From Pixels to Prognosis: The AI Twist in the Tale of Breast Cancer

• Yashna

  GD Goenka University

  Author

• Suyash Srivastava

  GD Goenka University

  Author

AI (Artificial intelligence) and ML (machine learning ) are significantly advancing the area of cancer imaging, with enhanced diagnostic accuracy, earlier detection of tumors, and improved clinical workflow. This paper consolidates evidence from seminal studies to provide a vivid description of how AI algorithms is being implemented, especially in breast cancer. Let’s explore prominent AI methods like convolutional neural networks, support vector machines, and radiomics and demonstrate how these technologies improve diagnostic accuracy and provide real-world benefits in medical practice. But we have to be enlightened of the existing challenges in the form of data variability, clinical validation, ethics, as well as the hurdles of regulatory approvals. These need to be addressed to enable AI products to be transparent, explainable, and equitable to all patients. Ahead, incorporating heterogeneous sources of data and implementing federated learning have tantalizing potential to deliver personalized cancer therapy. The paper emphasizes the incredible potential for AI to improve established imaging modalities and, over time, augment patient outcomes and underscores the requirement for comprehensive verification and vigilant utilization in practical, real-world use.

[1] Aerts, H. J. W. L., et al. (2014). Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics (Aerts et al., 2014) approach. Nature Communications, 5, 4006.

[2] Arik, S. O., & Pfister, T. (2019). TabNet: Attentive interpretable tabular learning. arXiv preprint arXiv:1910.04858.

[3] Buda, M., Saha, A., & Mazurowski, M. A. (2019). Association of genomic subtypes of lower-grade gliomas with shape features automatically extracted by a deep learning algorithm. Scientific Reports, 9(1), 1-10.

[4] Cheng, J., et al. (2016). Enhanced performance of brain tumor classification via tumor region augmentation and partition. Neurocomputing, 219, 10–18.

[5] Cruz, J. A., & Wishart, D. S. (2006). Applications of machine learning in cancer prediction and prognosis. Current Medicinal Chemistry, 13(19), 2345–2357.

[6] Esteva, A., et al. (2017). Dermatologist-level classification of skin cancer with deep neural networks. Nature, 542(7639), 115–118.

[7] Ha, R., et al. (2021). Breast cancer molecular subtype classifier that incorporates MRI (Yasaka et al., 2018) features. European Journal of Radiology, 134, 109408.

[8] Holzinger, A., et al. (2019). Causability and explainability of artificial intelligence in medicine. WIREs Data Mining and Knowledge Discovery, 9(4), e1312.

[9] Huang, Y., et al. (2020). Radiomics signature: a potential biomarker for the prediction of disease-free survival in early-stage (I or II) non–small cell lung cancer. Radiology, 281(3), 947–957.

[10] Jiang, Y., et al. (2021). Radiomics signature of breast cancer: a biomarker for the preoperative prediction of lymph node metastasis. npj Breast Cancer, 7(1), 1–9.

[11] Kermany, D. S., et al. (2018). Identifying medical diagnoses and treatable diseases by image-based deep learning (Kermany et al., 2018). Cell, 172(5), 1122–1131.

[12] Lee, R. S., et al. (2017). Curated Breast Imaging Subset of DDSM. Radiology: Artificial Intelligence, 1(3), e180001.

[13] Litjens, G., et al. (2017). A survey on deep learning in medical image analysis. Medical Image Analysis, 42, 60–88.

[14] Lindholm, P., et al. (2022). Use of AI in mammography screening: a population-based observational study. European Radiology, 32(2), 1321–1329.

[15] Mobadersany, P., et al. (2018). Predicting cancer outcomes from histology and genomics using convolutional networks. PNAS, 115(13), E2970–E2979.

[16] Nie, D., et al. (2017). Medical image synthesis with deep convolutional adversarial networks. IEEE Transactions on Biomedical Engineering, 65(12), 2720–2730.

[17] Shen, L., et al. (2019). Deep learning to improve breast cancer detection on screening mammography. Nature Communications, 10(1), 5641.

[18] Topol, E. J. (2019). High-performance medicine: the convergence of human and artificial intelligence. Nature Medicine, 25(1), 44–56.

[19] Wang, X., et al. (2017). ChestX-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases. CVPR.

[20] Yala, A., et al. (2021). Toward robust mammography-based models for breast cancer risk. Science Translational Medicine, 13(578), eaba4373.

[21] Yasaka, K., et al. (2018). Deep learning with convolutional neural network in radiology. Radiology, 286(3), 889–897.

[22] Zhang, L., et al. (2021). Radiomics-based machine learning models for prediction of malignancy risk in BI-RADS 4 breast lesions. Frontiers in Oncology, 11, 652968.

[23] Zhou, S. K., et al. (2021). A review of deep learning in medical imaging: Imaging traits, technology trends, case studies with progress highlights, and future promises. Nature Biomedical Engineering, 5(6), 522–534.