocessing techniques with a customized DenseNet121 convolutional neural network architecture enhanced by global average pooling, dropout layers, and multi-label classification dense layers corresponding to vertebrae C1 through C7. This system is designed to assist radiologists by providing rapid, accurate, and reliable fracture predictions in real-time, significantly reducing diagnostic time and human error in emergency medical settings. The invention further incorporates an intuitive web-based graphical user interface built with Flask, HTML and CSS, enabling seamless upload of DICOM files and instant visualization of results indicating whether a fracture is present and the specific vertebrae affected. Through extensive training on the RSNA 2022 dataset with optimized callbacks for early stopping, model checkpointing, and learning rate reduction, the system achieves superior performance metrics, including approximately 94.75% accuracy, while maintaining strong generalization capabilities. This novel integration of domain-specific preprocessing, tailored neural network modifications, and user-friendly deployment offers a scalable solution that enhances clinical decision-making and patient outcomes in trauma care.
Disclaimer: Curated by HT Syndication.
