This article is a summary of a physician podcast discussion around AI in healthcare. The contents and conclusions of the following article are solely those of the speaker, unless otherwise cited. The speaker received funding from Covidien LP, a Medtronic company, for this engagement.
For the first time, artificial intelligence (AI) has been integrated seamlessly into colonoscopy. This advancement might help endoscopists improve adenoma detection rates (ADR) and help prevent colorectal cancer (CRC). But what does AI mean in this context, and why is it important now?
When discussing AI, what is being referenced is a wide-ranging field of computer science that consists of machines and software with the ability to perform tasks that typically or previously required human intelligence alone.1
A simple example of AI is facial recognition. When someone unlocks their phone or another device using facial recognition software, an AI algorithm recognizes facial patterns. The algorithm compares it to a set of images that have been provided to complete the designated task. This happens with a reasonable degree of certainty.
As groundbreaking as all of this can seem, modern AI has its roots in technology that was developed more than half a century ago. In fact, it dates all the way back to the 1950s, when early work began on building neural networks, or “thinking machines” that could recognize patterns and correlations in raw data sets.2
Toward the end of the 20th century, AI evolved to include what is referred to as machine learning, or AI systems that can learn and improve from experience. This happens without being explicitly programmed by a human. AI is the ability to mimic human intelligence using data, and machine learning is the specific ability of AI to mimic human learning.2
More recently, AI has gone much deeper than just analyzing data sets. Since so many of the machines used are connected to the internet, advanced neural networks and machine learning algorithms can cooperate. These systems can use data from many different sources like text, images, and our voices to put massive computing power at our fingertips.2
It is easy to see how AI can help in a healthcare context — any system that can examine data, pull from health records, compile notes, or translate speech. All of these areas have huge potential to improve efficiency in healthcare.
AI-powered solutions now have an opportunity to play a significant role as it relates to colonoscopy and the prevention of CRC. Data shows 1 in 23 adults will be diagnosed with CRC in their lifetime.3 It is currently the second-deadliest cancer in the U.S.3 Adenoma miss rates during colonoscopy can be as high as 26%,4 and this can contribute to false-negative results. The lesions missed during colonoscopy are responsible for 50-60% of interval cancers.4 Furthermore, sessile serrated polyps, which have been associated with interval cancer, can be difficult to identify.5 As with many cancers, early detection is crucial for increasing positive healthcare outcomes. It has been reported that 90% of CRC patients can beat it, if it is caught early.6
While colonoscopy is considered the gold standard for CRC screening, it is not perfect, and it can be difficult. There are precautions physicians can take to improve the quality of colonoscopies, like limiting fatigue, mitigating bias, and improving bowel preparation. AI is showing now that it can be a powerful tool to help increase the quality of colonoscopies and increase detection rates.
Given the clear value of early CRC detection and maximizing the effectiveness of colonoscopy as a screening tool, AI represents the next step in the journey to improve patient outcomes.
What is Artificial Intelligence? How Does AI Work? (2019). Built In. https://builtin.com/artificial-intelligence.
Artificial Intelligence – What it is and why it matters. (2017). SAS. https://www.sas.com/en_us/insights/analytics/what-is-artificial-intelligence.html.
Facts and Statistics about Colorectal Cancer. (2020). Colorectal Cancer Alliance. https://www.ccalliance.org/colorectal-cancer-information/facts-and-statistics.
Zhao, S., Wang, S., Pan, P., Xia, T., Chang, X., Yang, X., et al. (2019). Magnitude, Risk Factors, and Factors Associated With Adenoma Miss Rate of Tandem Colonoscopy: A Systematic Review and Meta-analysis. Gastroenterology, 156(6), 1661–1674.e11.
Lee, Y. M., & Huh, K. C. (2017). Clinical and Biological Features of Interval Colorectal Cancer. Clinical endoscopy, 50(3), 254–260.
Is colorectal cancer preventable with screening? (2019, March 27). Colorectal Cancer Alliance. https://www.ccalliance.org/blog/prevention/is-colorectal-cancer-preventable-with-screening.
Disclaimer: All content from healthcare professionals is their individual conclusions, unless otherwise cited. All speaker or author engagement for content is noted to acknowledge funding from Covidien LP, a Medtronic company, for any consulting engagement.