AI Use in Surgeries for Prostate Cancer May Help Standardize Outcomes

Artificial intelligence (AI) may have the potential to better standardize surgery outcomes in patients undergoing prostatectomies and train doctors to more effectively perform them, according to Ketan K. Badani, MD

CancerNetwork® spoke with Badani, vice chairman of Urology and Robotics and director of the Comprehensive Kidney Cancer Center and Reconstructive Urology at Mount Sinai Health System, about how AI may be integrated into patient care.

Badani began by suggesting that the opportunities for AI use were of particularly high relevance on the surgical side of patient care. He first explained that AI can be utilized to standardize surgical quality given its ability to analyze and learn from observing video feed from a surgery.

To illustrate this, he compared regional discrepancies in prostatectomy outcomes, whereby someone from a more isolated or rural area may not have access to a surgeon with the same level of skill or experience as one from a dense, urban area. He expressed that AI would mitigate these discrepancies by studying the best practices of experienced surgeons and programming surgical robots to perform at that level, regardless of location.

Looking ahead, Badani expressed that the implementation of this technology may be feasible. Additionally, he further expanded upon the utility of AI in surgical practice by suggesting that it may help train surgeons to use the best practices observed by AI, effectively standardizing training. Badani then explained that AI-informed augmented reality could help train surgeons to perform a particular operation and adopt the most effective surgical practices.

Transcript:

[For] AI and its management of patient care, there is a laundry list of potential opportunities here. I do not think we will be able to touch on all of that. There are a couple important things on the surgical side of [AI] that are highly relevant. Number 1 is standardizing the quality of an operation. We now have AI systems in place that can look at video and can tell what is happening based on the video. Are you suturing an anastomosis for a prostatectomy? Are you doing a nerve sparing? Are you doing it too close to the prostate? Based on just video feed alone, AI has the technology to recognize what is happening.

Let us say you are a prostate cancer surgeon who has done a million robotic prostatectomies. You are the best of the best, and no one is better than you. Or you have prostate cancer surgeon B, and he or she has done some and is pretty good, but there is a difference in outcomes––complication rates, something bad happening, potency rates, urinary incontinence rates. If you are in Omaha, how do you get to someone who has the million-prostatectomy experience and get that operation [like patients who] live in the middle of New York City with 5 [surgeons] that have done it a million times and have access to that? The answer is AI.

AI will standardize the quality of how [a procedure is] done. It will study the moves of the surgeons that have done millions of these—the best of the best—and translate it to everyone else. Now you have democratized it...Now, I know [I am] dreaming [about] the future and being a bit too optimistic about it, but I think it is real. Everyone will get the same operation in the end because AI will show you what you need to do it [the optimal] way.

Now, the quality can be in Omaha, it can be in New York City, [or] it can be in the middle of Alaska. It does not matter. As long as you have the physical technology, in this case, a surgical robot, you should be able to experience the same quality outcome you would [elsewhere], no matter where you are. [With] AI, we will be able to [accomplish that] or at least get us close. That also gets us into training.

You only have so many doctors that do it [with such frequency]; that can only take care of so many patients. [There is a] need to train other doctors to be up to that level. Without doing 1000 of them, how are you going to get physicians to that level? We get back to augmented reality, AI, and machine learning helping surgeons [perform] the same operation. Then, it [becomes] available to more patients across the world.