Standalone Smartphone Photorefraction: High Accuracy for Timely Myopia Screening

Standalone Smartphone Photorefraction: High Accuracy for Timely Myopia Screening | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Standalone Smartphone Photorefraction: High Accuracy for Timely Myopia Screening Ling-yau Kiu, Lily YL Chan, Oi-lam Kwok, Ho-cheung Leung, Wing-yu Tam, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8873852/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract Background Recent smartphone-based health technologies emphasize engineering, but evidence on standalone devices, without attachments, for detecting uncorrected refractive errors remains limited. Validation is essential for scalable vision screening in resource-limited settings. This study was conducted to validate smartphone-based eccentric photorefraction against open-field autorefraction and assess comparable performance under non-cycloplegic (dry) and cycloplegic (wet) conditions. Methods This diagnostic study (2022–2024), baseline for the Smartphone AI Refraction System (STARS), enrolled 948 Chinese children and adults (aged 6–43 years; mean [SD], 19.8 [9.03] years; spherical equivalent, − 12.50 to + 6.88 D) with clear media, excluding presbyopia, tropia, or nystagmus. This study was conducted at the Hong Kong Polytechnic University Optometry Research Clinic using convenience sampling. Participants underwent open-field autorefraction and smartphone photorefraction at 1 m distance under non-cycloplegic and cycloplegic conditions. We evaluated the refractive error agreement with intraclass correlation coefficient (ICC) and Bland-Altman analysis. Diagnostic performance for myopia worse than − 2.00 D (2021 AAPOS criteria) was assessed using sensitivity, specificity, and diagnostic odds ratio (DOR). Results We analysed 5,350 non-cycloplegic and 4,919 cycloplegic eye photographs. Non-cycloplegic photorefraction demonstrated higher sensitivity (96%; 95% CI, 96–97) and specificity (83%; 95% CI, 81–85) for detecting significant myopia compared with cycloplegic photorefraction (91% (95% CI, 90–92); 79% (95% CI, 77–81)). ICCs were 0.77 (non-cycloplegic) and 0.67 (cycloplegic). Non-cycloplegic photorefraction also showed higher DOR than cycloplegic photorefraction. Conclusions Standalone smartphone photorefraction without external attachments offers high accuracy, particularly under noncycloplegic conditions, and supports large-scale refractive errors screening in underserved areas. Myopia Teleophthalmology Smartphone Refraction Photoscreening Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Uncorrected refractive errors are the leading cause of visual impairment worldwide, particularly in low- and middle-income regions, where they often emerge during childhood [ 1 ]. Untreated before age 7 to 8 years, these errors can lead to amblyopia and permanent vision loss [ 2 ]. Early detection is critical, given effective myopia control strategies now available [ 3 ]. Among schoolchildren, myopia prevalence is rising alarmingly [ 4 , 5 ], projected to affect nearly half the global population by 2050 across developing and developed regions [ 6 , 7 ]. Uncorrected cases may progress to high myopia, increasing the subsequent risks of ocular complications such as glaucoma [ 8 , 9 ] and retinal detachment [ 10 ], underscoring the need for scalable screening tools to prevent avoidable visual disability as a public health priority. Instrument-based objective screening methods, including photoscreeners [ 11 – 13 ], offer established protocols superior to traditional retinoscopy or autorefraction, requiring minimal expertise and no bulky equipment. These devices leverage eccentric photorefraction, described by Bobier and Braddick [ 14 ], in which camera flashlight rays reflect off the retina as a crescent-shaped reflex that indicates refractive error. Specialized polarized lights, optical filters, and sometimes infrared components enhance contrast and reduce glare but increase complexity and cost, restricting deployment in resource-limited settings. Consequently, underserved children remain undiagnosed [ 15 ], potentially impairing academic performance [ 16 ], daily activities, and psychosocial well-being [ 17 , 18 ] despite treatable conditions. Smartphones' global ubiquity offers an ideal platform for adapting eccentric photorefraction. Existing smartphone-based techniques for estimating refractive errors require external attachments [ 19 , 20 ], limiting scalability and widespread adoption. In addition, like other photoscreeners, these devices also typically operate at close working distances, potentially inducing accommodation artifacts that compromise accuracy. This study develops and validates a foundational model, the Smartphone AI Refraction System (STARS), that uses only the smartphone’s built-in camera and flash for photoscreening, eliminating the need for add-on hardware. Its performance was evaluated against open-field autorefraction under non-cycloplegic (dry) and cycloplegic (wet) conditions, and accommodation effects were assessed in both children and adults. These investigations position standalone smartphones as effective, low-cost diagnostic tool for myopia screening in resource-limited settings. Methods Study Design In this prospective diagnostic study, smartphone photoscreening (index test) was compared with open-field autorefraction (reference standard). Participants Children and adults were recruited via convenience sampling at the Optometry Research Clinic, Centre for Myopia Research, The Hong Kong Polytechnic University, between 2022 and 2024. This diagnostic study adhered to the Declaration of Helsinki, was approved by the university's Human Subjects Ethics Sub-Committee (HSEARS #20210401006), and followed the Standards for Reporting of Diagnostic Accuracy Studies (STARD) reporting guideline. Written informed consent was obtained from all participants or their legal guardians (for those aged < 18 years). Eligible participants, screened by experienced optometrists for clear ocular media and absence of tropia, presbyopia, or nystagmus, underwent non-cycloplegic followed by cycloplegic assessments. Standard optometric pre-testing confirmed no contraindications to cycloplegia. Objective Refractive Error Assessment All participants underwent open-field autorefraction (WAM-5500; Grand Seiko Co Ltd, Tokyo, Japan) to measure non-cycloplegic (dry) refractive error (Auto Dry ) while viewing fixation targets at 3 m, representing distance refractive error, and at 1 m, replicating smartphone photorefraction conditions. Open-field autorefraction served as the reference standard for comparison with smartphone photorefraction results. Subjective refraction was performed before cycloplegia, followed by cycloplegic (wet) autorefraction (Auto Wet ) at 3 m and 1 m for additional comparisons. Smartphone Testing (Image Capture) Photorefraction images were captured using the default iPhone XS telephoto lens camera (iOS 15.0; Apple Inc; Cupertino, CA; 12-megapixel resolution; focal length, 6 mm [f/2.4]). Participants were seated 1 m from the smartphone, which was mounted on a tripod with a rotatable knob to align the camera at eye level and ensure accurate orientation. All measurements were conducted under dim lighting (~ 15 lux). Images were obtained at 3 meridians (horizontal 180°, vertical 90°, oblique 135°) to assess meridional refractive powers (Fig. 1 ), with a 3-second rest between captures. Testing was performed under non-cycloplegic (Photo Dry ) and cycloplegic (Photo Wet ) conditions when dilation was suitable. Test order (open-field autorefraction vs. smartphone photorefraction) was randomized and conducted by the same examiner. Smartphone Photo Analysis All smartphone-captured images were extracted for computation. Refractive error (X) was determined from pupil diameter and crescent width, with myopic errors identified by crescent location using equations previously described by Chan [ 21 , 22 ] (Fig. 2 ). Myopic X = \(\:\frac{e}{d(2r-s)}+\frac{1}{d}\) where d = distance from entrance pupil to the eye and e is the eccentricity of the flashlight. Pupil diameter ( 2r ) and crescent width ( s ) were measured using ImageJ software (National Institutes of Health) by image readers masked to refraction results. Unreadable images were defined as those with unclear pupil margins, indistinct crescents, or crescents obstructed by eyelids. Images with distinct pupil margins but no crescent reflex were classified as "no crescent". To compare photorefraction-derived refractive error with autorefraction result, the latter was converted to vector components P aligned with the photo-taking meridians (horizontal, vertical, oblique) using the equation: \(\:P\left(\theta\:\right)=S+C\times\:{sin}^{2}(\alpha\:-\theta\:)\) , where \(\:\theta\:\) is the meridian-of-interest (180°, 90°, 135° in radians), S is the spherical power, C is the cylindrical power and \(\:\alpha\:\) is the cylinder axis in radians [ 23 ]. Data and Statistical Analysis Photorefraction and autorefraction were first compared at the same viewing distance of 1 m under non-cycloplegic (dry) and cycloplegic (wet) conditions. Agreement between methods and test reliability were assessed using intraclass correlation coefficients (ICCs, 2-way mixed, single measures) with 95% confidence interval (CI).95% limit of agreement was estimated with Bland-Altman analyses. Mean absolute errors (MAEs) and standard errors of mean (SEM) between autorefraction and photorefraction were computed for refractive error categories (> 0 D, − 0.25 to − 2.00D, <−2.00 to − 4.00 D, <−4.00 to − 6.00D and < − 6.00D) under dry and wet conditions. The performance of dry and wet smartphone photorefraction was evaluated using receiver operating characteristic (ROC) curves across myopic thresholds from − 1.00 to − 6.00 D. Sensitivity and specificity (with 95% CI) for dry and wet photorefraction were calculated using the 2021 American Association for Pediatric Ophthalmology and Strabismus (AAPOS) instrument-based referral criterion of myopia greater than 2.00 D [ 24 ]. Diagnostic odd ratios (DORs) [ 25 ], defined as \(\:\frac{True\:Positive}{False\:Negative}/\frac{False\:Positive}{True\:Negatvie}\) , with 95% CIs were computed for both conditions to assess whether smartphone photorefraction performed differently under dry versus wet conditions. To account for accommodation induced by the closer working distance of smartphone-based photorefraction, mean difference between non-cycloplegic (dry) autorefraction (Auto Dry ) at distance (3 m) and near (1 m), and between cycloplegic (wet) autorefraction (Auto Wet ) at 3 m and 1 m, were calculated. Generalized linear models with robust standard errors (HC0) were used to examine the effects of age group (adults vs. children) and refractive error category (five severity levels) on the distance–near differences under dry and wet conditions. Mean distance-near autorefraction differences specific to age group and myopic severity under dry and wet conditions were used to derive calibration factors, which were incorporated into the corresponding photorefraction measurements. This adjustment compensated for the near working distance inherent to photorefraction and improved the accuracy of distance refractive error estimation. The calibrated photorefraction values were then compared with distance autorefraction and subjective refraction in terms of MAE, sensitivity, specificity, and DOR. Analyses were conducted using SPSS version 28 (IBM, Armonk, NY), with R version 4.4.1 (R Foundation for Statistical Computing; Vienna, Austria) used for graphical representations. Statistical significance was set at P < 0.05. Results Photorefraction Image Reading A total of 984 participants were recruited, of whom 36 were excluded because of tropia, media opacities, or presbyopia. Among 948 Chinese participants included in the analysis (aged 6–43 years; mean [standard deviation (SD)] age, 19.8 [9.03] years; 38% male; 59% adults ≥ 18 years), 5,350 non-cycloplegic eye photographs (Photo Dry ) were analyzed (spherical equivalent (SE) range, − 12.50 to + 6.88 D; mean [standard error of mean (SEM)] SE, − 3.50 [0.06] D). Of these, 876 participants (aged 6–43 years; mean [SD] age, 20.1 [9.09] years; 38% male; 61% adults) completed cycloplegic testing (Photo Wet : 4,919 images; SE range, − 12.13 to + 8.88 D; mean [SEM] SE, − 3.26 [0.06] D), with non-completion primarily due to refusal of dilation or narrow angles (Fig. 3 ). Unreadable images were excluded (6.17%). Mean refractions of subjective refraction, dry/wet photorefraction, and autorefraction were shown in Table 1 . Table 1 Detailed Summary between Dry and Wet Groups Mean refractions of non-cycloplegic and cycloplegic measurements Dry Wet Method Mean [SEM]; (range) a , D Mean (SEM), D Photorefraction (1 m) -3.61 [0.03]; (-12.28 to + 4.60) -3.04 [0.03]; (-12.53 to + 3.75) Autorefraction (1 m) -3.66 [0.03]; (-13.65 to + 6.70) -3.29 [0.04]; (-12.75 to + 8.75) Autorefraction (3 m) -3.50 [0.04]; (-13.49 to + 7.23) -3.26 [0.04]; (-12.97 to + 9.48) Subjective Refraction (6 m) -3.31 [0.03]; (-11.99 to + 6.19) NA Agreement between photorefraction and autorefraction at 1 m Accuracy Dry Wet 180-meridian ICC (95% CI) 0.795 (0.776–0.812) 0.727 (0.703–0.749) Sensitivity 0.95 (0.94–0.97) 0.91 (0.89–0.93) Specificity 0.86 (0.83–0.88) 0.78 (0.75–0.81) PPV 0.93 (0.91–0.94) 0.87 (0.85–0.89) NPV 0.90 (0.88–0.93) 0.84 (0.81–0.87) 90-meridian ICC (95% CI) 0.763 (0.741–0.783) 0.645 (0.614–0.673) Sensitivity 0.98 (0.97–0.99) 0.92 (0.90–0.94) Specificity 0.77 (0.74–0.81) 0.82 (0.78–0.85) PPV 0.92 (0.90–0.93) 0.93 (0.91–0.94) NPV 0.93 (0.91–0.96) 0.81 (0.77–0.84) 135-meridian ICC (95% CI) 0.750 (0.726–0.772) 0.655 (0.624–0.683) Sensitivity 