Development of Machine Learning Models for Predicting Radiation Dermatitis in Breast Cancer Patients Using Clinical Risk Factors, Patient-Reported Outcomes, and Serum Cytokine Biomarkers
Bibliographic record
Abstract
BACKGROUND: Radiation dermatitis (RD) is a significant side effect of radiotherapy experienced by breast cancer patients. Severe symptoms include desquamation or ulceration of irradiated skin, which impacts quality of life and increases healthcare costs. Early identification of patients at risk for severe RD can facilitate preventive management and reduce severe symptoms. This study evaluated the utility of subjective and objective factors, such as patient-reported outcomes (PROs) and serum cytokines, for predicting RD in breast cancer patients. The performance of machine learning (ML) and logistic regression-based models were compared. PATIENTS AND METHODS: Data from 147 breast cancer patients who underwent radiotherapy was analyzed to develop prognostic models. ML algorithms, including neural networks, random forest, XGBoost, and logistic regression, were employed to predict clinically significant Grade 2+ RD. Clinical factors, PROs, and cytokine biomarkers were incorporated into the risk models. Model performance was evaluated using nested cross-validation with separate loops for hyperparameter tuning and calculating performance metrics. RESULTS: Feature selection identified 18 predictors of Grade 2+ RD including smoking, radiotherapy boost, reduced motivation, and the cytokines interleukin-4, interleukin-17, interleukin-1RA, interferon-gamma, and stromal cell-derived factor-1a. Incorporating these predictors, the XGBoost model achieved the highest performance with an area under the curve (AUC) of 0.780 (95% CI: 0.701-0.854). This was not significantly improved over the logistic regression model, which demonstrated an AUC of 0.714 (95% CI: 0.629-0.798). CONCLUSION: Clinical risk factors, PROs, and serum cytokine levels provide complementary prognostic information for predicting severe RD in breast cancer patients undergoing radiotherapy. ML and logistic regression models demonstrated comparable performance for predicting clinically significant RD with AUC>0.70.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 source (direct Gemma or distilled Codex), 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".