AI Science Tracker
EN

Sun Yat-Sen Memorial Hospital

1 discovery from 2026

Discoveries

Multimodal learning evaluates vertebroplasty cement-leakage risk

Source check: Source checked. Not independent scientific confirmation.

Finding: Six learned classifiers combined clinical, conventional CT and radiomic features to evaluate cement leakage after vertebroplasty. Development used 378 patients, with additional external and later-acquired validation cohorts of 45 and 50 patients. Patient grouping was retained during splits and tuning. Because injected cement volume is a predictor, this is a perioperative risk analysis rather than a purely preoperative forecast; treatment benefit was not tested.

Technology: XGBoost; random forest; support vector machine; learned classifiers

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

AI contribution
Learned classifiers integrate clinical, CT and selected radiomic variables to estimate perioperative cement-leakage probability.
Test and result
Patient-grouped training/tuning, internal testing and unchanged-model application to external and later-acquired cohorts.
Limitation / still unknown
Prospectively timed predictions before outcome ascertainment and benefit from model-guided injection planning remain unestablished.
Evidence sources and annotation context

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

Evidence sources: https://pmc.ncbi.nlm.nih.gov/articles/PMC13745800/

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: 21 Nov 2026. Review scope: Primary relevant Methods/results/limitations and exact front/canonical identity; current primary PubMed linked correction/retraction metadata checked..

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 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open the interactive profile for the timeline, or browse all organizations and model families.