0.95 (0.94–0.97) 0.91 (0.90–0.93) Specificity 0.84 (0.81–0.87) 0.79 (0.75–0.82) PPV 0.93 (0.92–0.95) 0.90 (0.88–0.91) NPV 0.89 (0.87–0.92) 0.82 (0.79–0.85) All meridians ICC (95% CI) 0.771 (0.759–0.783) 0.670 (0.654–0.686) Sensitivity 0.96 (0.96–0.97) 0.91 (0.90–0.92) Specificity 0.83 (0.81–0.85) 0.79 (0.77–0.81) PPV 0.93 (0.92–0.93) 0.90 (0.89–0.91) NPV 0.91 (0.89–0.92) 0.82 (0.80–0.84) Comparison of photorefraction and near autorefraction at 1 m Dry Wet Refractive Error Range No. Gradable Photos MAE [SEM], D within ± 1.00 D, % No. Gradable Photos MAE [SEM], D within ± 1.00 D, % > 0 D b 154 1.21 [0.08] 56 418 1.52 [0.05] 40 -0.25 to -2.00 D 725 0.77 [0.02] 73 873 0.93 [0.03] 65 < -2.00 to -4.00 D 1548 0.59 [0.01] 84 1579 0.73 [0.01] 75 < -4.00 to -6.00 D 1258 1.05 [0.02] 52 1056 1.40 [0.02] 37 < -6.00 D b 869 2.30 [0.04] 24 630 2.96 [0.06] 14 Overall at 1 m 4554 c 1.09 [0.02] 61 4556 c 1.30 [0.02] 53 Abbreviation: D, dioptres, ICC, Intraclass Correlation Coefficient; MAE, Mean Absolute Error; PPV, Positive Predictive Value; NPV, Negative Predictive Value; No. Number; SEM, Standard Error of Mean a Mean, standard error of mean (SEM) and ranges were calculated based on all meridional refractive errors at 180-, 90- and 135- meridians. b Meridional refractive error for near autorefractions ranged from − 13.65 to + 6.70 D (Dry) and − 12.75 to + 8.75 D (Wet) respectively. c Photos graded as “no-crescent” were excluded in this analysis. Comparison of Photorefraction with Autorefraction Readings Photorefraction results were compared with corresponding autorefraction measurements for each of the three meridians at the same viewing distance of 1 m. Each meridian showed good agreement with autorefraction under both non-cycloplegic (dry) and cycloplegic (wet) conditions (Table 1 ). ICC values for all meridians were 0.77 (95% CI, 0.76–0.78) under dry conditions and 0.67 (95% CI, 0.65–0.69) under wet conditions. MAE [SEM] between autorefraction and photorefraction was smallest when refractive error range fell within < − 2.00 to − 4.00 D, 0.59 [0.01] and 0.73 [0.01] in dry and wet conditions, respectively (Table 1 ). Bland-Altman plots demonstrated that most differences between methods lay within the 95% limits of agreement in both conditions (Fig. 4 ). Across all meridians, the area under the receiver operating characteristic curve (AUC) exceeded 0.90 under dry conditions and 0.84 under wet conditions for myopia thresholds from − 1.00 D to − 6.00 D, indicating good discriminative performance (Fig. 4 ). When analysed according to the 2021 AAPOS instrument-based referral criterion (myopia more than − 2.00 D), dry photorefraction showed slightly higher sensitivity and specificity (96% (95% CI, 96–97) and 83% (95%CI, 81–85), respectively) than wet photorefraction (91% (95% CI, 90–92) and 79% (95% CI, 77–81), respectively), with AUCs of 0.97 (95%CI, 0.96–0.97) (dry) and 0.91 (95% CI, 0.90–0.92) (wet). Dry photorefraction had a higher diagnostic odd ratio (DOR) of 123.1 (95% CI, 99.5-152.2) than wet photorefraction (41.2; 95% CI, 34.7–48.9). Comparison of Accommodation Effect on Smartphone-based Photorefraction The mean [SEM] distance–near autorefraction difference was 0.15 [0.01] D under dry conditions and 0.03 [0.01] D under wet conditions (P 0.05), whereas refractive error category showed a significant effect (P 0.05) nor myopic refractive error category (P > 0.05) showed significant effects on the distance-near accommodation, and autorefraction differences across refractive categories and between age groups were not clinically significant. Comparison between Photorefraction, Distance Autorefraction and Subjective Refraction Based on the distance–near autorefraction differences, calibration factors accounting for photorefraction working distance and stratified by age group and myopia severity were derived and applied to both dry and wet photorefraction measurements. The performance of photorefraction compared with conventional distance autorefraction is shown in Fig. 6 (dry: sensitivity 94% (95%CI, 93–95), specificity 85% (95% CI, 83–87); wet: sensitivity 91% (95% CI, 90–92), specificity 80% (95%CI, 78–82)), demonstrating consimilar performance across age groups under both conditions. To align with clinical standards, calibrated non-cycloplegic photorefraction measurements were compared with subjective refraction. Photorefraction demonstrated strong diagnostic accuracy for detecting myopia, with a sensitivity of 96% (95% CI, 95–96), specificity of 82% (95% CI, 81–84), DOR of 102.1 (95% CI, 83.5-124.9) and an AUC of 0.96 (95% CI, 0.96–0.97) (Fig. 6 ). MAE [SEM] was 0.79 [0.01] D for myopia fell between − 0.25 to − 6.00 D. When benchmarked against cycloplegic autorefraction, calibrated dry photorefraction maintained the robust performance for screening significant myopia (sensitivity: 95% (95% CI, 95–96); specificity: 81% (95% CI, 79–83); DOR: 89.0 (95% CI, 72.6–109.0); AUC: 0.96 (95% CI, 0.95–0.96)), with an MAE [SEM] of 0.83 [0.01] D within − 0.25 to − 6.00 D of myopia. Discussion This study represents a large-scale, first-in-class diagnostic evaluation of a standalone smartphone-based platform that uses only its built-in camera and flashlight to detect refractive errors, benchmarked against clinical standards under both non-cycloplegic and cycloplegic conditions. The baseline STARS model validated the core eccentric photorefraction principles, providing a robust foundation for subsequent integration of AI-based image analysis. Non-cycloplegic photorefraction demonstrated superior performance, achieving 96% sensitivity and 83% specificity for detecting significant myopia, following AAPOS 2021 guidelines, outperforming cycloplegic readings (sensitivity 91%, specificity 79%). Multiple myopia thresholds were evaluated, with corresponding AUCs values exceeding 0.90 across low to high myopia ranges, indicating that STARS can not only screen for myopia but also estimate the magnitude of refractive error with high accuracy. Accommodative effort was comparable between myopic children and adults across conditions, supporting the method’s utility for vision screening across age groups and settings. Previous research has largely focused on comparing non-cycloplegic photoscreening with cycloplegic refraction [ 26 – 30 ], and relatively few studies [ 31 , 32 ] have directly contrasted non-cycloplegic and cycloplegic photoscreening against a cycloplegic standard. In this context, the present findings, demonstrating better performance of smartphone photoscreening without eyedrops, support that non-cycloplegic STARS photoscreening performs consistently and accurately relative to cycloplegic autorefraction, reinforcing the feasibility of this standalone platform without the need for cycloplegia. Beyond its internal validation metrics, it is important to contextualize the performance of STARS against existing photoscreening technologies. The performance of STARS is comparable to that of commercially available photoscreeners in terms of mean difference and screening accuracy. For example, the 2WIN (Adaptica SRL, Padova, Italy) demonstrated a mean difference of 0.60 D compared with an autorefractor in 44 undilated myopic participants [ 33 ]. The Spot Vision Screener (Welch Allyn Inc, Skaneateles Falls, NY) has reported sensitivity and specificity values of 60–96% and 94–99%, respectively, for screening significant myopia greater than − 2.00 D without cycloplegia [ 13 , 34 ]. Yan and colleagues [ 35 ] reported a myopic screening performance of 46/99% (sensitivity/specificity) by PlusoptiX A12C (PlusoptiX GmbH, Nuremberg, Germany) in a pediatric population. GoCheck Kids (Goquity Mobile Health), a smartphone-based photoscreener using photorefraction with an auxiliary attachment, has reported sensitivity and specificity of 52–91% and 68–90% for detecting amblyopic risk factors [ 19 , 20 , 36 – 38 ]. For detecting significant myopia according to 2021 AAPOS guideline, its sensitivity and specificity were 100 and 79.1%, respectively, although the PPV was 38% [ 38 ], which is lower than the PPV of 93% observed in STARS models for predicting myopia. In addition, Luo et al. developed a smartphone application tested in 201 myopic eyes with refractive errors up to − 10.00 D, yielding a MAE (SD) of 0.65 (0.83) D, relying on active patient feedback [ 39 ]. Taken together, these findings suggest that the STARS model provides an accurate and sensitive objective refraction method based on portable and accessible technology that does not require additional hardware accessories or patient response. Given smartphones have gained widespread adoption especially over the past decades where coverage has extended to even less developed countries [ 40 , 41 ], the application value of our investigated smartphone system, with its validated agreement in detecting significant refractive error under non-cycloplegic conditions, is generalizable to typical screening environments where diagnostic agents may be unavailable or constrained to administer. This also highlights the public health relevance of our novel approach. Even though screening accuracy for myopia was similarly high in both dry and wet conditions, as reflected by their MAEs against conventional standards, the dry photorefraction screening framework ultimately offers a methodology that is more practical and cost-effective. This advantageous strategy reduces the need for professional manpower during vision screening compared to wet photorefraction. In addition, as we disseminated the specific concerns of research participants, we estimate that approximately one in ten more parents would be willing to have their child undergo preliminary vision screening if no eye drops were required, based on the proportion of participants who opted out of wet photorefraction during data collection. Our study also assessed a diverse age range where a similarly good performance was observed between the two age groups. Moreover, the applicability of our model is expected to apply relevantly to East and South Asian ethnic groups who are disproportionally affected by the myopia epidemic, with some of the highest prevalence rates globally over the past few decades [ 7 ]. Pupil diameter is often more difficult to estimate in darkly coloured irises; however, the smart device was able to keep up to this need. Limitations This study did not directly compare the smartphone-based method with existing portable photoscreeners, which are not universally available internationally and are primarily designed for preliminary detection rather than serving as clinical standards. Potential pseudo-myopia arising from the near working distance was addressed through calibration; although the refractive estimates were statistically greater under dry than wet conditions, the mean differences of 0.03–0.15 D were clinically negligible. In addition, further studies and validation are needed in broader populations, including younger children (< 6 years) and individuals with hyperopia. The controlled clinic-based design supported the underlying optical principles but did not evaluate real-world community deployment, in which environmental variability may affect measurement performance. Given the refractive error profile and age distribution of the participants, the model shows strong potential as a low-cost, cycloplegia-free option for large-scale myopia screening. Future studies should evaluate usability, scalability, and health economic impact in diverse settings and primary care settings. Implications for Practice Global organizations, including the World Health Organization, have deployed mobile applications for visual acuity screening [ 42 – 45 ] that typically rely on self-reported responses. Integrating objective refractive-error assessment with visual acuity testing would better align community screening with clinical standards and improve diagnostic structure, particularly in settings lacking traditional equipment. External validation across diverse populations and field environments is a critical next step to broaden access to reliable vision screening in underserved areas. From a feasibility perspective, the STARS framework offers a promising, scalable approach for community-level implementation and could serve as a cost‑effective alternative for large-scale deployment. Given the projected public‑health burden of myopia, prioritizing on‑field evaluation and implementation research will be essential to determine real‑world effectiveness, acceptability, and health‑economic impact. Conclusions A standalone smartphone-based photorefraction system shows promise as a scalable, translational solution for vision screening in under-resourced communities. Images captured by the device contain sufficient information to accurately quantify refractive errors, particularly myopia, with performance comparable to conventional clinical techniques. These findings support its potential as a cost-effective screening tool to detect visually significant myopia and facilitate timely management and intervention. Abbreviations AAPOS American Association for Pediatric Ophthalmology and Strabismus AUC Area under curve DOR Diagnostic odd ratio Dry Non-cycloplegic condition CI Confidence interval D Diopter ICC Intraclass correlation coefficients MAE Mean absolute error PPV Positive Predictive Value NPV Negative Predictive Value ROC Receiver Operating Characteristic SD Standard deviation SE Spherical equivalent SEM Standard errors of mean Wet Cycloplegic condition Declarations Ethnics approval and consent to participate This diagnostic study adhered to the Declaration of Helsinki, was approved by the university's Human Subjects Ethics Sub-Committee (HSEARS #20210401006), and followed the Standards for Reporting of Diagnostic Accuracy Studies (STARD) reporting guideline. Written informed consent was obtained from all participants or their legal guardians (for those aged < 18 years). Consent for publication Not applicable. The manuscript does not contain personal data (smartphone photorefraction photos), only refractive errors analysed and calculated from the photos. Funding This work was supported by Health Medical Research Fund (18191351), Innovation and Technology Fund (ITS/049/23), and Hong Kong PhD Fellowship Scheme (PF23-93959) by the Hong Kong Government, the Hong Kong Polytechnic University (Grants no.: 1-WZ1B, 1-WZ0L, 1-BD50), and the Tianjin Key Medical Discipline Construction Project (TJYXZDXK-3-004A-2). Authors’ contributions CWD, LC, GN, EF, and HVL contributed to the funding acquisition, conception and design of the study. CWD, GN, BD, and RW contributed to project administration and supervision. LYK, OLK, and HCL contributed to data acquisition. WYT and YYW did the image analysis. PL contributed to the data analysis plan. LYK did the data analysis. LC and LYK drafted the original manuscript. CWD reviewed and edited the manuscript. PL accessed and verified the raw data of the study. All authors read and approved the final manuscript. Acknowledgement We thank Dr. Wing Tang, Miss Lena Li, Mr. Kenny Chung and Miss Lotus Lai for their valuable contributions during the early stages of the conceptual work in preliminary testing and validation. Availability of data and materials The datasets generated and/or analysed during the current study are not publicly available due to an AI algorithm in development but are available from the corresponding author on reasonable request. Competing interests The authors declare that they have no competing interests. References Tarczy-Hornoch K, Cotter SA, Borchert M, et al. Prevalence and causes of visual impairment in Asian and non-Hispanic white preschool children: Multi-ethnic Pediatric Eye Disease Study. Ophthalmology. 