AI Science Tracker
EN

University of Nottingham

1 discovery from 2025

Discoveries

Random forests identify surface geometry relationships associated with reduced bacterial attachment

Source check: Source checked. Not independent scientific confirmation.

Finding: A screen of 2176 polymer surface patterns supplied measurements for random-forest models linking geometry to bacterial attachment. Held-out topographies supported predictive associations between attachment and surface coverage, feature size and spacing; the models outperformed a simpler roughness descriptor. The anti-attachment hits themselves came from experimental screening, not AI selection. The represented result is a learned structure–response relationship, without attributing the separate biological mechanism or animal findings to AI.

Technology: Random forest; SHAP feature interpretation

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

AI contribution
Random forests learned geometry-to-attachment mappings from 68 topographical descriptors; SHAP identified influential geometric variables and conditional design relationships beyond a single roughness descriptor.
Test and result
Random-forest models trained on 70% of measured surface-attachment data and evaluated on 30% across 50 resamples; source-reported test R² values were 0.851 and 0.810 for the two modeled bacterial species. A simpler roughness descriptor performed less well.
Limitation / still unknown
No newly fabricated AI-selected surface was tested as a prospective design prediction. The experimental screen selected the best surfaces. Mechanistic and animal experiments are not represented as AI-enabled findings, and clinical device benefit remains unestablished.
Evidence sources and annotation context

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

Evidence sources: https://doi.org/10.1038/s41467-025-60567-x · https://pmc.ncbi.nlm.nih.gov/articles/PMC12177056/

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: Exact primary article-front identity/date; full relevant surface-screening, learned-model Methods and result paragraphs; fresh PubMed publication-status relationships. Only the material geometry/attachment association is represented..

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.