AI can predict pancreatic cancer 3 years before diagnosis

Credit: Unsplash+

An AI tool has been developed that can predict those most at risk of developing pancreatic cancer.

The tool can identify patients at risk up to three years before diagnosis, by analyzing their medical records.

The research was led by scientists at Harvard Medical School and the University of Copenhagen, along with VA Boston Healthcare System, Dana-Farber Cancer Institute, and the Harvard T.H. Chan School of Public Health.

Pancreatic cancer is notoriously difficult to diagnose, meaning that it is often caught in advanced stages when treatments are less effective.

While some patients with a genetic predisposition or family history of pancreatic cancer are targeted for screening, the disease can be missed in other patients who fall outside of these categories.

An AI-based population screening tool for pancreatic cancer could be used to identify those at the highest risk of developing the disease, allowing for earlier detection and treatment.

How the AI tool works

The AI algorithm was trained on two datasets containing 9 million patient records from Denmark and the United States.

The model was asked to look for telltale signs of pancreatic cancer based on the data contained in the records.

By analyzing combinations of disease codes and the timing of their occurrence, the model was able to predict which patients were likely to develop pancreatic cancer in the future.

Testing the AI tool

The researchers tested different versions of the AI models on their ability to detect people at elevated risk for disease development within different time frames – 6 months, one year, two years, and three years.

Each version of the AI algorithm was substantially more accurate at predicting who would develop pancreatic cancer than current population-wide estimates of disease incidence.

The tool could be used on any and all patients for whom health records and medical history are available, not just those with known family history or genetic predisposition for the disease.

This is especially important because many patients at high risk may not even be aware of their genetic predisposition or family history.

Importance of the tool

An AI tool that identifies those at the highest risk for pancreatic cancer would ensure that clinicians test the right population while sparing others unnecessary testing and additional procedures, the researchers said.

Only 12% of pancreatic cancer cases are diagnosed in the early stages when the cancer is most treatable, meaning that earlier detection and treatment is critical for improving survival rates.

By identifying those at highest risk, AI-based population screening could lead to earlier detection and intervention, improving patient outcomes and prolonging life.

Implications for future research

The researchers suggest that AI-based screening is an opportunity to alter the trajectory of pancreatic cancer.

Their work shows that an AI tool that can zero in on those at the highest risk for pancreatic cancer who stand to benefit most from further tests could go a long way toward improving clinical decision-making.

If you care about cancer, please read studies about a safer, more effective cancer therapy, and vitamin D supplements could strongly reduce cancer death.

For more information about health, please see recent studies about how drinking milk affects the risks of heart disease and cancer and results showing vitamin D supplements could strongly reduce cancer death.

The study was published in Nature Medicine.

Copyright © 2023 Knowridge Science Report. All rights reserved.