AI Science Tracker
EN

University Hospital Basel

2 discoveries from 2023

Discoveries

Neural ECG classifier reduces inconclusive readings across four devices

Source check: Source checked. Not independent scientific confirmation.

Finding: A prospective diagnostic study compared a residual neural classifier with four device algorithms against physician-interpreted ECGs. Among 824 recordings from 206 patients, inconclusive results fell from 16% to 1.2%. This selected hospital study does not establish population-screening or workload benefit.

Technology: PulseAI residual convolutional neural network

Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.

AI contribution
A trained residual CNN classifies rhythm from raw single-lead signals.
Test and result
Near-contemporaneous physician-interpreted 12-lead references; blinded adjudication. Training contained no smart-device recordings.
Limitation / still unknown
52 of 258 recruited patients lacked complete recordings. Selected electrophysiology patients, vectorized exports and unrepeated inconclusive recordings limit generalization. Four recordings per participant are correlated. Vendor-affiliated authors participated; no clinical-outcome benefit tested.
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/PMC10879015/

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: Indexed primary clinical/model Methods, Results and limitations; exact front/date/current PubMed notices..

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.

Learned Holter analysis agrees with physician AF-burden estimates

Source check: Source checked. Not independent scientific confirmation.

Finding: A trained ECG-analysis system estimated atrial-fibrillation burden in seven-day recordings. In 47 recordings with intermediate burden, correlation with physician review was 0.998 and agreement bias was −0.006. This validates measurement agreement in a selected cohort, not treatment or outcome benefit.

Technology: Cardiomatics trained ECG rhythm-classification system

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

AI contribution
Previously trained algorithms classify ECG rhythm and aggregate detected AF duration into burden.
Test and result
100 recordings from82 patients;53 zero/full-burden recordings were separated from the47 intermediate-burden comparisons. Repeated recordings were assessed with a patient-random-intercept sensitivity analysis.
Limitation / still unknown
Small sample and few women limit transfer. Paced rhythms were excluded. Training architecture and training/test overlap were not fully documented in the study; no efficiency or clinical-outcome benefit measured.
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/PMC10123500/

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 indexed clinical/reference/software/statistical Methods, results and study limitations; exact fresh PubMed date/status..

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.