Grey Wolf-Optimised Convolutional Extreme Learning Machine for Early-Stage Cancer Detection Through Thermal Imaging
Abstract
Breast cancer remains the most prevalent malignancy among women, rendering early detection crucial for improving survival. Diagnostic methods, including mammography, MRI, and ultrasound, are invasive, expensive, and unsuitable for large-scale screening. Infrared thermography is non-invasive, cost-effective, but relies on robust computational models. We introduce a hybrid deep learning framework that integrates a convolutional neural network (CNN) for automated feature extraction with extreme learning machine (ELM) classifier, optimised using the grey wolf optimisation algorithm. The CNN–ELM model achieved 99.56% accuracy, 99.12% sensitivity and 100% specificity, outperforming conventional and transfer-learning-based models, demonstrating high efficiency, robustness, and real-time breast cancer screening capability.
Fields of Study & Subjects
Machine Learning and ELM
Infrared Thermography in Medicine
Brain Tumor Detection and Classification
Publication details
- Journal / source
- IETE Journal of Research
- Publisher
- Informa UK Limited
- Volume
- 72
- Issue
- 5-6
- Pages
- 1409–1421
- Publication date
- 18 January 2026
- Language
- EN