AI Science Tracker

Neural network ensemble for prenatal detection of complex congenital heart disease

University of California, San Francisco; Harvard Medical School · 14 May 2021 · medicine

Discovery

Neural network ensemble for prenatal detection of complex congenital heart disease

An ensemble of neural networks trained on 107,823 fetal ultrasound images reached an AUC of 0.99, 95% sensitivity and 96% specificity in an internal test set of 4,108 fetal surveys with 0.9% CHD. Its sensitivity was comparable to that of clinicians and it stayed robust on outside-hospital and lower-quality images. Caveat: retrospective study with a low-prevalence internal test set.

Technology: Ensemble of convolutional neural networks with segmentation models

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

Result, AI contribution and limits

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

  • Publication status: Peer-reviewed paper
  • Test type: retrospective or benchmark
  • Tested: Trained on 107,823 images from 1,326 retrospective studies; on 4,108 fetal surveys (0.9% CHD), AUC 0.99, 95% sensitivity, 96% specificity, 100% negative predictive value; comparable to clinicians and robust on outside-hospital images.
  • Still unknown: Prospective screening performance is not reported in the abstract.
  • AI role: An ensemble of neural networks identifies cardiac views and classifies normal versus complex congenital heart disease.

Evidence sources: https://doi.org/10.1038/s41591-021-01342-5

Source checked · editorial details

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

Source type: paper. Paper: linked separately.

Review scope and date are not recorded; source checked does not establish full-text review.

Citations: in the top 1% of works of the same type, year and subfield (OpenAlex work metrics, 345 citations; as of 15 Nov 2026). Attention, not a measure of correctness; not part of the score.

Editorial assessment: Impact 4/5 · Validation 2/5 · Novelty 4/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Follow-up and relationships

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.

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