2013;120(6):1220–6. 10.1016/j.ophtha.2012.12.029 . Webber AL, Wood J. Amblyopia: prevalence, natural history, functional effects and treatment. Clin Exp Optom. 2005;88(6):365–75. 10.1111/j.1444-0938.2005.tb05102.x . Zaabaar E, Asiamah R, Kyei S, Ankamah S. Myopia control strategies: A systematic review and meta-meta-analysis. Ophthalmic Physiol Opt. 2025;45(1):160–76. 10.1111/opo.13417 . Do CW, Chan LYL, Tse ACY, et al. Association between Time Spent on Smart Devices and Change in Refractive Error: A 1-Year Prospective Observational Study among Hong Kong Children and Adolescents. Int J Environ Res Public Health. 2020;17(23):8923. 10.3390/ijerph17238923 . Zhang XJ, Zhang Y, Kam KW, et al. Prevalence of Myopia in Children Before, During, and After COVID-19 Restrictions in Hong Kong. JAMA Netw Open. 2023;6(3):e234080. 10.1001/jamanetworkopen.2023.4080 . Holden BA, Fricke TR, Wilson DA, et al. Ophthalmology. 2016;123(5):1036–42. 10.1016/j.ophtha.2016.01.006 . Global Prevalence of Myopia and High Myopia and Temporal Trends from 2000 through 2050. Liang J, Pu Y, Chen J, et al. Global prevalence, trend and projection of myopia in children and adolescents from 1990 to 2050: a comprehensive systematic review and meta-analysis. Br J Ophthalmol. 2025;109(3):362–71. 10.1136/bjo-2024-325427 . Xu L, Wang Y, Wang S, Wang Y, Jonas JB. High myopia and glaucoma susceptibility the Beijing Eye Study. Ophthalmology. 2007;114(2):216–20. 10.1016/j.ophtha.2006.06.050 . Ha A, Kim CY, Shim SR, Chang IB, Kim YK. Degree of Myopia and Glaucoma Risk: A Dose-Response Meta-analysis. Am J Ophthalmol. 2022;236:107–19. 10.1016/j.ajo.2021.10.007 . Baba T, Ohno-Matsui K, Futagami S, et al. Prevalence and characteristics of foveal retinal detachment without macular hole in high myopia. Am J Ophthalmol. 2003;135(3):338–42. 10.1016/s0002-9394(02)01937-2 . Watts P, Walker K, Beck L. Photoscreening for refractive errors in children and young adults with severe learning disabilities using the MTI photoscreener. Eye (Lond). 1999;13(3):363–8. 10.1038/eye.1999.92 . Moghaddam AA, Kargozar A, Zarei-Ghanavati M, Najjaran M, Nozari V, Shakeri MT. Screening for amblyopia risk factors in pre-verbal children using the Plusoptix photoscreener: a cross-sectional population-based study. Br J Ophthalmol. 2012;96(1):83–6. 10.1136/bjo.2010.190405 . Peterseim MMW, Trivedi RH, Monahan SR et al. Effectiveness of the Spot Vision Screener using updated 2021 AAPOS guidelines. J AAPOS. 2023, 27(1): 24.e1–24.e7. 10.1016/j.jaapos.2022.11.019 Bobier WR, Braddick OJ. Eccentric photorefraction: optical analysis and empirical measures. Am J Optom Physiol Opt. 1985;62(9):614–20. World Health Organization. Report of the 2030 targets on effective coverage of eye care. World Health Organization; 2022. Latif MZ, Hussain I, Afzal S, et al. Impact of Refractive Errors on the Academic Performance of High School Children of Lahore. Front Public Health. 2022;10:869294. 10.3389/fpubh.2022.869294 . Wong HB, Machin D, Tan SB, Wong TY, Saw SM. Visual impairment and its impact on health-related quality of life in adolescents. Am J Ophthalmol. 2009;147(3):505–e5111. 10.1016/j.ajo.2008.09.025 . Li D, Chan VF, Virgili G, et al. Impact of Vision Impairment and Ocular Morbidity and Their Treatment on Depression and Anxiety in Children: A Systematic Review. Ophthalmology. 2022;129(10):1152–70. 10.1016/j.ophtha.2022.05.020 . Arnold RW, Arnold AW, Hunt-Smith TT, Grendahl RL, Winkle RK. The Positive Predictive Value of Smartphone Photoscreening in Pediatric Practices. J Pediatr Ophthalmol Strabismus. 2018;55(6):393–6. 10.3928/01913913-20180710-01 . Debert I, Costa DRD, Polati M, Falabretti JG, Susanna Junior R. Vision screening using a smartphone platform. Rev Paul Pediatr. 2022;40:e2020021. 10.1590/1984-0462/2022/40/2020021IN . Chan OY. Hong Kong Polytechnic. Dept. of Optometry and Radiography. Vision Screening in Pre-School Children. [dissertation]. The Hong Kong Polytechnic University: HKP Dept. of Optometry and Radiography; 1994. Chan OY, Edwards M, Brown B. Calibration and validity of an eccentric photorefractor. Ophthalmic Physiol Opt. 1996;16(3):203–10. Naeser K, Behrens JK. Correlation between polar values and vector analysis. J Cataract Refract Surg. 1997;23(1):76–81. 10.1016/s0886-3350(97)80154-x . Arnold RW, Donahue SP, Silbert DI et al. AAPOS uniform guidelines for instrument-based pediatric vision screen validation 2021. J AAPOS. 2022, 26(1): 1.e1–1.e6. 10.1016/j.jaapos.2021.09.009 Glas AS, Lijmer JG, Prins MH, Bonsel GJ, Bossuyt PM. The diagnostic odds ratio: a single indicator of test performance. J Clin Epidemiol. 2003;56(11):1129–35. 10.1016/s0895-4356(03)00177-x . Bloomberg JD, Suh DW. The accuracy of the plusoptiX A08 photoscreener in detecting risk factors for amblyopia in central Iowa. J AAPOS. 2013;17(3):301–4. 10.1016/j.jaapos.2013.03.014 . Peterseim MM, Papa CE, Wilson ME, et al. Photoscreeners in the pediatric eye office: compared testability and refractions on high-risk children. Am J Ophthalmol. 2014;158(5):932–8. 10.1016/j.ajo.2014.07.041 . Yan XR, Jiao WZ, Li ZW, Xu WW, Li FJ, Wang LH. Performance of the Plusoptix A09 photoscreener in detecting amblyopia risk factors in Chinese children attending an eye clinic. PLoS ONE. 2015;10(6):e0126052. 10.1371/journal.pone.0126052 . Huang D, Chen X, Zhang X et al. Pediatric vision screening using the plusoptiX A12C photoscreener in Chinese preschool children aged 3 to 4 years. Sci Rep. 2017, 7(1): 2041. 10.1038/s41598-017-02246-6 Vilà-de Muga M, Van Esso D, Alarcon S, et al. Instrument-based screening for amblyopia risk factors in a primary care setting in children aged 18 to 30 months. Eur J Pediatr. 2021;180(5):1521–7. 10.1007/s00431-020-03904-0 . Abrahamsson M, Ohlsson J, Björndahl M, Abrahamsson H. Clinical evaluation of an eccentric infrared photorefractor: the PowerRefractor. Acta Ophthalmol Scand. 2003;81(6):605–10. 10.1046/j.1600-0420.2003.0149.x . Ayse YK, Onder U, Suheyla K. Accuracy of Plusoptix S04 in children and teens. Can J Ophthalmol. 2011;46(2):153–7. 10.3129/i10-110 . Kanclerz P, Przewłócka K, Arnold RW. Agreement in non-cycloplegic and cycloplegic refraction between a photoscreener and a calibrated autorefractor. BMC Ophthalmol. 2024;24(1):130. 10.1186/s12886-024-03375-z . Mudie LI, Pickett K, Ross K, McCourt E, Enzenauer R. Performance of the Spot Vision Screener in children with Down syndrome and other special needs. J AAPOS. 2023, 27(5): 274.e1-274.e7. 10.1016/j.jaapos.2023.07.011 Yan Q, Li R, Qian Y, et al. Instrument referral criteria for PlusoptiX and SureSight based on 2021 AAPOS guidelines: A population-based study. Front Public Health. 2022;10:959757. 10.3389/fpubh.2022.959757 . Walker M, Duvall A, Daniels M et al. Effectiveness of the iPhone GoCheck Kids smartphone vision screener in detecting amblyopia risk factors. J AAPOS. 2020, 24(1): 16.e1–16.e5. 10.1016/j.jaapos.2019.10.007 Otto H, Deschoemaeker M, Van Overmeire B, Casteels I, Cassiman C. Validation of the eye screening tool GoCheck Kids for the detection of amblyopia risk factors in toddlers in Flanders. J AAPOS. 2024;28(5):104008. 10.1016/j.jaapos.2024.104008 . Applebaum SS, Sopeyin A, Mohamedali A, et al. Comparison of the GoCheck Kids and Spot Screener photoscreening devices for the detection of amblyopia risk factors using 2021 AAPOS recommendations. J AAPOS. 2024;28(6):104035. 10.1016/j.jaapos.2024.104035 . Luo G, Lee CY, Shivshanker P, et al. Preliminary Evaluation of a Smartphone App for Refractive Error Measurement. Transl Vis Sci Technol. 2022;11(2):40. 10.1167/tvst.11.2.40 . Hoffmann IM, Andersen AM, Lund S, Nygaard U, Joshua D, Poulsen A. Smartphone apps hold promise for neonatal emergency care in low-resource settings. Acta Paediatr. 2024;113(12):2526–33. 10.1111/apa.17410 . Poulsen A, Hickie IB, Alam M, et al. Access to mobile health in lower middle-income countries: a review. Health Technol. 2025;15:333–43. 10.1007/s12553-025-00948-w . Bastawrous A, Rono HK, Livingstone IAT, et al. Development and Validation of a Smartphone-Based Visual Acuity Test (Peek Acuity) for Clinical Practice and Community-Based Fieldwork. JAMA Ophthalmol. 2015;133(8):930–7. 10.1001/jamaophthalmol.2015.1468 . World Health Organization. World report on vision. Geneva: World Health Organization. 2019. Available from: https://www.who.int/publications/i/item/9789241516570 Abdul Rahman SNA, Naing NN, Othman AM, et al. Validity and Reliability of Vis-Screen Application: A Smartphone-Based Distance Vision Testing for Visual Impairment and Blindness Vision Screening. Med (Kaunas). 2023;59(5):912. 10.3390/medicina59050912 . Aman S, Ahmad NU, Sadiq MAA, et al. Implementation outcomes of a school-based visual screening program using the peek tool in low-resource settings in Pakistan. BMC Public Health. 2025;25(1):4120. 10.1186/s12889-025-25112-x . Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 24 Feb, 2026 Reviewers invited by journal 24 Feb, 2026 Editor assigned by journal 16 Feb, 2026 First submitted to journal 13 Feb, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8873852","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":596439590,"identity":"9086ad8b-d27e-4dfe-9030-9e735804f062","order_by":0,"name":"Ling-yau Kiu","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Ling-yau","middleName":"","lastName":"Kiu","suffix":""},{"id":596439591,"identity":"fd9cff1c-ad2a-4bda-93eb-69513d893146","order_by":1,"name":"Lily YL Chan","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Lily","middleName":"YL","lastName":"Chan","suffix":""},{"id":596439592,"identity":"cd840b69-ae2e-4615-9b0f-1cadb206cbb1","order_by":2,"name":"Oi-lam Kwok","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Oi-lam","middleName":"","lastName":"Kwok","suffix":""},{"id":596439593,"identity":"4ccba68d-e9e6-4468-a02d-2c9338b77a69","order_by":3,"name":"Ho-cheung Leung","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Ho-cheung","middleName":"","lastName":"Leung","suffix":""},{"id":596439594,"identity":"91d8c076-615d-4a5c-9154-fc5085f26a2b","order_by":4,"name":"Wing-yu Tam","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Wing-yu","middleName":"","lastName":"Tam","suffix":""},{"id":596439595,"identity":"844c21f9-4e22-4956-aa3d-02bf7beb8cf3","order_by":5,"name":"Yui-yi Wong","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Yui-yi","middleName":"","lastName":"Wong","suffix":""},{"id":596439596,"identity":"54651233-9f79-4f58-a651-7ee10d943176","order_by":6,"name":"Eugene Fu","email":"","orcid":"","institution":"The Education University of Hong Kong","correspondingAuthor":false,"prefix":"","firstName":"Eugene","middleName":"","lastName":"Fu","suffix":""},{"id":596439597,"identity":"02cfa382-2452-45d5-90e8-5aba8e4a65f3","order_by":7,"name":"Paul H Lee","email":"","orcid":"","institution":"University of Southampton","correspondingAuthor":false,"prefix":"","firstName":"Paul","middleName":"H","lastName":"Lee","suffix":""},{"id":596439598,"identity":"ca00d2fd-c4d7-4a41-b1ed-4a910e6f95bb","order_by":8,"name":"Hong-va Leong","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Hong-va","middleName":"","lastName":"Leong","suffix":""},{"id":596439599,"identity":"1276e26c-cc2d-4237-9960-7b47c53dc069","order_by":9,"name":"Grace Ngai","email":"","orcid":"","institution":"The Hong Kong Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Grace","middleName":"","lastName":"Ngai","suffix":""},{"id":596439600,"identity":"c8372990-8579-40c6-b97d-8bea3b8f4021","order_by":10,"name":"Bei Du","email":"","orcid":"","institution":"Tianjin Medical University Eye Hospital","correspondingAuthor":false,"prefix":"","firstName":"Bei","middleName":"","lastName":"Du","suffix":""},{"id":596439601,"identity":"70e458e4-03fb-4e1e-a575-6a2d677470f0","order_by":11,"name":"Wei