Intellectual Property India Publishes Patent Application for Deep Learning Based System For Automated Classification Of Leukemic Lymphoblast Subtypes

MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202621094623 A) filed by Symbiosis International Deemed University on August 04, 2026, for Deep Learning Based System For Automated Classification Of Leukemic Lymphoblast Subtypes.

Inventors include Dr. Gaurav Vishnu Londhe; Arya Bhende; Sukhada Bhoyar; and Arya Jawanjal.

The application for the patent was published on September 25, 2026, under issue no. 39/2026.

Abstract: ABSTRACT DEEP LEARNING BASED SYSTEM FOR AUTOMATED CLASSIFICATION OF LEUKEMIC LYMPHOBLAST SUBTYPES The present invention discloses an automated diagnostic system (100) for classification of leukemic lymphoblast subtypes from microscopic blood smear images. The system (100) comprises an image acquisition module (110), an image preprocessing module (120) employing LAB colour space conversion and K-means clustering based cell segmentation (126) along with resizing (122), normalization (124), and noise removal (128), a deep learning feature extraction module (130), and a multi-model classification module (140) incorporating MobileNetV2 (142), ResNet50 (144), DenseNet121 (146), and EfficientNetB0 (148) architectures adapted through transfer learning (150). The system classifies blood smear images into benign hematogones and malignant B-cell ALL subtypes including Early Pre-B, Pre-B, and Pro-B. An output prediction module (160), training module (170), and evaluation module (180) facilitate result generation and model optimization. The EfficientNetB0 network achieves highest classification accuracy exceeding 98 percent. [

Disclaimer: Curated by HT Syndication.