Authors:
S. Rubin Bose, J. Angelin Jeba, V. Karrthik Kishore, Gnaneswari Gnanaguru, T. Shynu
Addresses:
Department of Electronics and Communication Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India, rubinbos@srmist.edu.in. Department of Electronics and Communication Engineering, S.A. Engineering College, Chennai, Tamil Nadu, India, angelinjeba@saec.ac.in. Department of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India, vk5619@srmist.edu.in. Department of Computer Applications, CMR Institute of Technology, Bengaluru, Karnataka, India, gnaneswari@yahoo.com. Department of Electronics and Communication Engineering, Dhaanish Ahmed College of Engineering, Chennai, Tamil Nadu, India, shynu469@gmail.com.
This paper presents an advanced method for cancer classification using the YOLOv8 architecture, emphasizing its application in detecting Acute Lymphoblastic Leukemia (ALL). The study explores the YOLOv8 framework, which integrates Darknet-53 as the backbone for feature extraction, an FPN neck for multi-scale object detection, and a detection head for bounding box prediction and class probability estimation. The proposed system demonstrates significant accuracy in classifying and localizing cancerous cells across eight cancer types, focusing on Acute Lymphoblastic Leukemia (ALL). By leveraging the YOLOv8 architecture, the system achieves high precision, recall, and F1 scores, indicating its potential for improving automated cancer diagnosis. The paper also delves into the mathematical modeling, loss functions, and evaluation metrics employed to optimize the model’s performance. Future work will enhance the system’s efficiency and extend its application to other types of cancers, thereby contributing to the broader medical imaging and diagnostics field.
Keywords: Cancer Classification; Acute Lymphoblastic Leukemia; Machine Learning; Medical Imaging; Object Detection; Convolutional Neural Networks; Detection and Diagnosis.
Received on: 05/01/2024, Revised on: 02/03/2024, Accepted on: 29/04/2024, Published on: 03/06/2024
AVE Trends in Intelligent Health Letters, 2024 Vol. 1 No. 2, Pages: 110-124