Ruihua","email":"","orcid":"","institution":"Tianjin Medical University Eye Hospital","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Ruihua","suffix":""},{"id":596439602,"identity":"534fbaf8-2042-4837-9468-8d604bfe5799","order_by":12,"name":"Chi-Wai Do","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9UlEQVRIiWNgGAWjYDACCcYGEMnAcID5AANDAVjMgIGxIYEYLWwJIMXEaIExDvAYEKeFf3ZzAzNPjQUD3/Gebw8+GBxObGBv3ibBuCMNtyV3DgK1HJNgkDxzdrvhDJAWnmNlEoxncnBqMZBIbP+dwybBYHAjd5s0D0iLRI6ZBGNbBT4tDcw5/0Bacp5BtMi/IUJLbhtYCxvUFh6QFtwOk7gB1PK3T4JH8swxc6Bf0o3beNKKLRLbcHuff0b6A8YZ3+rk+I43P3vwocJatp/98MYbH9uScWqBAR4gZgPiZjDJkEBQAwSAFNcRqXYUjIJRMApGEgAAeq1Q+MgLk2oAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0002-7720-7507","institution":"The Hong Kong Polytechnic University","correspondingAuthor":true,"prefix":"","firstName":"Chi-Wai","middleName":"","lastName":"Do","suffix":""}],"badges":[],"createdAt":"2026-02-13 16:19:34","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8873852/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8873852/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104399227,"identity":"b162dcfa-b4fe-4b65-a250-1aa75b53ddae","added_by":"auto","created_at":"2026-03-11 12:05:11","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":570090,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSmartphone Orientation\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8873852/v1/b90bcb19f2dafae388161094.jpeg"},{"id":103596889,"identity":"729619f0-05d8-4216-8a82-605f7a73fd9a","added_by":"auto","created_at":"2026-02-27 13:20:49","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":136170,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePhotorefraction Optic Principle in a Myopic Eye\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8873852/v1/6dd61ab673b53467ae38f0ee.png"},{"id":103596887,"identity":"1eeaee16-6f90-409e-aa30-f950eba56c46","added_by":"auto","created_at":"2026-02-27 13:20:49","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":76484,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePatient Recruitment Flowchart\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8873852/v1/1f250436e50f5b97b63c792f.png"},{"id":103596892,"identity":"af56d8aa-471c-45e7-8a7c-2327b2d0b3a3","added_by":"auto","created_at":"2026-02-27 13:20:50","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1427831,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDry and Wet ROC curves for myopic thresholds and Bland-Altman across meridians (for 180-/90-/135-/all meridians)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) ROC curves for non-cycloplegic (dry) conditions at individual meridians; (B) ROC curves for cycloplegic (wet) conditions at individual meridians; (C) Bland-Altman plots for dry conditions; (D) Bland-Altman plots for wet conditions.\u003c/p\u003e\n\u003cp\u003eHigher AUCs under dry conditions (Row A-B) reflect superior pass/fail classification per AAPOS 2021 guidelines and other myopic thresholds.\u003c/p\u003e\n\u003cp\u003eOn the Bland-Altman plots (Row C-D), most points fall within 95% limits of agreement and expected refractive error ranges, with minimal outliers at low/moderate errors\u003cem\u003e.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8873852/v1/91242aa7f40bc871da2d9f79.jpeg"},{"id":103596890,"identity":"daa160e3-fea9-43dc-973f-074ef638f2ab","added_by":"auto","created_at":"2026-02-27 13:20:49","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":381922,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDistance–near autorefraction differences across age groups and refractive error under non-cycloplegic and cycloplegic conditions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAge groups were divided into 2 groups, children (\u0026lt;18 years old) and adults (≥18 years old).\u003c/p\u003e\n\u003cp\u003eDistance-near autorefraction difference was significantly higher under dry than wet condition. Under dry condition, age groups did not significantly affect the distance-near autorefraction differences, while hyperopic participants demonstrated significantly higher accommodation. Under wet condition, neither age groups nor refractive errors significantly affected the distance-near autorefraction differences.\u003c/p\u003e","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8873852/v1/d421c876b6b2426dc05c97fd.jpeg"},{"id":103596891,"identity":"e08c2903-892d-43c8-8f3d-59831521b2c3","added_by":"auto","created_at":"2026-02-27 13:20:49","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":232850,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePerformance of calibrated photorefraction in children and adults (dry/wet),ROC curves compared with autorefraction/ subjective refraction\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Performance of photorefraction with calibration factors in children and adults under dry and wet conditions; (B) Performance of calibrated photorefraction compared with standard distance autorefraction and subjective refraction; (C) ROC curves for calibrated photorefraction compared with standard autorefraction (dry and wet respectively); (D) ROC curve of dry calibrated photorefraction compared with subjective refraction; (E) ROC curve of dry calibrated photorefraction compared to wet standard autorefraction.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8873852/v1/d5000afce0a9065c34244956.png"},{"id":104407528,"identity":"8b032d8d-6f67-4c66-9f8d-870fd76dae3a","added_by":"auto","created_at":"2026-03-11 12:38:35","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4054959,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8873852/v1/01237153-e65b-401f-8497-8d1089c9d063.pdf"}],"financialInterests":"","formattedTitle":"Standalone Smartphone Photorefraction: High Accuracy for Timely Myopia Screening","fulltext":[{"header":"Introduction","content":"\u003cp\u003eUncorrected refractive errors are the leading cause of visual impairment worldwide, particularly in low- and middle-income regions, where they often emerge during childhood [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Untreated before age 7 to 8 years, these errors can lead to amblyopia and permanent vision loss [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Early detection is critical, given effective myopia control strategies now available [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Among schoolchildren, myopia prevalence is rising alarmingly [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e], projected to affect nearly half the global population by 2050 across developing and developed regions [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Uncorrected cases may progress to high myopia, increasing the subsequent risks of ocular complications such as glaucoma [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e] and retinal detachment [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e], underscoring the need for scalable screening tools to prevent avoidable visual disability as a public health priority.\u003c/p\u003e \u003cp\u003eInstrument-based objective screening methods, including photoscreeners [\u003cspan additionalcitationids=\"CR12\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], offer established protocols superior to traditional retinoscopy or autorefraction, requiring minimal expertise and no bulky equipment. These devices leverage eccentric photorefraction, described by Bobier and Braddick [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], in which camera flashlight rays reflect off the retina as a crescent-shaped reflex that indicates refractive error. Specialized polarized lights, optical filters, and sometimes infrared components enhance contrast and reduce glare but increase complexity and cost, restricting deployment in resource-limited settings. Consequently, underserved children remain undiagnosed [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], potentially impairing academic performance [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e], daily activities, and psychosocial well-being [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] despite treatable conditions.\u003c/p\u003e \u003cp\u003eSmartphones' global ubiquity offers an ideal platform for adapting eccentric photorefraction. Existing smartphone-based techniques for estimating refractive errors require external attachments [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e], limiting scalability and widespread adoption. In addition, like other photoscreeners, these devices also typically operate at close working distances, potentially inducing accommodation artifacts that compromise accuracy. This study develops and validates a foundational model, the Smartphone AI Refraction System (STARS), that uses only the smartphone\u0026rsquo;s built-in camera and flash for photoscreening, eliminating the need for add-on hardware. Its performance was evaluated against open-field autorefraction under non-cycloplegic (dry) and cycloplegic (wet) conditions, and accommodation effects were assessed in both children and adults. These investigations position standalone smartphones as effective, low-cost diagnostic tool for myopia screening in resource-limited settings.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy Design\u003c/h2\u003e \u003cp\u003eIn this prospective diagnostic study, smartphone photoscreening (index test) was compared with open-field autorefraction (reference standard).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eParticipants\u003c/h3\u003e\n\u003cp\u003eChildren and adults were recruited via convenience sampling at the Optometry Research Clinic, Centre for Myopia Research, The Hong Kong Polytechnic University, between 2022 and 2024. This diagnostic study adhered to the Declaration of Helsinki, was approved by the university's Human Subjects Ethics Sub-Committee (HSEARS #20210401006), and followed the Standards for Reporting of Diagnostic Accuracy Studies (STARD) reporting guideline. Written informed consent was obtained from all participants or their legal guardians (for those aged\u0026thinsp;\u0026lt;\u0026thinsp;18 years).\u003c/p\u003e \u003cp\u003eEligible participants, screened by experienced optometrists for clear ocular media and absence of tropia, presbyopia, or nystagmus, underwent non-cycloplegic followed by cycloplegic assessments. Standard optometric pre-testing confirmed no contraindications to cycloplegia.\u003c/p\u003e\n\u003ch3\u003eObjective Refractive Error Assessment\u003c/h3\u003e\n\u003cp\u003eAll participants underwent open-field autorefraction (WAM-5500; Grand Seiko Co Ltd, Tokyo, Japan) to measure non-cycloplegic (dry) refractive error (Auto\u003csub\u003eDry\u003c/sub\u003e) while viewing fixation targets at 3 m, representing distance refractive error, and at 1 m, replicating smartphone photorefraction conditions. Open-field autorefraction served as the reference standard for comparison with smartphone photorefraction results. Subjective refraction was performed before cycloplegia, followed by cycloplegic (wet) autorefraction (Auto\u003csub\u003eWet\u003c/sub\u003e) at 3 m and 1 m for additional comparisons.\u003c/p\u003e\n\u003ch3\u003eSmartphone Testing (Image Capture)\u003c/h3\u003e\n\u003cp\u003ePhotorefraction images were captured using the default iPhone XS telephoto lens camera (iOS 15.0; Apple Inc; Cupertino, CA; 12-megapixel resolution; focal length, 6 mm [f/2.4]). Participants were seated 1 m from the smartphone, which was mounted on a tripod with a rotatable knob to align the camera at eye level and ensure accurate orientation. All measurements were conducted under dim lighting (~\u0026thinsp;15 lux).\u003c/p\u003e \u003cp\u003eImages were obtained at 3 meridians (horizontal 180\u0026deg;, vertical 90\u0026deg;, oblique 135\u0026deg;) to assess meridional refractive powers (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), with a 3-second rest between captures. Testing was performed under non-cycloplegic (Photo\u003csub\u003eDry\u003c/sub\u003e) and cycloplegic (Photo\u003csub\u003eWet\u003c/sub\u003e) conditions when dilation was suitable. Test order (open-field autorefraction vs. smartphone photorefraction) was randomized and conducted by the same examiner.\u003c/p\u003e\n\u003ch3\u003eSmartphone Photo Analysis\u003c/h3\u003e\n\u003cp\u003eAll smartphone-captured images were extracted for computation. Refractive error (X) was determined from pupil diameter and crescent width, with myopic errors identified by crescent location using equations previously described by Chan [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eMyopic X = \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\frac{e}{d(2r-s)}+\\frac{1}{d}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003ewhere \u003cem\u003ed\u003c/em\u003e\u0026thinsp;=\u0026thinsp;distance from entrance pupil to the eye and \u003cem\u003ee\u003c/em\u003e is the eccentricity of the flashlight. Pupil diameter (\u003cem\u003e2r\u003c/em\u003e) and crescent width (\u003cem\u003es\u003c/em\u003e) were measured using ImageJ software (National Institutes of Health) by image readers masked to refraction results.\u003c/p\u003e \u003cp\u003eUnreadable images were defined as those with unclear pupil margins, indistinct crescents, or crescents obstructed by eyelids. Images with distinct pupil margins but no crescent reflex were classified as \"no crescent\". To compare photorefraction-derived refractive error with autorefraction result, the latter was converted to vector components \u003cem\u003eP\u003c/em\u003e aligned with the photo-taking meridians (horizontal, vertical, oblique) using the equation: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:P\\left(\\theta\\:\\right)=S+C\\times\\:{sin}^{2}(\\alpha\\:-\\theta\\:)\\)\u003c/span\u003e\u003c/span\u003e, where \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\theta\\:\\)\u003c/span\u003e\u003c/span\u003e is the meridian-of-interest (180\u0026deg;, 90\u0026deg;, 135\u0026deg; in radians), \u003cem\u003eS\u003c/em\u003e is the spherical power, \u003cem\u003eC\u003c/em\u003e is the cylindrical power and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\alpha\\:\\)\u003c/span\u003e\u003c/span\u003e is the cylinder axis in radians [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e].