MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641049023 A) filed by Seshadri Rao Gudlavalleru Engineering College; Dhana Lakshmi Lanka; Sayyad Aafrin; Raavi Lakshmi Sowjanya; Edulamudi Princy Dayana; and Shaik Saida, Gudlavalleru, Andhra Pradesh, on April 17, for 'kidney tumor identification from ct images using an integrated 3d cnn and transformer-based framework.'
Inventor(s) include Seshadri Rao Gudlavalleru Engineering College; Dhana Lakshmi Lanka; Sayyad Aafrin; Raavi Lakshmi Sowjanya; Edulamudi Princy Dayana; and Shaik Saida.
The application for the patent was published on May 1, under issue no. 18/2026.
According to the abstract released by the Intellectual Property India: "Automated kidney abnormality detection has become an important tool in modern healthcare for early diagnosis and effective clinical decision-making. This project focuses on developing a system that analyzes kidney CT scan images and predicts disease categories using advanced deep learning techniques. Unlike traditional diagnostic methods that rely on manual interpretation by radiologists, the proposed system utilizes a hybrid deep learning approach combining convolutional neural networks and transformer-based models to capture both spatial and global features from CT images, ensuring improved accuracy and robustness. The system reduces dependency on manual analysis and provides a scalable solution suitable for real-time clinical applications. The proposed system employs a hybrid architecture consisting of ConvNeXt and Vision Transformer models for feature extraction. Input CT scan images are preprocessed through resizing, normalization, and data augmentation techniques to improve data quality and model performance. The hybrid model effectively extracts spatial patterns and contextual relationships within the images, enabling accurate identification of kidney conditions such as Normal, Cyst, Stone, and Tumor. The system generates class probabilities along with confidence scores, supporting reliable multi-class classification. To further improve classification performance, the system integrates Fuzzy Neural Network (FNN) and Generalized Fuzzy Neural Network (GFNN) classifiers, which help handle uncertainty and variations in medical image data. The system is designed with a user-friendly interface that allows users to upload CT scan images and receive instant predictions. Additionally, the results are stored for future reference and analysis. Experimental results demonstrate high accuracy and efficient performance. Overall, the proposed framework provides an accurate, reliable, and efficient solution for kidney abnormality detection, contributing to advancements in computer-aided diagnosis and healthcare support systems."
Disclaimer: Curated by HT Syndication.
