In a groundbreaking development, researchers at the University of California San Diego have unveiled a new artificial intelligence (AI) tool that can predict the risk of colorectal cancer in patients with ulcerative colitis (UC). This chronic inflammatory bowel disease makes individuals up to four times more likely to develop colorectal cancer compared to the general population. The study, published on February 17 in Clinical Gastroenterology and Hepatology, highlights the potential of AI to significantly enhance patient counseling and decision-making.
Ulcerative colitis patients often develop low-grade dysplasia (LGD), abnormal or precancerous lesions that can serve as early warning signs. However, only a small percentage of UC-LGD cases progress to cancer, complicating the decision-making process for both clinicians and patients. The AI model, integrated with biostatistical risk models, aims to provide clarity by accurately predicting which UC-LGD patients are most at risk.
AI Model: A New Era in Patient Care
The researchers developed a fully automated AI workflow that analyzed the medical records of 55,000 patients within the U.S. Department of Veterans Affairs (VA) healthcare system. This extensive dataset, the largest of its kind in the U.S., allowed the AI to identify UC-LGD patients and assess their individual cancer risk.
“Large language models accurately derived colitis-associated colorectal cancer risk factors — such as how big the low-grade dysplasia lesion is, whether there are multiple lesions, and if the colon is extremely inflamed — from the narrative clinical notes themselves,” explained Dr. Kit Curtius, assistant professor of medicine at UC San Diego and a member of Moores Cancer Center.
AI Workflow and Statistical Risk Model Predictions
- Correctly grouped patients into five risk categories based on four established factors: dysplasia size, lesion resection completeness and visibility, number of dysplastic sites, and severity of inflammation.
- Matched real-world patient outcomes with high accuracy for more than a decade after diagnosis.
- Classified nearly half of the patients into the lowest-risk group, correctly determining that almost 99% will avoid cancer diagnosis within two years.
“A lot of people are low risk — they have small dysplastic lesions — and it’s been hard to know what to confidently tell these people until now,” said Curtius. “With this tool, there may be a potential to increase the surveillance interval so patients who are at this low risk don’t have to come back so often.”
Implications for Clinical Practice
The AI model also revealed that patients with unresectable visible lesions — those that cannot be safely and completely removed through surgery due to size, location, or extent of spreading — are at significantly higher risk than many clinicians typically estimate. This insight could lead to more tailored patient care strategies.
The study suggests that AI models can integrate seamlessly into clinical workflows, offering precise, automated risk assessments to guide clinician and patient decision-making. This includes determining the timing of the next colonoscopy or when to consider surgery, ultimately reducing the burden on care teams.
“Currently, the process of advising people about levels of risk is a somewhat subjective thing, and doctors don’t have enough data to back up what they feel,” said Curtius. “This AI pipeline could read the clinical notes and tell you your risk score, rather than just having a list of risk factors and no real way to turn that into a number during a patient visit.”
Future Directions and Broader Implications
The next steps for this promising technology include validating the AI tool in patient populations outside of the VA system and incorporating emerging risk factors and patient genetic information. “We know that genomics play a big part in driving cancer progression,” Curtius noted.
Additional co-authors on the study include Brian Johnson and Hyrum Eddington at UC San Diego; Samir Gupta and Shailja C. Shah at UC San Diego and VA San Diego Healthcare System; and Misha Kabir at University College London Hospitals NHS Trust. The study received funding from the U.S. Department of Veterans Affairs Biomedical Laboratory Research and Development Service and the National Institutes of Health.
As AI continues to revolutionize healthcare, this study represents a significant step forward in the fight against colorectal cancer. By providing more accurate risk assessments, AI tools like this one could lead to earlier interventions and improved outcomes for patients with ulcerative colitis.
