Machine Learning for Predicting Postoperative Functional Disability and Mortality Among Older Patients With Cancer: Retrospective Cohort Study
Bibliographic record
Abstract
Background: The global cancer burden is rapidly increasing, with 20 million new cases estimated in 2022. The world population aged ≥65 years is also increasing, projected to reach 15.9% by 2050, making cancer control for older patients urgent. Surgical resection is important for cancer treatment; however, predicting postoperative disability and mortality in older patients is crucial for surgical decision-making, considering the quality of life and care burden. Currently, no model directly predicts postoperative functional disability in this population. Objective: We aimed to develop and validate machine-learning models to predict postoperative functional disability (≥5-point decrease in the Barthel Index) or in-hospital death in patients with cancer aged ≥ 65 years. Methods: This retrospective cohort study included patients aged ≥65 years who underwent surgery for major cancers (lung, stomach, colorectal, liver, pancreatic, breast, or prostate cancer) between April 2016 and March 2023 in 70 Japanese hospitals across 6 regional groups. One group was randomly selected for external validation, while the remaining 5 groups were randomly divided into training (70%) and internal validation (30%) sets. Predictor variables were selected from 37 routinely available preoperative factors through electronic medical records (age, sex, income, comorbidities, laboratory values, and vital signs) using crude odds ratios (P<.1) and the least absolute shrinkage and selection operator method. We developed 6 machine-learning models, including category boosting (CatBoost), extreme gradient boosting (XGBoost), logistic regression, neural networks, random forest, and support vector machine. Model predictive performance was evaluated using the area under the receiver operating characteristic curve (AUC) with 95% CI. We used the Shapley additive explanations (SHAP) method to evaluate contribution to the predictive performance for each predictor variable. Results: This study included 33,355 patients in the training, 14,294 in the internal validation, and 6711 in the external validation sets. In the training set, 1406/33,355 (4.2%) patients experienced worse discharge. A total of 24 predictor variables were selected for the final models. CatBoost and XGBoost achieved the largest AUCs among the 6 models: 0.81 (95% CI 0.80-0.82) and 0.81 (95% CI 0.80-0.82), respectively. In the top 15 influential factors based on the mean absolute SHAP value, both models shared the same 14 factors such as dementia, age ≥85 years, and gastrointestinal cancer. The CatBoost model showed the largest AUCs in both internal (0.77, 95% CI 0.75-0.79) and external validation (0.72, 95% CI 0.68-0.75). Conclusions: The CatBoost model demonstrated good performance in predicting postoperative outcomes for older patients with cancer using routinely available preoperative factors. The robustness of these findings was supported by the identical top influential factors between the CatBoost and XGBoost models. This model could support surgical decision-making while considering postoperative quality of life and care burden, with potential for implementation through electronic health records.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".