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

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories. / Placido, Davide; Yuan, Bo; Hjaltelin, Jessica X; Zheng, Chunlei; Haue, Amalie D; Chmura, Piotr J; Yuan, Chen; Kim, Jihye; Umeton, Renato; Antell, Gregory; Chowdhury, Alexander; Franz, Alexandra; Brais, Lauren; Andrews, Elizabeth; Marks, Debora S; Regev, Aviv; Ayandeh, Siamack; Brophy, Mary T; Do, Nhan V; Kraft, Peter; Wolpin, Brian M; Rosenthal, Michael H; Fillmore, Nathanael R; Brunak, Søren; Sander, Chris.

In: Nature Medicine, Vol. 29, 2023, p. 1113-1122.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Placido, D, Yuan, B, Hjaltelin, JX, Zheng, C, Haue, AD, Chmura, PJ, Yuan, C, Kim, J, Umeton, R, Antell, G, Chowdhury, A, Franz, A, Brais, L, Andrews, E, Marks, DS, Regev, A, Ayandeh, S, Brophy, MT, Do, NV, Kraft, P, Wolpin, BM, Rosenthal, MH, Fillmore, NR, Brunak, S & Sander, C 2023, 'A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories', Nature Medicine, vol. 29, pp. 1113-1122. https://doi.org/10.1038/s41591-023-02332-5

APA

Placido, D., Yuan, B., Hjaltelin, J. X., Zheng, C., Haue, A. D., Chmura, P. J., Yuan, C., Kim, J., Umeton, R., Antell, G., Chowdhury, A., Franz, A., Brais, L., Andrews, E., Marks, D. S., Regev, A., Ayandeh, S., Brophy, M. T., Do, N. V., ... Sander, C. (2023). A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories. Nature Medicine, 29, 1113-1122. https://doi.org/10.1038/s41591-023-02332-5

Vancouver

Placido D, Yuan B, Hjaltelin JX, Zheng C, Haue AD, Chmura PJ et al. A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories. Nature Medicine. 2023;29:1113-1122. https://doi.org/10.1038/s41591-023-02332-5

Author

Placido, Davide ; Yuan, Bo ; Hjaltelin, Jessica X ; Zheng, Chunlei ; Haue, Amalie D ; Chmura, Piotr J ; Yuan, Chen ; Kim, Jihye ; Umeton, Renato ; Antell, Gregory ; Chowdhury, Alexander ; Franz, Alexandra ; Brais, Lauren ; Andrews, Elizabeth ; Marks, Debora S ; Regev, Aviv ; Ayandeh, Siamack ; Brophy, Mary T ; Do, Nhan V ; Kraft, Peter ; Wolpin, Brian M ; Rosenthal, Michael H ; Fillmore, Nathanael R ; Brunak, Søren ; Sander, Chris. / A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories. In: Nature Medicine. 2023 ; Vol. 29. pp. 1113-1122.

Bibtex

@article{d6d0b64773cb4d7a83e6da8040cb2623,
title = "A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories",
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.",
author = "Davide Placido and Bo Yuan and Hjaltelin, {Jessica X} and Chunlei Zheng and Haue, {Amalie D} and Chmura, {Piotr J} and Chen Yuan and Jihye Kim and Renato Umeton and Gregory Antell and Alexander Chowdhury and Alexandra Franz and Lauren Brais and Elizabeth Andrews and Marks, {Debora S} and Aviv Regev and Siamack Ayandeh and Brophy, {Mary T} and Do, {Nhan V} and Peter Kraft and Wolpin, {Brian M} and Rosenthal, {Michael H} and Fillmore, {Nathanael R} and S{\o}ren Brunak and Chris Sander",
note = "{\textcopyright} 2023. The Author(s).",
year = "2023",
doi = "10.1038/s41591-023-02332-5",
language = "English",
volume = "29",
pages = "1113--1122",
journal = "Nature Medicine",
issn = "1078-8956",
publisher = "nature publishing group",

}

RIS

TY - JOUR

T1 - A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories

AU - Placido, Davide

AU - Yuan, Bo

AU - Hjaltelin, Jessica X

AU - Zheng, Chunlei

AU - Haue, Amalie D

AU - Chmura, Piotr J

AU - Yuan, Chen

AU - Kim, Jihye

AU - Umeton, Renato

AU - Antell, Gregory

AU - Chowdhury, Alexander

AU - Franz, Alexandra

AU - Brais, Lauren

AU - Andrews, Elizabeth

AU - Marks, Debora S

AU - Regev, Aviv

AU - Ayandeh, Siamack

AU - Brophy, Mary T

AU - Do, Nhan V

AU - Kraft, Peter

AU - Wolpin, Brian M

AU - Rosenthal, Michael H

AU - Fillmore, Nathanael R

AU - Brunak, Søren

AU - Sander, Chris

N1 - © 2023. The Author(s).

PY - 2023

Y1 - 2023

N2 - 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.

AB - 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.

U2 - 10.1038/s41591-023-02332-5

DO - 10.1038/s41591-023-02332-5

M3 - Journal article

C2 - 37156936

VL - 29

SP - 1113

EP - 1122

JO - Nature Medicine

JF - Nature Medicine

SN - 1078-8956

ER -

ID: 346457476