MUMBAI, India, July 30 -- Intellectual Property India has published a patent application (202641087021 A) filed by Vallurupalli Nageswara Rao Vignana Jyothi Institute Of Engineering And Technology on July 16, 2026, for System And Method For Automated White Blood Cell Classification Using Convolutional Neural Networks And Transfer Learning-Based Medical Image Analysis.
Inventors include Mrs. Soujanya Ambala; M. Varshini; M. Deeksha; Md Aman Mouzan; and M. Geethika.
The application for the patent was published on July 24, 2026, under issue no. 30/2026.
Abstract: ABSTRACT The present invention discloses an Automated White Blood Cell Classification System (AWBCS) utilizing Convolutional Neural Networks for precise, rapid, and scalable differentiation of four primary leukocyte types Eosinophil, Lymphocyte, Monocyte, and Neutrophil from microscopic blood smear images. The system processes the publicly available Kaggle Blood Cell Image Dataset comprising approximately 12,500 labeled JPEG images through systematic preprocessing involving resizing to 64×64 pixels and pixel normalization. A custom CNN architecture featuring multiple convolutional and max-pooling layers for hierarchical feature extraction is complemented by a fine-tuned MobileNetV2 transfer learning variant, achieving validation accuracies of approximately 73% and 80% respectively after training on GPU-accelerated environments. The trained models are seamlessly integrated into a Flask-based RESTful backend API, paired with a responsive HTML/CSS/JavaScript frontend that enables medical professionals to upload images and receive instantaneous classification results with confidence scores. This end- to-end deployable framework addresses the inefficiencies and variability of manual differential counts, offering a practical assistive tool for early diagnosis of infections, leukemias, and inflammatory conditions while maintaining computational efficiency suitable for clinical and resource-constrained settings.
Disclaimer: Curated by HT Syndication.
