Skip to main content
2025 ASCRS Annual Scientific Meeting Main banner
Final Program


Visit ASCRS U
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Presentation Icons
CME
CME
Invitation/Ticketed Event
Invitation/Ticketed Event
Invited Lectureship
Invited Lectureship
Award Recipient
Award Recipient
Self-Assessment
Self-Assessment
Social
Social
Industry
Industry
Back

Session: ePoster: Colon and Colorectal Cancer 6 (Sun225)

EP504 - Improving risk stratification for colorectal cancer recurrence after curative surgery with advanced machine learning model

Sunday, May 11, 2025
2:35 PM – 2:40 PM PT
Location: Monitor 6, Sails Pavilion, Upper Level

    ePoster Presenter(s)

  • Jesung Park, MD, PhD

    Fellow
    Seoul National University Hospital
    Seoul, Seoul-t'ukpyolsi, Republic of Korea

    All of the relevant financial relationships listed below have been mitigated.
    This individual has no financial relationships with ineligible companies.

    • Author(s)

  • KJ

    Ki-Hyun Jeon, M.D.

    Professor
    Seoul National University Bundang Hospital
    Seoul, Seoul-t'ukpyolsi, Republic of Korea

    Disclosure information not submitted.

  • JC

    Jin Sun Choi, M.D.

    Professor
    SEOUL METROPOLITAN GOVERNMENT SEOUL NATIONAL UNIVERSITY BORAMAE MEDICAL CENTER
    Seoul, Seoul-t'ukpyolsi, Republic of Korea

    Disclosure information not submitted.

  • MK

    Min Jung Kim, MD, PhD

    Visiting Investigator
    Memorial Sloan Kettering
    New York, New York, United States

    Disclosure information not submitted.

  • SR

    Seung-Bum Ryoo, M.D., Ph.D.

    Professor
    Colorectal Surgery division, Surgery department, Seoul national university hospital
    Seoul, Seoul-t'ukpyolsi, Republic of Korea

    All of the relevant financial relationships listed below have been mitigated.
    This individual has no financial relationships with ineligible companies.

  • SJ

    Seung-Yong Jeong, n/a

    Professor
    SNUH
    Seoul, Seoul-t'ukpyolsi, Republic of Korea

    Disclosure information not submitted.

  • KP

    Kyu Joo Park, n/a

    Professor
    SNUH
    Seoul, Seoul-t'ukpyolsi, Republic of Korea

    All of the relevant financial relationships listed below have been mitigated.
    This individual has no financial relationships with ineligible companies.

  • SH

    Seung Chul Heo, n/a

    Clinical Professor
    BRMSNUH
    Seoul, Seoul-t'ukpyolsi, Republic of Korea

    Disclosure information not submitted.

  • JP

    Ji Won Park, n/a

    Clinical Professor
    SNUH
    Seoul, Seoul-t'ukpyolsi, Republic of Korea

    Disclosure information not submitted.

    • Senior Author(s)

  • RS

    Rumi Shin, n/a

    Clinical Professor
    Seoul National University Boramae Medical Center
    서울특별시, Seoul-t'ukpyolsi, Republic of Korea

    All of the relevant financial relationships listed below have been mitigated.
    This individual has no financial relationships with ineligible companies.

All of the relevant financial relationships listed below have been mitigated.
Jesung Park, MD, PhD: This individual has no financial relationships with ineligible companies.
Seung-Bum Ryoo, M.D., Ph.D.: This individual has no financial relationships with ineligible companies.
Kyu Joo Park, n/a: This individual has no financial relationships with ineligible companies.
Rumi Shin, n/a: This individual has no financial relationships with ineligible companies.
Image or Table (Do Not add text here):

Purpose/Background: Colorectal cancer recurrence remains a significant concern in patient outcomes. This study aimed to develop a machine learning model to predict recurrence in patients who underwent curative surgery for colorectal cancer.

Methods/Interventions: We conducted a retrospective analysis of 6,436 patients who underwent curative operations for colorectal cancer between 2012 and 2021. The model was developed using 37 clinically relevant features derived from patient demographics, cancer characteristics, and pathology data. Various machine learning algorithms were evaluated, with the ensemble model identified as the best-performing algorithm.

Results/Outcomes: The final ensemble model demonstrated robust performance with a receiver operating characteristic (ROC) area under the curve (AUC) of 0.793 (95% CI: 0.756-0.829) and a precision-recall curve (PRC) AUC of 0.366 (95% CI: 0.291-0.449). At the optimal threshold, the model exhibited a sensitivity of 0.7134 (95% CI: 0.6436-0.7852) and a specificity of 0.7648 (95% CI: 0.7390-0.7898). In the independent test set of 1,224 patients, those classified as high-risk by the model had significantly higher recurrence rates compared to the low-risk group (HR 6.806, 95% CI: 4.810-9.641, P< 0.001). This risk stratification remained significant within Stages II, III, and IV respectively.

Conclusion/Discussion: Our machine learning model shows promise in predicting colorectal cancer recurrence following curative surgery, offering a valuable tool for personalized risk stratification across different cancer stages and potentially improving patient outcomes through targeted interventions and surveillance protocols.