Part 3 of an ongoing series
David Letsa, MD
Integrating artificial intelligence (AI) and deep learning into the diagnostic process has begun to transform the landscape of cancer care. With their capability to sift through immense quantities of data and to detect intricate patterns within that data, these technologies are enhancing the accuracy and speed of cancer diagnoses.1
Automation of Medical Image Analysis
One of the most promising applications of AI in cancer diagnosis lies in the analysis of medical imaging. Deep learning algorithms, a specific subset of machine learning, are particularly effective at interpreting complex visual data in images such as CT, MRI, and mammograms. Programmers can train AI algorithms to recognize the characteristic features of malignant and benign lesions, often with a precision that surpasses human experts.
For example, the detection and removal of precancerous polyps via colonoscopy is the gold standard for colon cancer prevention. However, machine-learning algorithms can detect polyps in clinical colonoscopies in real-time and with high sensitivity and specificity. A deep-learning algorithm was developed using data from 1,290 patients and validated on various data sets, showing high sensitivity and specificity in detecting polyps with impressive performance metrics. It demonstrated processing at least 25 frames per second with low latency in real-time video analysis. The software has the potential to aid endoscopists during colonoscopies and assess differences in polyp and adenoma detection performance among endoscopists.2
Pattern Recognition in Patient Data
AI systems can also process and analyze other forms of patient data, such as genetic information, blood tests, and electronic health records. By identifying patterns that may indicate the presence of cancer, these systems can help clinicians in the early detection and diagnosis of cancer, which is crucial for the successful treatment of the disease. For instance, the Vision Transformer architecture known as ViT-Patch is effective for both malignant detection and tumor localization.3
Improving Diagnostic Accuracy
The substantial advantage of AI over traditional methods is its ability to consistently and quickly evaluate these patterns, leading to more accurate diagnoses. For instance, in January 2020, an AI system based on a Google DeepMind algorithm outperformed human specialists in breast cancer detection based on imaging data.4
Enhanced Efficiency for Advanced Practice Providers (APPs):
AI tools can take on time-consuming tasks for advanced practice providers, freeing clinicians to prioritize patient care requiring human expertise, such as empathetic communication and critical decision-making, which can reduce the risk of severe burnout among health care professionals.5 Moreover, APPs can leverage AI to interpret patient data more quickly, which can be especially valuable in settings where the volume of patients is high or resources are limited.6
As AI becomes more integrated into clinical settings, health care providers must remain informed about these advancements and how they apply to everyday practice. The ultimate goal is to work alongside these technologies to improve the standard of care for cancer patients, achieving earlier detection, more precise diagnoses, and, consequently, better clinical outcomes.
In summary, the transformative potential of deep learning and AI in cancer care is vast. It not only streamlines the process of diagnosing cancer but also opens the door to personalized medicine, where treatments are customized based on AI-driven insights into the individual characteristics of each patient’s tumor. As we continue to harness the power of AI and deep learning, we can expect ongoing improvements in cancer care, significantly impacting patients’ lives.
David Letsa, MD, is a medical doctor and medical writer. Dr. Letsa completed his medical degree at Semmelweis University and held certificates in global health innovation and entrepreneurship from the Barcelona Institute for Global Health (ISGlobal) and Health-based Artificial Intelligence Online Training from GE Healthcare. Find David on LinkedIn.
References
- Zhang, B., Shi, H., & Wang, H. (2023). Machine Learning and AI in Cancer Prognosis, Prediction, and Treatment Selection: A Critical Approach. Journal of multidisciplinary healthcare, 16, 1779–1791. https://doi.org/10.2147/JMDH.S410301
- Wang, P., Xiao, X., Glissen Brown, J. R., Berzin, T. M., Tu, M., Xiong, F., Hu, X., Liu, P., Song, Y., Zhang, D., Yang, X., Li, L., He, J., Yi, X., Liu, J., & Liu, X. (2018). Development and validation of a deep-learning algorithm for the detection of polyps during colonoscopy. Nature biomedical engineering, 2(10), 741–748. https://doi.org/10.1038/s41551-018-0301-3
- Feng, H., Yang, B., Wang, J., Liu, M., Yin, L., Zheng, W., Yin, Z., Liu, C. (2023). Identifying malignant breast ultrasound images using ViT-patch. Appl sci. 13(6), 3489.https://doi.org/10.3390/app13063489
- McKinney, S. M., Sieniek, M., Godbole, V., Godwin, J., Antropova, N., Ashrafian, H., Back, T., Chesus, M., Corrado, G. S., Darzi, A., Etemadi, M., Garcia-Vicente, F., Gilbert, F. J., Halling-Brown, M., Hassabis, D., Jansen, S., Karthikesalingam, A., Kelly, C. J., King, D., Ledsam, J. R., … Shetty, S. (2020). International evaluation of an AI system for breast cancer screening. Nature, 577(7788), 89–94. https://doi.org/10.1038/s41586-019-1799-6
- Orozco, J. M., Furman, J., McAndrews, K. K., Keenan, M. M., Roman, C., Guthrie, J., Lloyd, C. J., & Wilson, A. B. (2019). Assessing Burnout Among Advanced Practice Providers (APPs) Compared with APP Trainees. Medical science educator, 29(4), 1023–1031. https://doi.org/10.1007/s40670-019-00799-x
- Abbasgholizadeh Rahimi, S., Cwintal, M., Huang, Y., Ghadiri, P., Grad, R., Poenaru, D., Gore, G., Zomahoun, H. T. V., Légaré, F., & Pluye, P. (2022). Application of Artificial Intelligence in Shared Decision Making: Scoping Review. JMIR medical informatics, 10(8), e36199. https://doi.org/10.2196/36199
