MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202641072046 A) filed by Dr. S. Ganeshmoorthy; Karpagam College Of Engineering; Vishwa A; Priyadharshini G; Abishek S; Vinitha P; Arul Kumaran S; Jalagandeswaran V; Kavipriyadharsan P; and Barath Aswin S on June 10, 2026, for Detection Of Cancer In Human Blood Cell Using Cnn Classification Algorithm.
Inventors include Dr. S. Ganeshmoorthy; Vishwa A; Priyadharshini G; Abishek S; Vinitha P; Arul Kumaran S; Jalagandeswaran V; Kavipriyadharsan P; and Barath Aswin S.
The application for the patent was published on June 19, 2026, under issue no. 25/2026.
Abstract: DETECTION OF CANCER IN HUMAN BLOOD CELL USING CNN CLASSIFICATION ALGORITHM Abstract: Blood cancer is one of the most serious diseases affecting the production and function of blood cells. Early and accurate detection is essential for improving patient survival rates and enabling timely treatment. Traditional diagnostic methods often require significant time and expert analysis, creating a need for automated and efficient detection systems. This paper presents a deep learning-based approach for the detection and classification of human blood cancer cells using a Convolutional Neural Network (CNN) with the MobileNetV2 architecture. Transfer learning is employed to improve classification performance while reducing training time and computational complexity. The proposed model classifies blood cell images into four categories: Benign, Early, Pre, and Pro. Image preprocessing and data augmentation techniques are applied to enhance the quality and diversity of the dataset. Experimental results demonstrate that the model achieved a training accuracy of 95.97% and a validation accuracy of 94.76%, indicating strong classification capability and good generalization performance. The proposed system provides a reliable and efficient solution for automated blood cancer detection and has the potential to assist medical professionals in clinical diagnosis and decision-making. Blood cancer is one of the most dangerous diseases affecting the circulatory and immune systems of the human body. Leukemia is a common type of blood cancer in which abnormal white blood cells grow uncontrollably inside the bone marrow and interfere with the production of healthy blood cells. As the disease progresses, patients may experience reduced immunity, fatigue, anemia, bleeding disorders, and infections due to the imbalance of blood cell production. Early detection of leukemia is extremely important because timely treatment can improve survival rates and reduce disease severity. In conventional clinical environments, leukemia diagnosis mainly depends on microscopic examination of blood smear samples and laboratory analysis performed by medical experts. This manual procedure requires significant experience and concentration because cancerous blood cells often resemble normal cells in appearance. In addition, manual diagnosis becomes difficult when large numbers of samples must be analyzed continuously. The advancement of Artificial Intelligence (AI) and medical image processing technologies has introduced new possibilities for automated disease detection systems. Deep Learning techniques have become highly effective in healthcare applications because of their capability to learn complex patterns directly from image data. Among Deep Learning methods, Convolutional Neural Networks (CNNs) have shown remarkable performance in image classification, object detection, and medical diagnosis tasks. CNN architectures automatically extract both low-level and high-level image features such as texture, shape, and structural abnormalities without requiring manual feature engineering. This capability makes CNN models highly suitable for leukemia detection using blood smear images. In this research work, a CNN-based blood cancer classification framework using MobileNetV2 architecture is proposed. The system analyzes Peripheral Blood Smear images and classifies blood cells into different leukemia categories. The objective of this work is to improve classification accuracy, reduce computational complexity, and provide an efficient automated support system for healthcare professionals.
Disclaimer: Curated by HT Syndication.
