Selection of safe artemisinin derivatives using a machine learning-based cardiotoxicity platform and in vitro and in vivo validation.
Source check: Source checked. Not independent scientific confirmation.
Finding: A random-forest cardiotoxicity model and additional toxicity filters reduced a 374-derivative library to seven candidates. Artemisinin B was then examined in AC16 cells and zebrafish larvae; these preclinical assays do not establish human safety.
Technology: Random-forest cardiotoxicity models
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
- AI contribution
- A leave-one-out random-forest model predicted arrhythmia, cardiac failure, heart block, hypertension and myocardial infarction; DataWarrior and Pro-Tox-II added other toxicity filters.
- Test and result
- The screen retained 197/374 derivatives, then 17, then seven. Five were available for AC16 viability assays (three independent experiments, six parallel measurements each); one showed moderate cytotoxicity at high concentration. Artemisinin B was tested in 210 zebrafish larvae across seven groups of 30; one death at 0.5 pg/mL and pericardial edema at 1 pg/mL were reported.
- Limitation / still unknown
- The seven-candidate list is a computational screening outcome. The reported cell and zebrafish assays do not establish human safety or clinical cardioprotection; transcriptomic markers are indirect endpoints.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://doi.org/10.1007/s00204-021-03058-4
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: 18 Nov 2026. Review scope: PMC screening funnel, AC16 viability Methods/Results and Figure 3 caption, zebrafish Methods/Results, and publication metadata..
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 3/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.