Download PDF

Stratification of Risk of Developing Early-Onset Intestinal Cancer Using Machine Learning Techniques and Nutrition/Socioeconomic Variables Among Low-Income Young Adults in Morocco: A Case-Control Study

Hicham ELLOUAD, Anass BELBACHIR, Bouchra ASSARAG, Anass Doukkali, Assia SAYDTAHIRI
1 Laboratory of Excellence in Biotechnology, Pharmaceutical Bioengineering, and Artificial Intelligence, FMPM, Cadi Ayyad University, Marrakech, Morocco.
2 Public Health Department, National School of Public Health, Rabat, Morocco.
3 Laboratory of Analytical Chemistry, Faculty of Medicine and Pharmacy, Mohammed V University in Rabat, Morocco.
4 LARTID, National School of Applied Sciences, Cadi Ayyad University, Marrakech, Morocco.
h.ellouad.ced@uca.ac.ma, an.belbachir@uca.ac.ma, bassarag1@gmail.com, doukkali73@gmail.com, a.saydtahiri.ced@uca.ac.ma
How to cite: Hicham ELLOUAD, Anass BELBACHIR, Bouchra ASSARAG, Anass Doukkali, Assia SAYDTAHIRI, "Stratification of Risk of Developing Early-Onset Intestinal Cancer Using Machine Learning Techniques and Nutrition/Socioeconomic Variables Among Low-Income Young Adults in Morocco: A Case-Control Study", Sciences Methods and Technologies International Journal (SciMeTech), (2026) Vol 2, Issue 2, p 40-47
Abstract
The incidence of intestinal cancer has increased rapidly among young adults in Morocco due to the nutrition transition that involves the gradual shift from adherence to the traditional Mediterranean diet to the intake of highly processed foods. Despite the emergence of strong epidemiological evidence of the risk, there exists no machine learning algorithm that incorporates culture-specific nutritional risk factors in the risk assessment of early onset of colorectal cancer among young adults in Morocco.

This study seeks to build an interpretable machine learning system for predicting the risk of colorectal cancer among young adults in low-income settings in Morocco based on sixteen nutritional risk factors that apply to their dietary habits.

Three group case-control analysis are done at the hospitals of the Marrakech-Safi region in Morocco (sample size: 1,353 individuals; 459 young cancer patients, 447 young healthy controls, 447 older cancer patients). A set of machine learning algorithms are built and benchmarked: Elastic Net Logistic Regression, Balanced Random Forest, and Extreme Gradient Boosting (XGBoost) using Bayesian optimization with five-fold cross-validation. Extreme Gradient Boosting (XGBoost) yielded the highest classification accuracy (area under curve \(= 0.935\), cross-validated area under curve \(= 0.926 \pm 0.024\)). The decision curve analysis showed net clinical benefit at 0.43 vs 0.25 for non-stratified mass screening. Ultra-processed food intake and Mediterranean diet scores were the most influential predictors. Importantly, dietary quality improvement lowered an individual's predicted risk by \(93\%\) - from 0.913 to 0.062.

The proposed questionnaire-based ML model does not require any laboratory tests and can be implemented in outpatient practices regardless of resources. Early-onset colorectal cancer in Moroccan low-income young adults is a nutritional and socioeconomic disorder.
Keywords: Intestinal cancer, Machine Learning, early-onset, risk stratification

