Validity of a Telemedicine System for the Evaluation of Acute-Phase Retinopathy of Prematurity
Bibliographic record
Abstract
IMPORTANCE: The present strategy to identify infants needing treatment for retinopathy of prematurity (ROP) requires repeated examinations of at-risk infants by physicians. However, less than 10% ultimately require treatment. Retinal imaging by nonphysicians with remote image interpretation by nonphysicians may provide a more efficient strategy. OBJECTIVE: To evaluate the validity of a telemedicine system to identify infants who have sufficiently severe ROP to require evaluation by an ophthalmologist. DESIGN, SETTING, AND PARTICIPANTS: An observational study of premature infants starting at 32 weeks' postmenstrual age was conducted. This study involved 1257 infants with birth weight less than 1251 g in neonatal intensive care units in 13 North American centers enrolled from May 25, 2011, through October 31, 2013. INTERVENTIONS: Infants underwent regularly scheduled diagnostic examinations by an ophthalmologist and digital imaging by nonphysician staff using a wide-field digital camera. Ophthalmologists documented findings consistent with referral-warranted (RW) ROP (ie, zone I ROP, stage 3 ROP or worse, or plus disease). A standard 6-image set per eye was sent to a central server and graded by 2 trained, masked, nonphysician readers. A reading supervisor adjudicated disagreements. MAIN OUTCOMES AND MEASURES: The validity of grading retinal image sets was based on the sensitivity and specificity for detecting RW-ROP compared with the criterion standard diagnostic examination. RESULTS: A total of 1257 infants (mean birth weight, 864 g; mean gestational age, 27 weeks) underwent a median of 3 sessions of examinations and imaging. Diagnostic examination identified characteristics of RW-ROP in 18.2% of eyes (19.4% of infants). Remote grading of images of an eye at a single session had sensitivity of 81.9% (95% CI, 77.4-85.6) and specificity of 90.1% (95% CI, 87.9-91.8). When both eyes were considered for the presence of RW-ROP, as would routinely be done in a screening, the sensitivity was 90.0% (95% CI, 85.4-93.5), with specificity of 87.0% (95% CI, 84.0-89.5), negative predictive value of 97.3%, and positive predictive value of 62.5% at the observed RW-ROP rate of 19.4%. CONCLUSIONS AND RELEVANCE: When compared with the criterion standard diagnostic examination, these results provide strong support for the validity of remote evaluation by trained nonphysician readers of digital retinal images taken by trained nonphysician imagers from infants at risk for RW-ROP. TRIAL REGISTRATION: clinicaltrials.gov Identifier:NCT01264276.
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 | 0.017 | 0.092 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| 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".