AI Science Tracker
EN

Predicting Primary Refractory Diffuse Large B-Cell Lymphoma: A Comparison of Machine Learning Models with the IPI Score.

Grigore T. Popa University of Medicine and Pharmacy, 700115 Iasi, Romania; Faculty of Medicine and Biological Sciences, ‘Ștefan cel Mare’ University, 720229 Suceava, Romania · 27 Aug 2026 · medicine

Discovery

Predicting Primary Refractory Diffuse Large B-Cell Lymphoma: A Comparison of Machine Learning Models with the IPI Score.

Source check: Source checked. Not independent scientific confirmation.

Finding: Machine-learning models estimated primary refractory DLBCL in 369 retrospective single-center patients, with IPI strongest on the 73-patient holdout and gradient boosting best among ML models; ML did not outperform IPI. The result is a disease-specific prognostic comparison, and authors call for external validation.

Technology: Logistic regression, random forest, gradient boosting, SVM-RBF and MLP on baseline clinical/lab variables. Stratified 80/20 holdout, nested CV, repeated random splits; IPI holdout AUC 0.692 vs best ML gradient boosting 0.658; internal neste

Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.

AI contribution
Logistic regression, random forest, gradient boosting, SVM-RBF and MLP on baseline clinical/lab variables. Stratified 80/20 holdout, nested CV, repeated random splits; IPI holdout AUC 0.692 vs best ML gradient boosting 0.658; internal nested CV GB 0.712±0.049.
Test and result
Machine-learning models estimated primary refractory DLBCL in 369 retrospective single-center patients, with IPI strongest on the 73-patient holdout and gradient boosting best among ML models; ML did not outperform IPI. The result is a disease-specific prognostic comparison, and authors call for external validation. Evaluation: Machine-learning models estimated primary refractory DLBCL in 369 retrospective single-center patients, with IPI strongest on the 73-patient holdout and gradient boosting best among ML models; ML did not outperform IPI.
Limitation / still unknown
The 369-patient study is retrospective and single-center; the IPI outperformed the best machine-learning model on the holdout, and external validation remains necessary.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://mdpi-res.com/d_attachment/jcm/jcm-15-06628/article_deploy/jcm-15-06628.pdf

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 2/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open in the full catalog for filters, charts and related discoveries. Browse medicine discoveries.