\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eData and Statistical Analysis\u003c/h2\u003e \u003cp\u003ePhotorefraction and autorefraction were first compared at the same viewing distance of 1 m under non-cycloplegic (dry) and cycloplegic (wet) conditions. Agreement between methods and test reliability were assessed using intraclass correlation coefficients (ICCs, 2-way mixed, single measures) with 95% confidence interval (CI).95% limit of agreement was estimated with Bland-Altman analyses. Mean absolute errors (MAEs) and standard errors of mean (SEM) between autorefraction and photorefraction were computed for refractive error categories (\u0026gt;\u0026thinsp;0 D, \u0026minus;\u0026thinsp;0.25 to \u0026minus;\u0026thinsp;2.00D, \u0026lt;\u0026minus;2.00 to \u0026minus;\u0026thinsp;4.00 D, \u0026lt;\u0026minus;4.00 to \u0026minus;\u0026thinsp;6.00D and \u0026lt;\u0026thinsp;\u0026minus;\u0026thinsp;6.00D) under dry and wet conditions. The performance of dry and wet smartphone photorefraction was evaluated using receiver operating characteristic (ROC) curves across myopic thresholds from \u0026minus;\u0026thinsp;1.00 to \u0026minus;\u0026thinsp;6.00 D. Sensitivity and specificity (with 95% CI) for dry and wet photorefraction were calculated using the 2021 American Association for Pediatric Ophthalmology and Strabismus (AAPOS) instrument-based referral criterion of myopia greater than 2.00 D [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. Diagnostic odd ratios (DORs) [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e], defined as \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\frac{True\\:Positive}{False\\:Negative}/\\frac{False\\:Positive}{True\\:Negatvie}\\)\u003c/span\u003e\u003c/span\u003e, with 95% CIs were computed for both conditions to assess whether smartphone photorefraction performed differently under dry versus wet conditions.\u003c/p\u003e \u003cp\u003eTo account for accommodation induced by the closer working distance of smartphone-based photorefraction, mean difference between non-cycloplegic (dry) autorefraction (Auto\u003csub\u003eDry\u003c/sub\u003e) at distance (3 m) and near (1 m), and between cycloplegic (wet) autorefraction (Auto\u003csub\u003eWet\u003c/sub\u003e) at 3 m and 1 m, were calculated. Generalized linear models with robust standard errors (HC0) were used to examine the effects of age group (adults vs. children) and refractive error category (five severity levels) on the distance\u0026ndash;near differences under dry and wet conditions.\u003c/p\u003e \u003cp\u003eMean distance-near autorefraction differences specific to age group and myopic severity under dry and wet conditions were used to derive calibration factors, which were incorporated into the corresponding photorefraction measurements. This adjustment compensated for the near working distance inherent to photorefraction and improved the accuracy of distance refractive error estimation. The calibrated photorefraction values were then compared with distance autorefraction and subjective refraction in terms of MAE, sensitivity, specificity, and DOR.\u003c/p\u003e \u003cp\u003eAnalyses were conducted using SPSS version 28 (IBM, Armonk, NY), with R version 4.4.1 (R Foundation for Statistical Computing; Vienna, Austria) used for graphical representations. Statistical significance was set at P\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003ePhotorefraction Image Reading\u003c/h2\u003e \u003cp\u003eA total of 984 participants were recruited, of whom 36 were excluded because of tropia, media opacities, or presbyopia. Among 948 Chinese participants included in the analysis (aged 6\u0026ndash;43 years; mean [standard deviation (SD)] age, 19.8 [9.03] years; 38% male; 59% adults\u0026thinsp;\u0026ge;\u0026thinsp;18 years), 5,350 non-cycloplegic eye photographs (Photo\u003csub\u003eDry\u003c/sub\u003e) were analyzed (spherical equivalent (SE) range, \u0026minus;\u0026thinsp;12.50 to +\u0026thinsp;6.88 D; mean [standard error of mean (SEM)] SE, \u0026minus;\u0026thinsp;3.50 [0.06] D). Of these, 876 participants (aged 6\u0026ndash;43 years; mean [SD] age, 20.1 [9.09] years; 38% male; 61% adults) completed cycloplegic testing (Photo\u003csub\u003eWet\u003c/sub\u003e: 4,919 images; SE range, \u0026minus;\u0026thinsp;12.13 to +\u0026thinsp;8.88 D; mean [SEM] SE, \u0026minus;\u0026thinsp;3.26 [0.06] D), with non-completion primarily due to refusal of dilation or narrow angles (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Unreadable images were excluded (6.17%). Mean refractions of subjective refraction, dry/wet photorefraction, and autorefraction were shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDetailed Summary between Dry and Wet Groups\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"8\" nameend=\"c8\" namest=\"c1\"\u003e \u003cp\u003eMean refractions of non-cycloplegic and cycloplegic measurements\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003eDry\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003eWet\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eMethod\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003eMean [SEM]; (range)\u003csup\u003ea\u003c/sup\u003e, D\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003eMean (SEM), D\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePhotorefraction\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(1 m)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e-3.61 [0.03];\u003c/p\u003e \u003cp\u003e(-12.28 to +\u0026thinsp;4.60)\u003c/p\u003e\u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e-3.04 [0.03];\u003c/p\u003e \u003cp\u003e(-12.53 to +\u0026thinsp;3.75)\u003c/p\u003e\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAutorefraction\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(1 m)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e-3.66 [0.03];\u003c/p\u003e \u003cp\u003e(-13.65 to +\u0026thinsp;6.70)\u003c/p\u003e\u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e-3.29 [0.04];\u003c/p\u003e \u003cp\u003e(-12.75 to +\u0026thinsp;8.75)\u003c/p\u003e\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAutorefraction\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003e(3 m)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e-3.50 [0.04];\u003c/p\u003e \u003cp\u003e(-13.49 to +\u0026thinsp;7.23)\u003c/p\u003e\u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e-3.26 [0.04];\u003c/p\u003e \u003cp\u003e(-12.97 to +\u0026thinsp;9.48)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSubjective\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003eRefraction (6 m)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e-3.31 [0.03];\u003c/p\u003e \u003cp\u003e(-11.99 to +\u0026thinsp;6.19)\u003c/p\u003e\u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003eNA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"8\" nameend=\"c8\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAgreement between photorefraction and autorefraction at 1 m\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eAccuracy\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e\u003cb\u003eDry\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e\u003cb\u003eWet\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003e\u003cb\u003e180-meridian\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eICC (95% CI)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.795 (0.776\u0026ndash;0.812)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.727 (0.703\u0026ndash;0.749)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.95 (0.94\u0026ndash;0.97)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.91 (0.89\u0026ndash;0.93)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.86 (0.83\u0026ndash;0.88)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.78 (0.75\u0026ndash;0.81)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePPV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.93 (0.91\u0026ndash;0.94)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.87 (0.85\u0026ndash;0.89)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.90 (0.88\u0026ndash;0.93)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.84 (0.81\u0026ndash;0.87)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003e\u003cb\u003e90-meridian\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eICC (95% CI)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.763 (0.741\u0026ndash;0.783)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.645 (0.614\u0026ndash;0.673)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.98 (0.97\u0026ndash;0.99)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.92 (0.90\u0026ndash;0.94)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.77 (0.74\u0026ndash;0.81)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.82 (0.78\u0026ndash;0.85)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePPV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.92 (0.90\u0026ndash;0.93)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.93 (0.91\u0026ndash;0.94)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.93 (0.91\u0026ndash;0.96)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.81 (0.77\u0026ndash;0.84)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003e\u003cb\u003e135-meridian\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eICC (95% CI)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.750 (0.726\u0026ndash;0.772)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.655 (0.624\u0026ndash;0.683)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.95 (0.94\u0026ndash;0.97)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.91 (0.90\u0026ndash;0.93)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.84 (0.81\u0026ndash;0.87)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.79 (0.75\u0026ndash;0.82)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePPV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.93 (0.92\u0026ndash;0.95)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.90 (0.88\u0026ndash;0.91)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.89 (0.87\u0026ndash;0.92)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.82 (0.79\u0026ndash;0.85)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"4\" rowspan=\"5\"\u003e \u003cp\u003e\u003cb\u003eAll meridians\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eICC (95% CI)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.771 (0.759\u0026ndash;0.783)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.670 (0.654\u0026ndash;0.686)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.96 (0.96\u0026ndash;0.97)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.91 (0.90\u0026ndash;0.92)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.83 (0.81\u0026ndash;0.85)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.79 (0.77\u0026ndash;0.81)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePPV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.93 (0.92\u0026ndash;0.93)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.90 (0.89\u0026ndash;0.91)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e0.91 (0.89\u0026ndash;0.92)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e0.82 (0.80\u0026ndash;0.84)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"8\" nameend=\"c8\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eComparison of photorefraction and near autorefraction at 1 m\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003e\u003cb\u003eDry\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003e\u003cb\u003eWet\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eRefractive Error Range\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003eNo. Gradable Photos\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003eMAE [SEM], D\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003ewithin \u0026plusmn;\u0026thinsp;1.00 D, %\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003eNo. Gradable Photos\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cb\u003eMAE [SEM], D\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e\u003cb\u003ewithin \u0026plusmn;\u0026thinsp;1.00 D, %\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;0 D\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e154\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.21 [0.08]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e56\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e418\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.52 [0.05]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e40\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e-0.25 to -2.00 D\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e725\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.77 [0.02]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e873\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.93 [0.03]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e65\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026lt; -2.00\u0026nbsp;to -4.00 D\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1548\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.59 [0.01]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1579\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.73 [0.01]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e75\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026lt; -4.00 to -6.00 D\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1258\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.05 [0.02]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1056\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.40 [0.02]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e37\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u0026lt; -6.00 D\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e869\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2.30 [0.04]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e630\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2.96 [0.06]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003eOverall at 1 m\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4554\u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.09 [0.02]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4556\u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.30 [0.02]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e53\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003eAbbreviation: D, dioptres, ICC, Intraclass Correlation Coefficient; MAE, Mean Absolute Error; PPV, Positive Predictive Value; NPV, Negative Predictive Value; No. Number; SEM, Standard Error of Mean\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003csup\u003ea\u003c/sup\u003eMean, standard error of mean (SEM) and ranges were calculated based on all meridional refractive errors at 180-, 90- and 135- meridians.