Two academics teamed up to write an elegant review of four areas in which Artificial Intelligence (AI) is making increasing contributions to the world of prostate cancer (PCa) diagnosis and treatment planning. One hails from the East Coast, the other from the West Coast. First, Dr. James B. Yu is a radiation oncologist in Hartford, CT, and former Professor of Radiation Oncology at the Yale School of Medicine. While he specializes in treating prostate cancer and others, he conducts research on new radiation technologies and their adoption. Second, fellow radiation oncologist Dr. Julian Hong at University of California/San Francisco not only applies radiation to treating prostate and other cancers, his research combines clinical knowledge with data science to develop AI-based clinical tools and evaluate their benefits. Together, they are an ideal pair to identify AI’s accelerating breakthrough PCa-related developments in four arenas, particularly in the field of Machine Learning (ML):

Image Analysis and Radiomics

Drawing upon broad work in computer vision, AI can correctly identify image features in medical scans such as MRI that indicated clinically significant PCa (CSPCa). Though the authors acknowledge the ML is not yet ready to replace human radiological readers, “they may aid less experienced radiologists in distinguishing between cancerous and noncancerous lesions in prostate MRI scans.”^[i]

Predictions of Health Outcomes

ML involves algorithms that predict outcomes such as PCa risk level, spread to lymph nodes, response to treatment and mortality risk. The authors note current challenges to integrating them in day-to-day practice, particularly use of electronic medical record systems and regulatory restrictions, but they anticipate that such algorithms will ultimately replace existing nomograms.

Characterization of PCa Based on Digitized Tissue Slide Images

ML that is trained on specially labeled slides of prostate tissue (histopathology from biopsy or surgery specimens) is able to evaluate image features that correlate with Gleason score. There are also models that use such features to predict treatment outcomes, e.g., “predictive of androgen deprivation therapy in combination with radiotherapy vs radiotherapy alone.”^[ii] The authors refer to this as possibly where AI is making the biggest impact.

Tumor Definition and Treatment Planning for Radiation Therapy

The authors state, “A rapid area of AI expansion is in aiding radiation oncologists in the automated definition of normal tissue and tumor definition.” Based on imaging, an AI program define the tumor and distinguish it from neighboring organs potentially exposed to radiation scatter. Normally, such “contouring” is time consuming, AI improves efficiency and accuracy, and also assists in treatment planning by calculating the target volume and radiation dose to be delivered.

Clinical prostate cancer care is not the only medical field benefiting from AI, as Yu & Hong rightly point out: “The innovations seen in the application of AI to prostate cancer care mirror those happening throughout health care and information technology.” As healthcare services in all fields move forward, AI waves a proud banner leading their progress.

NOTE: This content is solely for purposes of information and does not substitute for diagnostic or medical advice. Talk to your doctor if you are experiencing pelvic pain, or have any other health concerns or questions of a personal medical nature.

References

[i] Yu JB, Hong JC. “AI Use in Prostate Cancer: Potential Improvements in Treatments and Patient Care.” Cancernetwork Oncology, 2024 May 17; 38(5):208-209.
[ii] Ibid.