Sign in Subscribe
Cancer research

Machine Learning Model Predicts Postoperative Disability and Mortality in Older Cancer Patients

Strategic Insights

— 1 min read

Researchers from National Hospital Organization have developed and validated a machine-learning model to predict postoperative functional disability and mortality in older patients with cancer. According to the study, the global cancer burden is rapidly increasing, with an estimated 20 million new cases in 2022. The population aged 65 years is also increasing, projected to reach 15.9% by 2050, making cancer control for older patients urgent. The study aimed to develop a model to predict postoperative functional disability in this population, considering the quality of life and care burden.

Key Takeaways:

  • The global cancer burden is estimated to be 20 million new cases in 2022, with the population aged 65 years projected to reach 15.9% by 2050.
  • The researchers developed and validated a machine-learning model, CatBoost, to predict postoperative functional disability and mortality in older patients with cancer.
  • The model used routinely available preoperative factors, such as age, sex, income, comorbidities, laboratory values, and vital signs, to predict postoperative outcomes.
  • The CatBoost model showed good performance in predicting postoperative outcomes, with an AUC of 0.72 in external validation.
  • The model identified income, comorbidities, and laboratory values as significant predictors of postoperative functional disability and mortality.
  • The researchers concluded that the model could support surgical decision-making while considering postoperative quality of life and care burden.
  • The study highlights the importance of developing machine-learning models to predict postoperative outcomes in older patients with cancer.

Statistics:

  • 20 million new cases of cancer estimated in 2022 (global cancer burden)
  • 15.9% of the population projected to be aged 65 years by 2050
  • AUC of 0.72 for CatBoost model in external validation
  • 37 routinely available preoperative factors used in the model (age, sex, income, comorbidities, laboratory values, and vital signs)
  • 70 Japanese hospitals across 6 regional groups participated in the study
  • April 2016 to March 2023: the period during which the surgeries were performed

Sources:

  • Machine Learning for Predicting Postoperative Functional Disability and Mortality Among Older Patients With Cancer: Retrospective Cohort Study. JMIR Aging, 2025;8.
  • NewsRx. Researchers from National Hospital Organization Detail Findings in Cancer (Machine Learning for Predicting Postoperative Functional Disability and Mortality Among Older Patients With Cancer: Retrospective Cohort Study). Cancer Weekly. July 1, 2025; p 840.

Read more