\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003csup\u003eb\u003c/sup\u003e Meridional refractive error for near autorefractions ranged from \u0026minus;\u0026thinsp;13.65 to +\u0026thinsp;6.70 D (Dry) and \u0026minus;\u0026thinsp;12.75 to +\u0026thinsp;8.75 D (Wet) respectively.\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003csup\u003ec\u003c/sup\u003e Photos graded as \u0026ldquo;no-crescent\u0026rdquo; were excluded in this analysis.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eComparison of Photorefraction with Autorefraction Readings\u003c/b\u003e \u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003ePhotorefraction results were compared with corresponding autorefraction measurements for each of the three meridians at the same viewing distance of 1 m. Each meridian showed good agreement with autorefraction under both non-cycloplegic (dry) and cycloplegic (wet) conditions (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). ICC values for all meridians were 0.77 (95% CI, 0.76\u0026ndash;0.78) under dry conditions and 0.67 (95% CI, 0.65\u0026ndash;0.69) under wet conditions. MAE [SEM] between autorefraction and photorefraction was smallest when refractive error range fell within \u0026lt;\u0026thinsp;\u0026minus;\u0026thinsp;2.00 to \u0026minus;\u0026thinsp;4.00 D, 0.59 [0.01] and 0.73 [0.01] in dry and wet conditions, respectively (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBland-Altman plots demonstrated that most differences between methods lay within the 95% limits of agreement in both conditions (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Across all meridians, the area under the receiver operating characteristic curve (AUC) exceeded 0.90 under dry conditions and 0.84 under wet conditions for myopia thresholds from \u0026minus;\u0026thinsp;1.00 D to \u0026minus;\u0026thinsp;6.00 D, indicating good discriminative performance (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). When analysed according to the 2021 AAPOS instrument-based referral criterion (myopia more than \u0026minus;\u0026thinsp;2.00 D), dry photorefraction showed slightly higher sensitivity and specificity (96% (95% CI, 96\u0026ndash;97) and 83% (95%CI, 81\u0026ndash;85), respectively) than wet photorefraction (91% (95% CI, 90\u0026ndash;92) and 79% (95% CI, 77\u0026ndash;81), respectively), with AUCs of 0.97 (95%CI, 0.96\u0026ndash;0.97) (dry) and 0.91 (95% CI, 0.90\u0026ndash;0.92) (wet). Dry photorefraction had a higher diagnostic odd ratio (DOR) of 123.1 (95% CI, 99.5-152.2) than wet photorefraction (41.2; 95% CI, 34.7\u0026ndash;48.9).\u003c/p\u003e \u003cp\u003e \u003col start=2\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eComparison of Accommodation Effect on Smartphone-based Photorefraction\u003c/b\u003e \u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eThe mean [SEM] distance\u0026ndash;near autorefraction difference was 0.15 [0.01] D under dry conditions and 0.03 [0.01] D under wet conditions (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Under dry conditions, there was no significant effect of age group (P\u0026thinsp;\u0026gt;\u0026thinsp;0.05), whereas refractive error category showed a significant effect (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05), with hyperopic participants demonstrating greater accommodation in both adults and children (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Under wet conditions, neither age group (P\u0026thinsp;\u0026gt;\u0026thinsp;0.05) nor myopic refractive error category (P\u0026thinsp;\u0026gt;\u0026thinsp;0.05) showed significant effects on the distance-near accommodation, and autorefraction differences across refractive categories and between age groups were not clinically significant.\u003c/p\u003e \u003cp\u003e \u003col start=3\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003e \u003cb\u003eComparison between Photorefraction, Distance Autorefraction and Subjective Refraction\u003c/b\u003e \u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eBased on the distance\u0026ndash;near autorefraction differences, calibration factors accounting for photorefraction working distance and stratified by age group and myopia severity were derived and applied to both dry and wet photorefraction measurements. The performance of photorefraction compared with conventional distance autorefraction is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e (dry: sensitivity 94% (95%CI, 93\u0026ndash;95), specificity 85% (95% CI, 83\u0026ndash;87); wet: sensitivity 91% (95% CI, 90\u0026ndash;92), specificity 80% (95%CI, 78\u0026ndash;82)), demonstrating consimilar performance across age groups under both conditions.\u003c/p\u003e \u003cp\u003eTo align with clinical standards, calibrated non-cycloplegic photorefraction measurements were compared with subjective refraction. Photorefraction demonstrated strong diagnostic accuracy for detecting myopia, with a sensitivity of 96% (95% CI, 95\u0026ndash;96), specificity of 82% (95% CI, 81\u0026ndash;84), DOR of 102.1 (95% CI, 83.5-124.9) and an AUC of 0.96 (95% CI, 0.96\u0026ndash;0.97) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). MAE [SEM] was 0.79 [0.01] D for myopia fell between \u0026minus;\u0026thinsp;0.25 to \u0026minus;\u0026thinsp;6.00 D. When benchmarked against cycloplegic autorefraction, calibrated dry photorefraction maintained the robust performance for screening significant myopia (sensitivity: 95% (95% CI, 95\u0026ndash;96); specificity: 81% (95% CI, 79\u0026ndash;83); DOR: 89.0 (95% CI, 72.6\u0026ndash;109.0); AUC: 0.96 (95% CI, 0.95\u0026ndash;0.96)), with an MAE [SEM] of 0.83 [0.01] D within \u0026minus;\u0026thinsp;0.25 to \u0026minus;\u0026thinsp;6.00 D of myopia.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study represents a large-scale, first-in-class diagnostic evaluation of a standalone smartphone-based platform that uses only its built-in camera and flashlight to detect refractive errors, benchmarked against clinical standards under both non-cycloplegic and cycloplegic conditions. The baseline STARS model validated the core eccentric photorefraction principles, providing a robust foundation for subsequent integration of AI-based image analysis. Non-cycloplegic photorefraction demonstrated superior performance, achieving 96% sensitivity and 83% specificity for detecting significant myopia, following AAPOS 2021 guidelines, outperforming cycloplegic readings (sensitivity 91%, specificity 79%). Multiple myopia thresholds were evaluated, with corresponding AUCs values exceeding 0.90 across low to high myopia ranges, indicating that STARS can not only screen for myopia but also estimate the magnitude of refractive error with high accuracy. Accommodative effort was comparable between myopic children and adults across conditions, supporting the method\u0026rsquo;s utility for vision screening across age groups and settings. Previous research has largely focused on comparing non-cycloplegic photoscreening with cycloplegic refraction [\u003cspan additionalcitationids=\"CR27 CR28 CR29\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e], and relatively few studies [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e] have directly contrasted non-cycloplegic and cycloplegic photoscreening against a cycloplegic standard. In this context, the present findings, demonstrating better performance of smartphone photoscreening without eyedrops, support that non-cycloplegic STARS photoscreening performs consistently and accurately relative to cycloplegic autorefraction, reinforcing the feasibility of this standalone platform without the need for cycloplegia.\u003c/p\u003e \u003cp\u003eBeyond its internal validation metrics, it is important to contextualize the performance of STARS against existing photoscreening technologies. The performance of STARS is comparable to that of commercially available photoscreeners in terms of mean difference and screening accuracy. For example, the 2WIN (Adaptica SRL, Padova, Italy) demonstrated a mean difference of 0.60 D compared with an autorefractor in 44 undilated myopic participants [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. The Spot Vision Screener (Welch Allyn Inc, Skaneateles Falls, NY) has reported sensitivity and specificity values of 60\u0026ndash;96% and 94\u0026ndash;99%, respectively, for screening significant myopia greater than \u0026minus;\u0026thinsp;2.00 D without cycloplegia [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Yan and colleagues [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e] reported a myopic screening performance of 46/99% (sensitivity/specificity) by PlusoptiX A12C (PlusoptiX GmbH, Nuremberg, Germany) in a pediatric population. GoCheck Kids (Goquity Mobile Health), a smartphone-based photoscreener using photorefraction with an auxiliary attachment, has reported sensitivity and specificity of 52\u0026ndash;91% and 68\u0026ndash;90% for detecting amblyopic risk factors [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan additionalcitationids=\"CR37\" citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. For detecting significant myopia according to 2021 AAPOS guideline, its sensitivity and specificity were 100 and 79.1%, respectively, although the PPV was 38% [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e], which is lower than the PPV of 93% observed in STARS models for predicting myopia. In addition, Luo et al. developed a smartphone application tested in 201 myopic eyes with refractive errors up to \u0026minus;\u0026thinsp;10.00 D, yielding a MAE (SD) of 0.65 (0.83) D, relying on active patient feedback [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. Taken together, these findings suggest that the STARS model provides an accurate and sensitive objective refraction method based on portable and accessible technology that does not require additional hardware accessories or patient response.\u003c/p\u003e \u003cp\u003eGiven smartphones have gained widespread adoption especially over the past decades where coverage has extended to even less developed countries [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e], the application value of our investigated smartphone system, with its validated agreement in detecting significant refractive error under non-cycloplegic conditions, is generalizable to typical screening environments where diagnostic agents may be unavailable or constrained to administer. This also highlights the public health relevance of our novel approach. Even though screening accuracy for myopia was similarly high in both dry and wet conditions, as reflected by their MAEs against conventional standards, the dry photorefraction screening framework ultimately offers a methodology that is more practical and cost-effective. This advantageous strategy reduces the need for professional manpower during vision screening compared to wet photorefraction. In addition, as we disseminated the specific concerns of research participants, we estimate that approximately one in ten more parents would be willing to have their child undergo preliminary vision screening if no eye drops were required, based on the proportion of participants who opted out of wet photorefraction during data collection. Our study also assessed a diverse age range where a similarly good performance was observed between the two age groups. Moreover, the applicability of our model is expected to apply relevantly to East and South Asian ethnic groups who are disproportionally affected by the myopia epidemic, with some of the highest prevalence rates globally over the past few decades [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Pupil diameter is often more difficult to estimate in darkly coloured irises; however, the smart device was able to keep up to this need.