EUVIMEDEUVIMED entdecken
ValidierungsstudieOpen Access

Evaluation of the Applicability of Synthetic Data in the Development of Colorectal Cancer Survival Prediction Models: External Validation of Advanced Machine Learning Models Based on National Cancer Data Center Data

Yujeong Jang, Jae Hoon Kwon, Heeyong Kim, You-Jin Joung, Junho Nang, Chang Hyun Kim

Journal of Medical Internet Research · 07.07.2026 · 28 · e86087-e86087

DOI: 10.2196/86087 ↗

Kurz gefasst

Validierungsstudie in der Zeitschrift Journal of Medical Internet Research, online erschienen am 07.07.2026. Verfasst von Yujeong Jang und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Regulatory Science, Kyung Hee University, Seoul, Republic of Korea. Hauptthemen laut MEDLINE: Colorectal Neoplasms, Machine Learning. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Limited data availability and privacy constraints hinder the development of robust survival prediction models for personalized treatment. Synthetic data offers a promising solution, preserving the statistical properties of real clinical data.

Objective

This study aimed to quantitatively assess the feasibility of using synthetic data for survival prediction by evaluating model transfer performance to real-world hospital data, with a focus on model transfer strategies.

Methods

We developed and validated colorectal cancer survival prediction models using the National Cancer Data Center (NCDC) synthetic data (30,683 patients from 3 Korean institutions) for pretraining and real hospital data (2170 patients from Hwasun Jeonnam University Hospital) for external validation. We evaluated 3 model transfer strategies—domain adaptation, zero-shot, and ensemble—using extreme gradient boosting (XGBoost) and light gradient boosting machine (LightGBM). In total, 48 model configurations were tested, defined by the combination of algorithms (LightGBM and XGBoost), sampling technique (no-sampling, random undersampling [RUS], and synthetic minority oversampling technique combined with edited nearest neighbors [SMOTEENN]), model type (baseline, domain adaptation, zero-shot, and ensemble), and optimization objective (area under the precision-recall curve [AUPRC] and F 1 ). The outcome was 7-year overall survival, evaluated using the AUPRC and Brier scores. Performance was compared against a hospital-only baseline using absolute values and deltas (ΔAUPRC and ΔBrier). Differences and corresponding 95% CIs were estimated on the held-out test set using 2000 bootstrap samples.

Results

Zero-shot application reduced the AUPRC in most settings, and any marginal improvements observed in the remaining settings were not statistically significant. In contrast, the domain adaptation model improved AUPRC in 8/12 combinations, with 4 statistically significant gains; the best setting (XGBoost+RUS+ F 1 optimization) achieved AUPRC=0.5391 (Δ+0.1474; P

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Colorectal NeoplasmsMachine LearningHumansBoosting Machine Learning AlgorithmsSurvival AnalysisPrediction AlgorithmsRepublic of KoreaPredictive Learning ModelsFemale

Stichwörter der Autor:innen

colorectal cancerdomain adaptationexternal validationextreme gradient boostinglight gradient boosting machinemachine learningsurvival prediction modelsynthetic data

Autor:innen und Institutionen

  • Department of Regulatory Science, Kyung Hee University, Seoul, Republic of Korea.
  • Department of Surgery, Chonnam National University Hwasun Hospital, 322 Seoyang-ro, Hwasun, 58128, Republic of Korea.
  • C&R Research, Seoul, Republic of Korea.
  • Clinical Trial Center, Institute for Biomedical Science, Chonnam National University Hwasun Hospital, 322, Seoyang-ro, Hwasun-eup, Hwasun-gun, Jeollanam-do, Republic of Korea, 82 10-8940-1365.

Zitieren

APA

Jang, Y., Kwon, J. H., Kim, H., Joung, Y. J., Nang, J., & Kim, C. H. (2026). Evaluation of the Applicability of Synthetic Data in the Development of Colorectal Cancer Survival Prediction Models: External Validation of Advanced Machine Learning Models Based on National Cancer Data Center Data. Journal of Medical Internet Research, 28, e86087-e86087. https://doi.org/10.2196/86087

Vancouver

Jang Y, Kwon JH, Kim H, Joung YJ, Nang J, Kim CH. Evaluation of the Applicability of Synthetic Data in the Development of Colorectal Cancer Survival Prediction Models: External Validation of Advanced Machine Learning Models Based on National Cancer Data Center Data. Journal of Medical Internet Research. 2026;28:e86087-e86087. doi:10.2196/86087

BibTeX
@article{Jang2026,
  author = {Jang, Yujeong and Kwon, Jae Hoon and Kim, Heeyong and Joung, You-Jin and Nang, Junho and Kim, Chang Hyun},
  title = {Evaluation of the Applicability of Synthetic Data in the Development of Colorectal Cancer Survival Prediction Models: External Validation of Advanced Machine Learning Models Based on National Cancer Data Center Data},
  journal = {Journal of Medical Internet Research},
  year = {2026},
  volume = {28},
  pages = {e86087--e86087},
  doi = {10.2196/86087}
}
Kostenlos starten →