The Hydronephrosis Severity Index guides paediatric antenatal hydronephrosis management based on artificial intelligence applied to ultrasound images alone
Bibliographic record
Abstract
Antenatal hydronephrosis (HN) impacts up to 5% of pregnancies and requires close, frequent follow-up monitoring to determine who may benefit from surgical intervention. To create an automated HN Severity Index (HSI) that helps guide clinical decision-making directly from renal ultrasound images. We applied a deep learning model to paediatric renal ultrasound images to predict the need for surgical intervention based on the HSI. The model was developed and studied at four large quaternary free-standing paediatric hospitals in North America. We evaluated the degree to which HSI corresponded with surgical intervention at each hospital using area under the receiver-operator curve, area under the precision-recall curve, sensitivity, and specificity. HSI predicted subsequent surgical intervention with > 90% AUROC, > 90% sensitivity, and > 70% specificity in a test set of 202 patients from the same institution. At three external institutions, HSI corresponded with AUROCs ≥ 90%, sensitivities ≥ 80%, and specificities > 50%. It is possible to automatically and reliably assess HN severity directly from a single ultrasound. The HSI stratifies low- and high-risk HN patients thus helping to triage low-risk patients while maintaining very high sensitivity to surgical cases. HN severity can be predicted from a single patient ultrasound using a novel image-based artificial intelligence system.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".