A One-Step, Streamlined Children’s Vision Screening Solution Based on Smartphone Imaging for Resource-Limited Areas: Design and Preliminary Field Evaluation
Bibliographic record
Abstract
BACKGROUND: Young children's vision screening, as part of a preventative health care service, produces great value for developing regions. Besides yielding a high return on investment from forestalling surgeries using a low-cost intervention at a young age, it improves school performance and thus boosts future labor force quality. Leveraging low-skilled health care workers with smartphones and automated diagnosis to offer such programs can be a scalable model in resource-limited areas. OBJECTIVE: This study aimed to develop and evaluate an effective, efficient, and comprehensive vision screening solution for school children in resource-limited areas. First, such an exam would need to cover the major risk factors of amblyopia and myopia, 2 major sources of vision impairment effectively preventable at a young age. Second, the solution must be integrated with digital patient record-keeping for long-term monitoring and popular statistical analysis. Last, it should utilize low-skilled technicians and only low-cost tools that are available in a typical school in developing regions, without compromising quality or efficiency. METHODS: A workflow for the screening program was designed and a smartphone app was developed to implement it. In the standardized screening procedure, a young child went through the smartphone-based photoscreening in a dark room. The child held a smartphone in front of their forehead, displaying pre-entered personal information as a quick response code that duplexed as a reference of scale. In one 10-second procedure, the child's personal information and interpupillary distance, relative visual axis alignment, and refractive error ranges were measured and analyzed automatically using image processing and artificial intelligence algorithms. The child's risk for strabismus, myopia, and anisometropia was then derived and consultation given. RESULTS: A preliminary evaluation of the solution was conducted alongside yearly physical exams in Luoyang, Henan, People's Republic of China. It covered 20 students with suspected strabismus and 80 randomly selected students, aged evenly between 8 and 10. Each examinee took about 1 minute, and a streamlined workflow allowed 3 exams to run in parallel. The 1-shot and 2-shot measurement success rates were 87% and 100%, respectively. The sensitivity and specificity of strabismus detection were 0.80 and 0.98, respectively. The sensitivity and specificity of myopia detection were 0.83 and 1.00, respectively. The sensitivity and specificity of anisometropia detection were 0.80 and 1.00, respectively. CONCLUSIONS: The proposed vision screening program is effective, efficient, and scalable. Compared with previously published studies on utilizing a smartphone for an automated Hirschberg test and photorefraction screening, this comprehensive solution is optimized for practicality and robustness, and is thus better ready-to-deploy. Our evaluation validated the achievement of the program's design specifications.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".