The major effects of health-related quality of life on 5-year survival prediction among lung cancer survivors: applications of machine learning
Problem
Predicting 5-year survival among lung cancer survivors, and whether health-related quality of life (HRQOL) data improves prediction beyond clinical and sociodemographic variables.
Why it matters
Accurate survival prediction informs treatment and follow-up decisions for lung cancer patients.
What the AI did
Five machine learning algorithms (decision tree, logistic regression, bagging, random forest, AdaBoost) were trained on data from 809 surgery survivors and compared via fivefold cross-validation on two feature sets, with and without HRQOL factors.
Results
Feature set 1 model accuracies ranged 0.647–0.713; with HRQOL added (feature set 2), AdaBoost and random forest outperformed the others.
Impact
Shows HRQOL factors improve machine-learning-based 5-year survival prediction for lung cancer survivors.
Record
| Venue | Scientific Reports |
|---|---|
| Published | 2020 |
| Work type | article |
| Authors | Jin-Ah Sim, Young Ae Kim, Ju Han Kim, Jong Mog Lee, Moon Soo Kim, Young Mog Shim, Jae Ill Zo, Young Ho Yun |
| Citations | 63 (OpenAlex, at import time) |
| References | 62 |
| Model or method | decision tree, logistic regression, bagging, random forest, AdaBoost |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1038/s41598-020-67604-3 |
| Open access | gold |
| Keywords | machine learning, lung cancer, survival prediction, quality of life, AdaBoost, random forest |
Importance rating
real but narrower advance
Survival prediction informs treatment and follow-up decisions for lung cancer patients; HRQOL-augmented ML improved prediction among 809 surgery survivors — a real but narrow, single-cohort advance.