Dynamic Thermal Imaging and Frame-wise Deep Learning Analysis for Breast Cancer Diagnosis: A Comparative Study with Mammography

Authors

  • Sandhya C., Suresha D

Keywords:

Dynamic Thermal Imaging, Deep Learning, Breast Cancer Detection, Mammography, CNN-LSTM, Temporal Analysis, DMR-IR, CBIS-DDSM, Medical Image Classification, AI-Based Diagnostics

Abstract

Timely identification of breast cancer plays a vital role in enhancing survival rates. While mammography is a widely adopted imaging technique, it often presents limitations related to accessibility, sensitivity, and radiation exposure—particularly in low-resource environments. In this study, we investigate the application of dynamic thermal imaging (DMR-IR) as a radiation-free, non-invasive alternative, utilizing a frame-wise deep learning strategy.

A comparative evaluation is conducted using two datasets: the DMR-IR thermal image sequences and the established CBIS-DDSM mammographic dataset. The proposed methodology integrates convolutional neural networks (CNN), transfer learning models including ResNet50 and EfficientNetB0, along with an optional CNN-LSTM architecture to model both spatial and temporal dynamics present in thermal image frames. Evaluation metrics—namely accuracy, precision, recall, F1-score, and AUC—demonstrate that the DMR-IR dataset yields superior classification performance across all tested models when compared to mammography.

The findings underscore the effectiveness of temporal thermal patterns in identifying malignancies and present a promising, scalable solution for early breast cancer screening, particularly in resource-constrained clinical settings. This work establishes the groundwork for advancing AI-driven diagnostic solutions based on physiological imaging data.

DOI: https://doi.org/10.17762/ijisae.v12i21s.7717

Downloads

Download data is not yet available.

References

Carriero, A., Groenhoff, L., Vologina, E., Basile, P., & Albera, M. (2024). Deep learning in breast cancer imaging: State of the art and recent advancements in early 2024. Diagnostics, 14(8), 848..

Li, H., Zeng, Q., & Zhang, H. (2021). Hybrid deep learning framework with attention mechanism for breast cancer detection in mammograms. Journal of Imaging, 9(2), 36.

Mirasbekov, Y., Aidossov, N., Mashekova, A., Zarikas, V., Zhao, Y., Ng, E. Y. K., & Midlenko, A. (2024). Fully interpretable deep learning model using IR thermal images for possible breast cancer cases. Biomimetics, 9, 609.

Mohamed, E. A., Rashed, E. A., Gaber, T., & Karam, O. (2022). Deep learning model for fully automated breast cancer detection system from thermograms. PLoS ONE, 17, e0262349.

Roslidar, R., et al. (2020). A review on recent progress in thermal imaging and deep learning approaches for breast cancer detection. IEEE Access, 8, 116176–116194.

Sadaf, M. I., Shah, M. A., Abbas, N., & Khan, A. (2023). Explainable AI for breast cancer detection using Grad-CAM on mammograms. Computers in Biology and Medicine, 158, 106783.

Shen, L., Margolies, L. R., Rothstein, J. H., Fluder, E., McBride, R., & Sieh, W. (2017). Deep learning to improve breast cancer early detection on screening mammography. Scientific Reports, 7, 44865.

Tsietso, D., Yahya, A., Samikannu, R., Tariq, M. U., Babar, M., Qureshi, B., & Koubaa, A. (2023). Multi-input deep learning approach for breast cancer screening using thermal infrared imaging and clinical data. IEEE Access, 11, 52101–52116.

Tsochatzidis, L., Costaridou, L., & Pratikakis, I. (2019). Deep learning for breast cancer diagnosis from mammograms—A comparative study. Journal of Imaging, 5(3), 37.

Gupta, S., Verma, H., & Sahu, P. (2022). Fine-tuning pre-trained CNNs for breast cancer detection: A comparison of ResNet and EfficientNet architectures. Applied Soft Computing, 112, 107845. https://doi.org/10.1016/j.asoc.2021.107845

Khan, M. A., Hussain, M., Alshahrani, S., Ahmed, F., & Naeem, M. (2021). A multi-modal deep learning ensemble approach for breast cancer diagnosis using mammography and thermography. Computer Methods and Programs in Biomedicine, 209, 106320. https://doi.org/10.1016/j.cmpb.2021.106320

Rajpurkar, P., Chen, E., Banerjee, O., & Topol, E. J. (2023). EfficientNetB0-based deep neural network for rapid breast cancer screening. Journal of Digital Imaging, 36, 1237–1248.

Zhang, L., Wei, X., Feng, Y., & Chen, H. (2024). Attention-based feature fusion in multi-modal breast cancer imaging. IEEE Transactions on Neural Networks and Learning Systems, 35(5), 2194–2205.

Downloads

Published

28.08.2024

How to Cite

Sandhya C. (2024). Dynamic Thermal Imaging and Frame-wise Deep Learning Analysis for Breast Cancer Diagnosis: A Comparative Study with Mammography. International Journal of Intelligent Systems and Applications in Engineering, 12(21s), 5065 –. Retrieved from https://ijisae.org/index.php/IJISAE/article/view/7717

Issue

Section

Research Article