Research Progress on the Application of Predictive Models in the Treatment of Colorectal Cancer

Chenyu Wu; Tianqing Yan

doi:10.54097/3p6zfd42

Authors

Chenyu Wu
Tianqing Yan

DOI:

https://doi.org/10.54097/3p6zfd42

Keywords:

Colorectal cancer, Predictive model, Supervised learning, Unsupervised learning, Data-driven optimization.

Abstract

Colorectal cancer (CRC), as a highly prevalent and lethal malignant tumor worldwide, has become a major public health issue requiring urgent global attention, posing a serious threat to the health of people across the globe. With the continuous advancement of medical research, predictive models for colorectal cancer treatment have been extensively studied and developed. This paper systematically reviews the construction methods, datasets used, and performance evaluation metrics of various mainstream colorectal cancer models both domestically and internationally. Through research and synthesis, it is found that while most models demonstrate good performance in internal validation, they also reveal numerous limitations in clinical practice. These limitations manifest as insufficient external validation, weak model generalization, and inadequate reporting standards and data sharing. Addressing these issues, this paper proposes corresponding solutions to provide practical reference for optimizing colorectal cancer treatment prediction models. This aims to advance the field toward greater scientific rigor and standardization.

References

[1] Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin, 2021, 71(3):209-249.

[2] Vapnik V N. The nature of statistical learning theory. Springer, 1995.

[3] Bishop C M. Pattern recognition and machine learning. Springer, 2006.

[4] Hastie T, Tibshirani R, Friedman J. The elements of statistical learning. Springer, 2009.

[5] Liu J, Wu B, Wang X, Luo D, Qiao C, Yang Y. Exploration on building a smart colorectal cancer diagnosis and prognosis prediction model based on private cloud platform. Beijing Journal of Traditional Chinese Medicine, 2023, 42(4): 369-375.

[6] Abdul Rahman H, Ottom M A, Dinov I D. Machine learning-based colorectal cancer prediction using global dietary data. BMC Cancer, 2023.

[7] Ma F. Construction of a risk prediction model for low anterior resection syndrome after rectal cancer surgery based on machine learning algorithms. Chengde Medical College, 2025.

[8] Yang X, Qiu H, Wang L, Wang X. Predicting colorectal cancer survival using time-to-event machine learning. J Med Internet Res, 2023.

[9] Zhao L Y, Luo Y J, Cheng X L, Cheng C S, Chen H L, Yu H W, & Wang X G. Interpretable machine learning prediction model for delayed bleeding risk following endoscopic resection of colorectal polyps. Huazhong University of Science and Technology (Medical Edition), 2025, 15(3): 13-20.

[10] Zhang P. Construction and validation of a predictive model for the occurrence and recurrence risk of advanced colorectal adenomas. Yan’an University, 2024.

[11] Usher-Smith J A, Harshfield A, Saunders CL, et al. Performance of FIT-based and other risk prediction models for CRC. Clinical Medicine, 2023.

[12] Wang Y, Li X, Chen H, et al. Evaluating generalizability of oncology trial results to real-world patients. Nature Medicine, 2024.

[13] Collins G S, Moons K G M, Riley R D, et al. Transparent reporting of clustered data prediction models: TRIPOD-Cluster checklist. BMJ, 2023.