Nathanael R. Fillmore
Publications / 2023
paper
Nature Medicine2023doi:10.1038/s41591-023-02332-5

A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories.

Davide Placido†, Bo Yuan†, Jessica X. Hjaltelin†, Chunlei Zheng†, Amalie D. Haue, Piotr J. Chmura, Chen Yuan, Jihye Kim, Renato Umeton, Gregory Antell, Alexander Chowdhury, Alexandra Franz, Lauren Brais, Elizabeth Andrews, Debora S. Marks, Aviv Regev, Siamack Ayandeh, Mary T. Brophy, Nhan V. Do, Peter Kraft, Brian M. Wolpin, Michael H. Rosenthal, Nathanael R. Fillmore, Søren Brunak‡, Chris Sander‡

† Equal first authorship · ‡ Equal senior authorship

Summary

A deep learning algorithm can predict the risk of pancreatic cancer by analyzing patterns of previous medical codes in large patient datasets. This tool enables clinicians to identify high-risk individuals earlier, allowing for targeted surveillance and the potential for improved survival rates through prompt diagnosis.

Key result

The best-performing model predicted pancreatic cancer occurrence within 36 months with an area under the receiver operating characteristic curve of 0.88 and an estimated relative risk of 59 for the 1,000 highest-risk patients older than 50 years.

Abstract

Pancreatic cancer is an aggressive disease that typically presents late with poor outcomes, indicating a pronounced need for early detection. In this study, we applied artificial intelligence methods to clinical data from 6 million patients (24,000 pancreatic cancer cases) in Denmark (Danish National Patient Registry (DNPR)) and from 3 million patients (3,900 cases) in the United States (US Veterans Affairs (US-VA)). We trained machine learning models on the sequence of disease codes in clinical histories and tested prediction of cancer occurrence within incremental time windows (CancerRiskNet). For cancer occurrence within 36 months, the performance of the best DNPR model has area under the receiver operating characteristic (AUROC) curve = 0.88 and decreases to AUROC (3m) = 0.83 when disease events within 3 months before cancer diagnosis are excluded from training, with an estimated relative risk of 59 for 1,000 highest-risk patients older than age 50 years. Cross-application of the Danish model to US-VA data had lower performance (AUROC = 0.71), and retraining was needed to improve performance (AUROC = 0.78, AUROC (3m) = 0.76). These results improve the ability to design realistic surveillance programs for patients at elevated risk, potentially benefiting lifespan and quality of life by early detection of this aggressive cancer.

Key figure

Key figure from A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories.