Comparison of machine learning models to predict the risk of breast cancer-related lymphedema among breast cancer survivors: a cross-sectional study in China

Jiali Du ¹, Jing Yang ¹, Qing Yang ²

⁰Department of Breast Surgery, Sichuan Clinical Research Center for Cancer, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, Affiliated Cancer Hospital of University of Electronic Science and Technology of China, Chengdu, China.

Frontiers in Oncology +

|

February 27, 2024

View abstract on PubMed

Summary

Machine learning models can predict breast cancer-related lymphoedema (BCRL) risk. A logistic regression model demonstrated superior performance, enabling early identification and intervention for high-risk patients.

Area Of Science

Oncology
Medical Informatics
Biostatistics

Background

Postoperative breast cancer-related lymphoedema (BCRL) poses a significant clinical challenge.
Early identification of individuals at high risk for BCRL is crucial for effective management and prevention.
Current prediction methods may lack the precision needed for timely interventions.

Purpose Of The Study

To develop and validate machine learning models for predicting BCRL risk.
To identify high-risk individuals for early intervention and prevention of BCRL.
To enhance clinical decision-making in postoperative breast cancer care.

Main Methods

Retrospective study (2012-2022) involving 670 breast cancer patients.
Utilized Least Absolute Shrinkage and Selection Operator (LASSO) for feature selection.
Developed and evaluated nine classification models, selecting the optimal based on AUC.

Main Results

LASSO identified 13 key predictive features for BCRL.
The logistic regression (LR) model achieved the highest Area Under the Curve (AUC) of 0.87.
LR demonstrated strong performance: accuracy (0.81), sensitivity (0.79), specificity (0.82), precision (0.60), and F-score (0.68).

Conclusions

The developed logistic regression model accurately predicts BCRL risk.
This model aids in identifying patients at elevated risk for timely, tailored interventions.
Effective BCRL risk prediction supports nurses in preventing lymphoedema onset.

Keywords:

breast cancer cancer lymphoedema machine learning prediction model

Related Concept Videos

Cancer Survival Analysis

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...