\u003c/p\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eLimitations\u003c/h2\u003e \u003cp\u003eThis study did not directly compare the smartphone-based method with existing portable photoscreeners, which are not universally available internationally and are primarily designed for preliminary detection rather than serving as clinical standards. Potential pseudo-myopia arising from the near working distance was addressed through calibration; although the refractive estimates were statistically greater under dry than wet conditions, the mean differences of 0.03\u0026ndash;0.15 D were clinically negligible. In addition, further studies and validation are needed in broader populations, including younger children (\u0026lt;\u0026thinsp;6 years) and individuals with hyperopia. The controlled clinic-based design supported the underlying optical principles but did not evaluate real-world community deployment, in which environmental variability may affect measurement performance. Given the refractive error profile and age distribution of the participants, the model shows strong potential as a low-cost, cycloplegia-free option for large-scale myopia screening. Future studies should evaluate usability, scalability, and health economic impact in diverse settings and primary care settings.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eImplications for Practice\u003c/h2\u003e \u003cp\u003eGlobal organizations, including the World Health Organization, have deployed mobile applications for visual acuity screening [\u003cspan additionalcitationids=\"CR43 CR44\" citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e] that typically rely on self-reported responses. Integrating objective refractive-error assessment with visual acuity testing would better align community screening with clinical standards and improve diagnostic structure, particularly in settings lacking traditional equipment. External validation across diverse populations and field environments is a critical next step to broaden access to reliable vision screening in underserved areas.\u003c/p\u003e \u003cp\u003eFrom a feasibility perspective, the STARS framework offers a promising, scalable approach for community-level implementation and could serve as a cost‑effective alternative for large-scale deployment. Given the projected public‑health burden of myopia, prioritizing on‑field evaluation and implementation research will be essential to determine real‑world effectiveness, acceptability, and health‑economic impact.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusions","content":"\u003cp\u003eA standalone smartphone-based photorefraction system shows promise as a scalable, translational solution for vision screening in under-resourced communities. Images captured by the device contain sufficient information to accurately quantify refractive errors, particularly myopia, with performance comparable to conventional clinical techniques. These findings support its potential as a cost-effective screening tool to detect visually significant myopia and facilitate timely management and intervention.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAAPOS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAmerican Association for Pediatric Ophthalmology and Strabismus\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAUC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eArea under curve\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDOR\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eDiagnostic odd ratio\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDry\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eNon-cycloplegic condition\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eConfidence interval\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eD\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eDiopter\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eICC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eIntraclass correlation coefficients\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eMAE\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eMean absolute error\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePPV\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003ePositive Predictive Value\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eNPV\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eNegative Predictive Value\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eROC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eReceiver Operating Characteristic\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSD\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eStandard deviation\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSE\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eSpherical equivalent\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSEM\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eStandard errors of mean\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eWet\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eCycloplegic condition\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":" \u003cp\u003e \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003eEthnics approval and consent to participate\u003c/span\u003e \u003c/p\u003e \u003cp\u003eThis diagnostic study adhered to the Declaration of Helsinki, was approved by the university's Human Subjects Ethics Sub-Committee (HSEARS #20210401006), and followed the Standards for Reporting of Diagnostic Accuracy Studies (STARD) reporting guideline. Written informed consent was obtained from all participants or their legal guardians (for those aged\u0026thinsp;\u0026lt;\u0026thinsp;18 years).\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent for publication\u003c/strong\u003e \u003cp\u003eNot applicable. The manuscript does not contain personal data (smartphone photorefraction photos), only refractive errors analysed and calculated from the photos.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThis work was supported by Health Medical Research Fund (18191351), Innovation and Technology Fund (ITS/049/23), and Hong Kong PhD Fellowship Scheme (PF23-93959) by the Hong Kong Government, the Hong Kong Polytechnic University (Grants no.: 1-WZ1B, 1-WZ0L, 1-BD50), and the Tianjin Key Medical Discipline Construction Project (TJYXZDXK-3-004A-2).\u003c/p\u003e\u003ch2\u003eAuthors\u0026rsquo; contributions\u003c/h2\u003e \u003cp\u003eCWD, LC, GN, EF, and HVL contributed to the funding acquisition, conception and design of the study. CWD, GN, BD, and RW contributed to project administration and supervision. LYK, OLK, and HCL contributed to data acquisition. WYT and YYW did the image analysis. PL contributed to the data analysis plan. LYK did the data analysis. LC and LYK drafted the original manuscript. CWD reviewed and edited the manuscript. PL accessed and verified the raw data of the study. All authors read and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e \u003cp\u003eWe thank Dr. Wing Tang, Miss Lena Li, Mr. Kenny Chung and Miss Lotus Lai for their valuable contributions during the early stages of the conceptual work in preliminary testing and validation.\u003c/p\u003e\u003ch2\u003eAvailability of data and materials\u003c/h2\u003e \u003cp\u003eThe datasets generated and/or analysed during the current study are not publicly available due to an AI algorithm in development but are available from the corresponding author on reasonable request.\u003c/p\u003e \u003ch2\u003eCompeting interests\u003c/h2\u003e \u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eTarczy-Hornoch K, Cotter SA, Borchert M, et al. Prevalence and causes of visual impairment in Asian and non-Hispanic white preschool children: Multi-ethnic Pediatric Eye Disease Study. Ophthalmology. 2013;120(6):1220\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ophtha.2012.12.029\u003c/span\u003e\u003cspan address=\"10.1016/j.ophtha.2012.12.029\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWebber AL, Wood J. Amblyopia: prevalence, natural history, functional effects and treatment. Clin Exp Optom. 2005;88(6):365\u0026ndash;75. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/j.1444-0938.2005.tb05102.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1444-0938.2005.tb05102.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZaabaar E, Asiamah R, Kyei S, Ankamah S. Myopia control strategies: A systematic review and meta-meta-analysis. Ophthalmic Physiol Opt. 2025;45(1):160\u0026ndash;76. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/opo.13417\u003c/span\u003e\u003cspan address=\"10.1111/opo.13417\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDo CW, Chan LYL, Tse ACY, et al. Association between Time Spent on Smart Devices and Change in Refractive Error: A 1-Year Prospective Observational Study among Hong Kong Children and Adolescents. Int J Environ Res Public Health. 2020;17(23):8923. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/ijerph17238923\u003c/span\u003e\u003cspan address=\"10.3390/ijerph17238923\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang XJ, Zhang Y, Kam KW, et al. Prevalence of Myopia in Children Before, During, and After COVID-19 Restrictions in Hong Kong. JAMA Netw Open. 2023;6(3):e234080. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jamanetworkopen.2023.4080\u003c/span\u003e\u003cspan address=\"10.1001/jamanetworkopen.2023.4080\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHolden BA, Fricke TR, Wilson DA, et al. Ophthalmology. 2016;123(5):1036\u0026ndash;42. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ophtha.2016.01.006\u003c/span\u003e\u003cspan address=\"10.1016/j.ophtha.2016.01.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Global Prevalence of Myopia and High Myopia and Temporal Trends from 2000 through 2050.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiang J, Pu Y, Chen J, et al. Global prevalence, trend and projection of myopia in children and adolescents from 1990 to 2050: a comprehensive systematic review and meta-analysis. Br J Ophthalmol. 2025;109(3):362\u0026ndash;71. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1136/bjo-2024-325427\u003c/span\u003e\u003cspan address=\"10.1136/bjo-2024-325427\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu L, Wang Y, Wang S, Wang Y, Jonas JB. High myopia and glaucoma susceptibility the Beijing Eye Study. Ophthalmology. 2007;114(2):216\u0026ndash;20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ophtha.2006.06.050\u003c/span\u003e\u003cspan address=\"10.1016/j.ophtha.2006.06.050\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHa A, Kim CY, Shim SR, Chang IB, Kim YK. Degree of Myopia and Glaucoma Risk: A Dose-Response Meta-analysis. Am J Ophthalmol. 2022;236:107\u0026ndash;19. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ajo.2021.10.007\u003c/span\u003e\u003cspan address=\"10.1016/j.ajo.2021.10.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaba T, Ohno-Matsui K, Futagami S, et al. Prevalence and characteristics of foveal retinal detachment without macular hole in high myopia. Am J Ophthalmol. 2003;135(3):338\u0026ndash;42. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/s0002-9394(02)01937-2\u003c/span\u003e\u003cspan address=\"10.1016/s0002-9394(02)01937-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWatts P, Walker K, Beck L. Photoscreening for refractive errors in children and young adults with severe learning disabilities using the MTI photoscreener. Eye (Lond). 1999;13(3):363\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/eye.1999.92\u003c/span\u003e\u003cspan address=\"10.1038/eye.1999.92\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoghaddam AA, Kargozar A, Zarei-Ghanavati M, Najjaran M, Nozari V, Shakeri MT. Screening for amblyopia risk factors in pre-verbal children using the Plusoptix photoscreener: a cross-sectional population-based study. Br J Ophthalmol. 2012;96(1):83\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1136/bjo.2010.190405\u003c/span\u003e\u003cspan address=\"10.1136/bjo.2010.190405\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeterseim MMW, Trivedi RH, Monahan SR et al. Effectiveness of the Spot Vision Screener using updated 2021 AAPOS guidelines. J AAPOS. 2023, 27(1): 24.e1\u0026ndash;24.e7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jaapos.2022.11.019\u003c/span\u003e\u003cspan address=\"10.1016/j.jaapos.2022.11.019\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBobier WR, Braddick OJ. Eccentric photorefraction: optical analysis and empirical measures. Am J Optom Physiol Opt. 1985;62(9):614\u0026ndash;20.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWorld Health Organization. Report of the 2030 targets on effective coverage of eye care. World Health Organization; 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLatif MZ, Hussain I, Afzal S, et al. Impact of Refractive Errors on the Academic Performance of High School Children of Lahore. Front Public Health. 2022;10:869294. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fpubh.2022.869294\u003c/span\u003e\u003cspan address=\"10.3389/fpubh.2022.869294\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWong HB, Machin D, Tan SB, Wong TY, Saw SM. Visual impairment and its impact on health-related quality of life in adolescents. Am J Ophthalmol. 2009;147(3):505\u0026ndash;e5111. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ajo.2008.09.025\u003c/span\u003e\u003cspan address=\"10.1016/j.ajo.2008.09.025\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi D, Chan VF, Virgili G, et al. Impact of Vision Impairment and Ocular Morbidity and Their Treatment on Depression and Anxiety in Children: A Systematic Review. Ophthalmology. 2022;129(10):1152\u0026ndash;70. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ophtha.2022.05.020\u003c/span\u003e\u003cspan address=\"10.1016/j.ophtha.2022.05.020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArnold RW, Arnold AW, Hunt-Smith TT, Grendahl RL, Winkle RK. The Positive Predictive Value of Smartphone Photoscreening in Pediatric Practices. J Pediatr Ophthalmol Strabismus. 