Preprints with The Lancet is a collaboration between The Lancet Group of journals and SSRN to facilitate the open sharing of preprints for early engagement, community comment, and collaboration. Preprints available here are not Lancet publications or necessarily under review with a Lancet journal. These preprints are early-stage research papers that have not been peer-reviewed. The usual SSRN checks and a Lancet-specific check for appropriateness and transparency have been applied. The findings should not be used for clinical or public health decision-making or presented without highlighting these facts. For more information, please see the FAQs.
Predicting Mortality in Korean Colorectal Cancer Using Multi-Center Data Based on Explainable Ai: A Nationwide Multi-Center Cohort Study
Background: Colorectal cancer (CRC) ranks highest in incidence and third in mortality among all cancers in Korea. We aimed to develop a predictive classification model using artificial intelligence (AI) models and identify mortality risk factors.
Methods: All-cause and CRC-related mortality following CRC diagnosis were analyzed across all age groups. Data of patients diagnosed with CRC between 2012 and 2020 (221,197 individuals, with 63,610 deceased) registered in Korea’s Cancer Public Library Database were used. We predicted mortality following CRC diagnosis in 9,069 patients.
Findings: The XGB model excelled in predicting both all-cause and CRC-induced mortality (areas under the curve: 0·8432 (95% CI, 0·8218-0·8639) and 0·8734 (95% CI, 0·8513-0·8946), respectively). Common predictors of mortality included AJCC-T, N, and M stages; age and sex were significant among those aged ≥ 50 years.
Interpretation: Based on our findings, we propose a new model for identifying risk factors affecting CRC mortality in Korean patients.
Funding: This work was funded by the "Regional Innovation Strategy (RIS)" of the National Research Foundation of Korea (NRF), which is funded by the Ministry of Education (MOE) (2022RIS-005) and Institute for Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT) (No. RS-2022-II221196, Regional strategic Industry convergence security core talent training business).
Declaration of Interest: There is no conflict of interest.
Ethical Approval: We utilized data from the Cancer Public Library Database (CPLD) [19] in South Korea, with approval from the Institutional Review Board (IRB, KNUH-2024-01-002) of Kangwon National University Hospital.
Yeo, Nayoug and Kim, Tae-Hoon and Kang, Seonguk and Kim, Woo Jin and Kwon, Oh Beom and Nam, Seong-Joo and Yim, Inhyeok and Lee, Hui-Yonng and PARK, SANG WON, Predicting Mortality in Korean Colorectal Cancer Using Multi-Center Data Based on Explainable Ai: A Nationwide Multi-Center Cohort Study. Available at SSRN: https://ssrn.com/abstract=4964490 or http://dx.doi.org/10.2139/ssrn.4964490
Download This Paper