Advancing Breast Cancer Detection Through Thermal Imaging and Machine Learning: A Comprehensive Review of Techniques and Applications

• Version
• Download 46
• File Size 588.18 KB
• File Count 1
• Create Date 3 July, 2025
• Last Updated 12 July, 2025

Authors : Vishakha Dubey, Dr. Shanti Rathore, Dr. Rahul Gedam

DOI :  10.46335/IJIES.2025.10.7.17

Abstract – Breast thermography proves to be an evolving diagnostic support method that uses infrared imaging for vascular and metabolic change detection in breast cancer evaluations. This paper conducts an extensive analysis of thermographic imaging progress along with AI-based and ML-based diagnostic enhancement strategies. People conducting comparative studies find that thermography demonstrates utility as a support tool while AI-based convolutional neural networks (CNNs) reach a sensitivity level of 97% in specific applications. The study analyzes two main analysis methods which involve segmentation through K-means clustering and classification through kNN algorithms while evaluating results through combination of fuzzy logic with histogram analysis techniques. The combination of Pennes heat equation with genetic algorithms and finite element methods using simulation techniques delivers useful understanding about tumor dimension and metabolism metrics. Thermography shows potential for clinical use because of its ability to work together with conventional diagnosis methods despite its known temperature-sensitivity and imaging difficulty in dense breasts. Additional research needs to prioritize thermography optimization since this non-invasive approach provides a radiation-free method for detecting breast cancer at its earliest stage.