2018;55(6):393\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3928/01913913-20180710-01\u003c/span\u003e\u003cspan address=\"10.3928/01913913-20180710-01\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDebert I, Costa DRD, Polati M, Falabretti JG, Susanna Junior R. Vision screening using a smartphone platform. Rev Paul Pediatr. 2022;40:e2020021. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1590/1984-0462/2022/40/2020021IN\u003c/span\u003e\u003cspan address=\"10.1590/1984-0462/2022/40/2020021IN\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChan OY. Hong Kong Polytechnic. Dept. of Optometry and Radiography. Vision Screening in Pre-School Children. [dissertation]. The Hong Kong Polytechnic University: HKP Dept. of Optometry and Radiography; 1994.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChan OY, Edwards M, Brown B. Calibration and validity of an eccentric photorefractor. Ophthalmic Physiol Opt. 1996;16(3):203\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNaeser K, Behrens JK. Correlation between polar values and vector analysis. J Cataract Refract Surg. 1997;23(1):76\u0026ndash;81. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/s0886-3350(97)80154-x\u003c/span\u003e\u003cspan address=\"10.1016/s0886-3350(97)80154-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArnold RW, Donahue SP, Silbert DI et al. AAPOS uniform guidelines for instrument-based pediatric vision screen validation 2021. J AAPOS. 2022, 26(1): 1.e1\u0026ndash;1.e6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jaapos.2021.09.009\u003c/span\u003e\u003cspan address=\"10.1016/j.jaapos.2021.09.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGlas AS, Lijmer JG, Prins MH, Bonsel GJ, Bossuyt PM. The diagnostic odds ratio: a single indicator of test performance. J Clin Epidemiol. 2003;56(11):1129\u0026ndash;35. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/s0895-4356(03)00177-x\u003c/span\u003e\u003cspan address=\"10.1016/s0895-4356(03)00177-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBloomberg JD, Suh DW. The accuracy of the plusoptiX A08 photoscreener in detecting risk factors for amblyopia in central Iowa. J AAPOS. 2013;17(3):301\u0026ndash;4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jaapos.2013.03.014\u003c/span\u003e\u003cspan address=\"10.1016/j.jaapos.2013.03.014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeterseim MM, Papa CE, Wilson ME, et al. Photoscreeners in the pediatric eye office: compared testability and refractions on high-risk children. Am J Ophthalmol. 2014;158(5):932\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ajo.2014.07.041\u003c/span\u003e\u003cspan address=\"10.1016/j.ajo.2014.07.041\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYan XR, Jiao WZ, Li ZW, Xu WW, Li FJ, Wang LH. Performance of the Plusoptix A09 photoscreener in detecting amblyopia risk factors in Chinese children attending an eye clinic. PLoS ONE. 2015;10(6):e0126052. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pone.0126052\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0126052\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang D, Chen X, Zhang X et al. Pediatric vision screening using the plusoptiX A12C photoscreener in Chinese preschool children aged 3 to 4 years. Sci Rep. 2017, 7(1): 2041. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41598-017-02246-6\u003c/span\u003e\u003cspan address=\"10.1038/s41598-017-02246-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVil\u0026agrave;-de Muga M, Van Esso D, Alarcon S, et al. Instrument-based screening for amblyopia risk factors in a primary care setting in children aged 18 to 30 months. Eur J Pediatr. 2021;180(5):1521\u0026ndash;7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00431-020-03904-0\u003c/span\u003e\u003cspan address=\"10.1007/s00431-020-03904-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbrahamsson M, Ohlsson J, Bj\u0026ouml;rndahl M, Abrahamsson H. Clinical evaluation of an eccentric infrared photorefractor: the PowerRefractor. Acta Ophthalmol Scand. 2003;81(6):605\u0026ndash;10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1046/j.1600-0420.2003.0149.x\u003c/span\u003e\u003cspan address=\"10.1046/j.1600-0420.2003.0149.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAyse YK, Onder U, Suheyla K. Accuracy of Plusoptix S04 in children and teens. Can J Ophthalmol. 2011;46(2):153\u0026ndash;7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3129/i10-110\u003c/span\u003e\u003cspan address=\"10.3129/i10-110\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKanclerz P, Przewł\u0026oacute;cka K, Arnold RW. Agreement in non-cycloplegic and cycloplegic refraction between a photoscreener and a calibrated autorefractor. BMC Ophthalmol. 2024;24(1):130. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12886-024-03375-z\u003c/span\u003e\u003cspan address=\"10.1186/s12886-024-03375-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMudie LI, Pickett K, Ross K, McCourt E, Enzenauer R. Performance of the Spot Vision Screener in children with Down syndrome and other special needs. J AAPOS. 2023, 27(5): 274.e1-274.e7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jaapos.2023.07.011\u003c/span\u003e\u003cspan address=\"10.1016/j.jaapos.2023.07.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYan Q, Li R, Qian Y, et al. Instrument referral criteria for PlusoptiX and SureSight based on 2021 AAPOS guidelines: A population-based study. Front Public Health. 2022;10:959757. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fpubh.2022.959757\u003c/span\u003e\u003cspan address=\"10.3389/fpubh.2022.959757\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWalker M, Duvall A, Daniels M et al. Effectiveness of the iPhone GoCheck Kids smartphone vision screener in detecting amblyopia risk factors. J AAPOS. 2020, 24(1): 16.e1\u0026ndash;16.e5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jaapos.2019.10.007\u003c/span\u003e\u003cspan address=\"10.1016/j.jaapos.2019.10.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOtto H, Deschoemaeker M, Van Overmeire B, Casteels I, Cassiman C. Validation of the eye screening tool GoCheck Kids for the detection of amblyopia risk factors in toddlers in Flanders. J AAPOS. 2024;28(5):104008. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jaapos.2024.104008\u003c/span\u003e\u003cspan address=\"10.1016/j.jaapos.2024.104008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eApplebaum SS, Sopeyin A, Mohamedali A, et al. Comparison of the GoCheck Kids and Spot Screener photoscreening devices for the detection of amblyopia risk factors using 2021 AAPOS recommendations. J AAPOS. 2024;28(6):104035. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jaapos.2024.104035\u003c/span\u003e\u003cspan address=\"10.1016/j.jaapos.2024.104035\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuo G, Lee CY, Shivshanker P, et al. Preliminary Evaluation of a Smartphone App for Refractive Error Measurement. Transl Vis Sci Technol. 2022;11(2):40. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1167/tvst.11.2.40\u003c/span\u003e\u003cspan address=\"10.1167/tvst.11.2.40\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoffmann IM, Andersen AM, Lund S, Nygaard U, Joshua D, Poulsen A. Smartphone apps hold promise for neonatal emergency care in low-resource settings. Acta Paediatr. 2024;113(12):2526\u0026ndash;33. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/apa.17410\u003c/span\u003e\u003cspan address=\"10.1111/apa.17410\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePoulsen A, Hickie IB, Alam M, et al. Access to mobile health in lower middle-income countries: a review. Health Technol. 2025;15:333\u0026ndash;43. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12553-025-00948-w\u003c/span\u003e\u003cspan address=\"10.1007/s12553-025-00948-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBastawrous A, Rono HK, Livingstone IAT, et al. Development and Validation of a Smartphone-Based Visual Acuity Test (Peek Acuity) for Clinical Practice and Community-Based Fieldwork. JAMA Ophthalmol. 2015;133(8):930\u0026ndash;7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1001/jamaophthalmol.2015.1468\u003c/span\u003e\u003cspan address=\"10.1001/jamaophthalmol.2015.1468\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWorld Health Organization. World report on vision. Geneva: World Health Organization. 2019. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.who.int/publications/i/item/9789241516570\u003c/span\u003e\u003cspan address=\"https://www.who.int/publications/i/item/9789241516570\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbdul Rahman SNA, Naing NN, Othman AM, et al. Validity and Reliability of Vis-Screen Application: A Smartphone-Based Distance Vision Testing for Visual Impairment and Blindness Vision Screening. Med (Kaunas). 2023;59(5):912. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/medicina59050912\u003c/span\u003e\u003cspan address=\"10.3390/medicina59050912\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAman S, Ahmad NU, Sadiq MAA, et al. Implementation outcomes of a school-based visual screening program using the peek tool in low-resource settings in Pakistan. BMC Public Health. 2025;25(1):4120. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s12889-025-25112-x\u003c/span\u003e\u003cspan address=\"10.1186/s12889-025-25112-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"journal-of-translational-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jtrm","sideBox":"Learn more about [Journal of Translational Medicine](http://translational-medicine.biomedcentral.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/jtrm/default.aspx","title":"Journal of Translational Medicine","twitterHandle":"@BioMedCentral","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Myopia, Teleophthalmology, Smartphone Refraction, Photoscreening","lastPublishedDoi":"10.21203/rs.3.rs-8873852/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8873852/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eRecent smartphone-based health technologies emphasize engineering, but evidence on standalone devices, without attachments, for detecting uncorrected refractive errors remains limited. Validation is essential for scalable vision screening in resource-limited settings. This study was conducted to validate smartphone-based eccentric photorefraction against open-field autorefraction and assess comparable performance under non-cycloplegic (dry) and cycloplegic (wet) conditions.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eThis diagnostic study (2022\u0026ndash;2024), baseline for the Smartphone AI Refraction System (STARS), enrolled 948 Chinese children and adults (aged 6\u0026ndash;43 years; mean [SD], 19.8 [9.03] years; spherical equivalent, \u0026minus;\u0026thinsp;12.50 to +\u0026thinsp;6.88 D) with clear media, excluding presbyopia, tropia, or nystagmus. This study was conducted at the Hong Kong Polytechnic University Optometry Research Clinic using convenience sampling. Participants underwent open-field autorefraction and smartphone photorefraction at 1 m distance under non-cycloplegic and cycloplegic conditions. We evaluated the refractive error agreement with intraclass correlation coefficient (ICC) and Bland-Altman analysis. Diagnostic performance for myopia worse than \u0026minus;\u0026thinsp;2.00 D (2021 AAPOS criteria) was assessed using sensitivity, specificity, and diagnostic odds ratio (DOR).\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eWe analysed 5,350 non-cycloplegic and 4,919 cycloplegic eye photographs. Non-cycloplegic photorefraction demonstrated higher sensitivity (96%; 95% CI, 96\u0026ndash;97) and specificity (83%; 95% CI, 81\u0026ndash;85) for detecting significant myopia compared with cycloplegic photorefraction (91% (95% CI, 90\u0026ndash;92); 79% (95% CI, 77\u0026ndash;81)). ICCs were 0.77 (non-cycloplegic) and 0.67 (cycloplegic). Non-cycloplegic photorefraction also showed higher DOR than cycloplegic photorefraction.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eStandalone smartphone photorefraction without external attachments offers high accuracy, particularly under noncycloplegic conditions, and supports large-scale refractive errors screening in underserved areas.\u003c/p\u003e","manuscriptTitle":"Standalone Smartphone Photorefraction: High Accuracy for Timely Myopia Screening","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-27 13:20:42","doi":"10.21203/rs.3.rs-8873852/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2026-02-24T16:47:06+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-24T16:30:55+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-17T04:14:33+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Translational Medicine","date":"2026-02-13T11:19:26+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-translational-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jtrm","sideBox":"Learn more about [Journal of Translational Medicine](http://translational-medicine.biomedcentral.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/jtrm/default.aspx","title":"Journal of Translational Medicine","twitterHandle":"@BioMedCentral","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"fd0a8032-fd0c-4081-8ff5-5a86ffcbef24","owner":[],"postedDate":"February 27th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-10T17:22:45+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-27 13:20:42","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8873852","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8873852","identity":"rs-8873852","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}