MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641081093 A) filed by Mlr Institute Of Technology on July 01, 2026, for White Blood Cell Classification Using Convolutional Neural Networks.
Inventors include Mrs. A Laxmi Prasanna; Mr. P Shashikiran Reddy; Mr. A Yuvateja; and Mr. T Rishit.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: In this invention, “White Blood Cell Classification Using Convolutional Neural Networks” is disclosed as an intelligent medical image analysis system designed to automatically identify and classify white blood cells from microscopic blood smear images using deep learning techniques. The system acquires blood smear images through digital microscopes, laboratory imaging devices, or medical image repositories and performs preprocessing operations including image enhancement, normalization, noise reduction, segmentation, and cell isolation. The processed images are supplied to a Convolutional Neural Network (CNN)-based classification engine that automatically extracts hierarchical features related to cellular morphology, nucleus structure, cytoplasm characteristics, texture patterns, and staining properties. The extracted features are analyzed to classify white blood cells into categories including neutrophils, lymphocytes, monocytes, eosinophils, basophils, and other clinically relevant leukocyte types. The invention further incorporates model training, validation, and optimization mechanisms utilizing annotated hematological datasets to improve classification accuracy, robustness, and reliability. Classification results are associated with confidence scores and stored in a centralized database for future reference, reporting, and analytical purposes. A visualization and reporting module presents classification outcomes through user-friendly dashboards, graphical representations, and diagnostic reports that assist healthcare professionals in clinical decision-making. The system may additionally support adaptive learning, disease risk assessment, laboratory workflow automation, and integration with hospital information systems and laboratory information management systems. The proposed framework can be deployed in hospitals, diagnostic laboratories, research institutions, pathology centers, and healthcare facilities requiring efficient hematological analysis. By integrating image acquisition, preprocessing, CNN-based feature extraction, automated multiclass white blood cell classification, result visualization, and diagnostic support within a unified framework, the invention significantly improves diagnostic accuracy, reduces manual workload, enhances laboratory productivity, supports early detection of hematological disorders, and advances the application of artificial intelligence in modern healthcare diagnostics.
Disclaimer: Curated by HT Syndication.
