Multiple Cancer Types and Subtype Classification using Deep Learning
Rutvi Sindha, Vidhi Vansiya, Dhara Parikh
12Student, Information Technology Department, Krishna School of Emerging Technology & Applied Research, KPGU University, Varnama, Vadodara, Gujarat, India
3Assistant Professor, Department of Information Technology and Engineering, Krishna School of Emerging Technology & Applied Research, KPGU University, Varnama, Vadodara, Gujarat, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Cancer classification is crucial for effective treatment planning, especially given the diversity of cancer types and subtypes that pose unique diagnostic challenges. This study presents a deep learning-based system for multi-cancer classification, utilizing a comprehensive dataset comprising over 130,000 images across 8 main cancer types and 26 subtypes. Our approach leverages Convolutional Neural Networks (CNNs) to automatically extract intricate features from cancer images, enhancing classification accuracy across various cancer types, including Acute Lymphoblastic Leukemia, Brain Cancer, Breast Cancer, Cervical Cancer, Kidney Cancer, Lung and Colon Cancer, Lymphoma, and Oral Cancer. To improve the model’s robustness, data augmentation techniques—such as rotation, shifting, brightness adjustment, and resizing—were applied. The model was trained, validated, and tested using a balanced dataset, and achieved notable accuracy in both main cancer type and subtype classification. Experimental results demonstrate the potential of our approach to facilitate reliable and automated cancer detection, supporting clinical diagnostic processes and potentially aiding in earlier detection and treatment of multiple cancer types. This research contributes a novel deep learning framework for multi-class cancer identification and highlights its application for large-scale cancer image datasets.
Key Words: Multi-Cancer Detection, Deep Learning, Convolutional Neural Networks (CNN), Image Data Augmentation
