Truace
Subscribe

AI prediction models for breast cancer-related lymphedema

Source article: Artificial Intelligence for Prediction of Breast Cancer-Related Lymphedema: A Systematic Review

Background Breast cancer-related lymphedema (BCRL) impairs quality of life and function, and early detection is crucial for preventive intervention. AI models, particularly machine learning and deep learning, hold promise for improving BCRL prediction and personalized risk stratification. This systematic review evaluates AI applications for predicting BCRL and, where available, summarizes how their reported performance compared with traditional diagnostic or non-AI methods. Methods This systematic review followe…

By the Truvace desk1 min read1 sourcePeer-reviewedTRV-2026-1792· cite & verify
Artificial Intelligence for Prediction of Breast Cancer-Related Lymphedema: A Systematic Review
Cultivando la Salud: Breast and Cervical Cancer Education Program. Training Curriculum. Public domain
Trace impact readingNegative state
P 71The P score combines the specificity and measured human impact of the grounded problem claim with the strength of this Trace’s cited sources.

Both sides are scored from claims and sources, not community votes.

G 66The G score combines the specificity and measured human impact of the grounded gain claim with the strength of this Trace’s cited sources.

In brief

A systematic review searched major medical databases in July 2024 and included seven studies published between 2018 and 2024, covering 5284 patients. It examined how machine learning and deep learning models predict breast cancer-related lymphedema, a condition that harms function and quality of life.

The findings suggest AI could support earlier detection and more personalized risk stratification, which matters because early detection is key to preventive intervention. However, the evidence is not strong enough for clinical use, and larger multi-institutional datasets and prospective validation are still needed.

Main points

  1. Five of the seven studies reported classification performance, with accuracies ranging from 81% to 93.75%.
  2. Mean sensitivity across those studies was 85.5% and mean specificity was 87.6%.
  3. Commonly reported predictors were higher body mass index, greater lymph node burden, and chemotherapy or radiotherapy.
  4. Younger age and neoadjuvant chemotherapy were linked to higher predicted risk in some studies.

The gain

Across five studies reporting classification results, AI models predicted breast cancer-related lymphedema with a mean accuracy of 87.2%, mean sensitivity of 85.5%, and mean specificity of 87.6%.

The problem

The same BCRL prediction models have small samples, heterogeneous data, no external validation, and inconsistent calibration, so they are not yet clinically established.

The rundown

The review followed PRISMA guidelines and searched MEDLINE, EMBASE, the Cochrane Library, and Google Scholar. Studies were chosen using predefined inclusion and exclusion criteria, and accuracy, sensitivity, and specificity were extracted.

Where available, the authors summarized how reported AI performance compared with traditional diagnostic or non-AI methods. The abstract does not give the comparison results.

The authors conclude that the models are promising but exploratory, and call for standardized evaluation and prospective validation before they are integrated into clinical care.

What this doesn’t fix

The authors state that current models are exploratory, limited by small samples, heterogeneous data, absent external validation, and inconsistent calibration, so prospective validation is needed before clinical use.

Sources

  1. Peer-reviewedPlastic and Reconstructive Surgery - Global Open2026-10-06

The debate