References

  1. Bray, F., Laversanne, M., Sung, H., Ferlay, J., Siegel, R. L., Soerjomataram, I., & Jemal, A. (2024). Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: a cancer journal for clinicians, 74(3), 229-263.
  2. Morgan, E., Arnold, M., Gini, A., Lorenzoni, V., Cabasag, C. J., Laversanne, M., ... & Bray, F. (2023). Global burden of colorectal cancer in 2020 and 2040: incidence and mortality estimates from GLOBOCAN. Gut, 72(2), 338-344.
  3. Louise, V. M. L. I., Zhao, Q., & Liu, L. (2024). African colorectal cancer burden in 2022 and projections to 2050. ecancermedicalscience, 18.
  4. Registre des Cancers de la Region du Grand Casablanca 2018-2021. (2022). Fondation Lalla Salma de Prévention et Traitement des Cancers, Maroc. Disponible sur : www.contrelecaner.ma
  5. Deoula, M. M. S. O., Huybrechts, I., El Kinany, K., Boudouaya, H., Hatime, Z., El Asri, A., ... & El Rhazi, K. (2020). Behavioral, nutritional, and genetic risk factors of colorectal cancers in morocco: protocol for a multicenter case-control study. JMIR Research Protocols, 9(1), e13998.
  6. World Cancer Research Fund / American Institute for Cancer Research. (2018). Diet, Nutrition, Physical Activity and Colorectal Cancer. Continuous Update Project Expert Report. London: World Cancer Research Fund International. Disponible sur : www.wcrf.org/dietandcancer
  7. El Asri, A., Ouldim, K., Bouguenouch, L., Sekal, M., Moufid, F. Z., Kampman, E., ... & El Rhazi, K. (2022). Dietary fat intake and KRAS mutations in colorectal cancer in a Moroccan population. Nutrients, 14(2), 318.
  8. Bouzini, G., Sammoud, K., Amzerin, M., Najdi, A., & El M'Rabet, F. Z. (2026). Fast food consumption patterns in Moroccan adults and their association with the risk of early-onset colorectal cancer. Clinical Nutrition ESPEN, 102976.
  9. El Kinany, K., Hatime, Z., El Asri, A., Benslimane, A., Deoula, M. M. S., Zarrouq, B., ... & El Rhazi, K. (2025). Adherence to the Mediterranean diet and colorectal cancer risk: a large case control study in the Moroccan population. Public Health Nutrition, 28(1), e48.
  10. Nartowt, B. J., Hart, G. R., Muhammad, W., Liang, Y., Stark, G. F., & Deng, J. (2020). Robust machine learning for colorectal cancer risk prediction and stratification. Frontiers in big Data, 3, 6.
  11. Du, G., Lv, H., Liang, Y., Zhang, J., Huang, Q., Xie, G., ... & Sun, Y. (2025). Population-based colorectal cancer risk prediction using a SHAP-enhanced LightGBM model. Frontiers in Oncology, 15, 1575844.
  12. Pu, J., Zhou, B., Yao, Y., Wu, Z., Wen, Y., Xu, R., & Xu, H. (2025). Development and validation of a lifestyle-based 10-year risk prediction model of colorectal Cancer for early stratification: evidence from a longitudinal screening cohort in China. Nutrients, 17(11), 1898.
  13. Gozum, S., Nimri, O. F., Merzah, M. A., & Vitorino, R. (2025). Cancer Screening and Prevention in MENA and Mediterranean Populations: A Multi-Level Analysis of Barriers, Knowledge Gaps, and Interventions Across Indigenous and Diaspora Communities. Diseases, 14(1), 10.
  14. Breiman, L. (2001). Random forests. Machine learning, 45(1), 5-32.
  15. Chen, T., & Guestrin, C. (2016, August). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining (pp. 785-794).
  16. Zou, H., & Hastie, T. (2005). Regularization and variable selection via the elastic net. Journal of the Royal Statistical Society Series B: Statistical Methodology, 67(2), 301-320.
  17. Lemaître, G., Nogueira, F., & Aridas, C. K. (2017). Imbalanced-learn: A python toolbox to tackle the curse of imbalanced datasets in machine learning. Journal of machine learning research, 18(17), 1-5.
  18. DeLong, E. R., DeLong, D. M., & Clarke-Pearson, D. L. (1988). Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics, 837-845.
  19. Hosmer Jr, D. W., Lemeshow, S., & Sturdivant, R. X. (2013). Applied logistic regression. John Wiley & Sons.
SciMeTech · Vol 2, Issue 2 · p 40-47