Development of a Machine Learning Model Using Limited Features to Predict 6-Month Mortality at Treatment Decision
George Chalkidis1, Jordan McPherson2, Anna Beck2
1Hitachi, Ltd, Tokyo, Japan.
JCO Clinical Cancer Informatics
|April 25, 2022
Summary
A machine learning model accurately predicts 6-month mortality in advanced solid tumor patients using limited electronic health record data. This tool aids in treatment decisions and communication of survival chances.
Area of Science:
- Oncology
- Medical Informatics
- Machine Learning
Background:
- Patients with advanced solid tumors often undergo intensive end-of-life treatments.
- Predicting short-term survival is crucial for treatment decisions.
Purpose of the Study:
- To develop a machine learning (ML) model using limited features to predict 6-month mortality at treatment decision points (TDPs) for patients with advanced solid tumors.
Main Methods:
- A cohort of 4,192 adult patients with advanced solid tumors was analyzed.
- Extreme gradient boosting was used to develop ML models predicting 6-month mortality after TDPs.
- Models utilized a limited set of 45 features derived from electronic health records.
Main Results:
- The ML model with 45 features accurately predicted 6-month mortality (Area Under the Curve ≥ 0.80).
- At a 0.3 risk threshold, the model distinguished between low (34% survival) and higher (81% survival) chances of 6-month survival.
- The positive predictive value of the limited feature model was 0.66.
Conclusions:
- A validated ML model using 45 readily available EHR features can predict 6-month prognosis in advanced solid tumor patients.
- This model can support shared decision-making regarding subsequent lines of therapy.
More Related Videos
07:15Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
7.0K
07:13Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
Published on: April 18, 2025
208
Related Concept Videos
01:21Cancer Survival Analysis
472
Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...
472
01:24Kaplan-Meier Approach
285
The Kaplan-Meier estimator is a non-parametric method used to estimate the survival function from time-to-event data. In medical research, it is frequently employed to measure the proportion of patients surviving for a certain period after treatment. This estimator is fundamental in analyzing time-to-event data, making it indispensable in clinical trials, epidemiological studies, and reliability engineering. By estimating survival probabilities, researchers can evaluate treatment effectiveness,...
285