{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,9]],"date-time":"2026-03-09T07:13:19Z","timestamp":1773040399311,"version":"3.50.1"},"reference-count":20,"publisher":"MDPI AG","issue":"2","license":[{"start":{"date-parts":[[2026,1,20]],"date-time":"2026-01-20T00:00:00Z","timestamp":1768867200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background\/Objectives: Subjective refraction is the clinical gold standard but is time-consuming and examiner-dependent. Most artificial intelligence (AI)-based approaches rely on specialized imaging or biometric data not routinely available. This study aimed to predict subjective refraction using only routine, non-cycloplegic autorefraction and keratometric data and to identify factors associated with reduced prediction accuracy. Methods: This retrospective study included 1856 eyes from 1006 patients. A multi-output histogram gradient-boosting model predicted subjective spherical equivalent, cylindrical power, and astigmatic axis. Performance was evaluated on an independent test dataset using R2 and mean absolute error, with circular statistics for axis prediction. Prediction failure was assessed using clinically relevant tolerance thresholds (sphere\/cylinder \u2264 0.50 D; axis \u2264 10\u00b0) and multivariable logistic regression. Results: The model achieved high accuracy for spherical and cylindrical prediction (R2 = 0.987 and 0.933; MAE = 0.126 D and 0.137 D). Astigmatic axis prediction demonstrated strong circular agreement (\u03c1 = 0.898), with a mean absolute angular error of 4.65\u00b0 (median, 0.96\u00b0). Axis errors were higher in eyes with low cylinder magnitude (&lt;0.75 D) and oblique astigmatism. In multivariable analysis, steeper keratometry (K2; OR = 7.25, 95% CI 1.62\u201332.46, p = 0.010) and greater objective cylindrical power (OR = 2.79, 95% CI 1.87\u20138.94, p = 0.032) were independently associated with poor prediction. Conclusions: A machine-learning model based solely on routine, non-cycloplegic autorefractor and keratometric measurements can accurately estimate subjective refraction, supporting AI as a complementary decision-support tool rather than a replacement for conventional subjective refraction.<\/jats:p>","DOI":"10.3390\/diagnostics16020331","type":"journal-article","created":{"date-parts":[[2026,1,20]],"date-time":"2026-01-20T09:24:34Z","timestamp":1768901074000},"page":"331","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Artificial Intelligence\u2013Based Prediction of Subjective Refraction and Clinical Determinants of Prediction Error"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7967-0899","authenticated-orcid":false,"given":"Ozlem","family":"Candan","sequence":"first","affiliation":[{"name":"Department of Ophthalmology, Ankara Training and Research Hospital, University of Health Sciences, 06340 Ankara, T\u00fcrkiye"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Irem","family":"Saglam","sequence":"additional","affiliation":[{"name":"Department of Ophthalmology, Ankara Training and Research Hospital, University of Health Sciences, 06340 Ankara, T\u00fcrkiye"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gozde","family":"Orman","sequence":"additional","affiliation":[{"name":"Department of Ophthalmology, Ankara Training and Research Hospital, University of Health Sciences, 06340 Ankara, T\u00fcrkiye"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Nurten","family":"Unlu","sequence":"additional","affiliation":[{"name":"Department of Ophthalmology, Ankara Training and Research Hospital, University of Health Sciences, 06340 Ankara, T\u00fcrkiye"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ay\u015fe","family":"Burcu","sequence":"additional","affiliation":[{"name":"Department of Ophthalmology, Ankara Training and Research Hospital, University of Health Sciences, 06340 Ankara, T\u00fcrkiye"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yusuf","family":"Candan","sequence":"additional","affiliation":[{"name":"Energy Market Regulation Authority, 06510 Ankara, T\u00fcrkiye"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,1,20]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"1036","DOI":"10.1016\/j.ophtha.2016.01.006","article-title":"Global Prevalence of Myopia and High Myopia and Temporal Trends from 2000 through 2050","volume":"123","author":"Holden","year":"2016","journal-title":"Ophthalmology"},{"key":"ref_2","unstructured":"Elliott, D.B. (2009). Subjective refraction. Optometry: Science, Techniques and Clinical Management, Elsevier. [2nd ed.]."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"228","DOI":"10.1097\/ICU.0000000000000834","article-title":"Beyond traditional subjective refraction","volume":"33","author":"Dorronsoro","year":"2022","journal-title":"Curr. Opin. Ophthalmol."},{"key":"ref_4","unstructured":"Elliott, D.B. (2016). Objective refraction. Optometry: Science, Techniques and Clinical Management, Elsevier. [2nd ed.]."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"102361","DOI":"10.1016\/j.inffus.2024.102361","article-title":"A review of cancer data fusion methods based on deep learning","volume":"108","author":"Zhao","year":"2024","journal-title":"Inf. Fusion"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"8","DOI":"10.1167\/tvst.11.4.8","article-title":"Prediction of Subjective Refraction from Anterior Corneal Surface, Eye Lengths, and Age Using Machine Learning Algorithms","volume":"11","author":"Espinosa","year":"2022","journal-title":"Transl. Vis. Sci. Technol."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"S22","DOI":"10.1016\/j.optom.2022.03.001","article-title":"Prediction of manifest refraction using machine learning ensemble models on wavefront aberrometry data","volume":"15","author":"Gil","year":"2022","journal-title":"J. Optom."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Kim, S.R., Kang, D.H., Choe, G.S., and Kim, D.H. (2025). Ensemble machine learning prediction model for clinical refraction using partial interferometry measurements in childhood. PLoS ONE, 20.","DOI":"10.1371\/journal.pone.0328213"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"2861","DOI":"10.1167\/iovs.18-23887","article-title":"Deep Learning for Predicting Refractive Error from Retinal Fundus Images","volume":"59","author":"Varadarajan","year":"2018","journal-title":"Investig. Ophthalmol. Vis. Sci."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"164","DOI":"10.1016\/j.aopr.2024.06.005","article-title":"Ocular image-based deep learning for predicting refractive error: A systematic review","volume":"4","author":"Yew","year":"2024","journal-title":"Adv. Ophthalmol. Pract. Res."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Yang, D., Li, M., Wang, Y., Niu, L., Shen, Y., Zhang, X., Fu, B., and Zhou, X. (2022). Prediction of refractive error based on ultrawide field images with deep learning models in myopia patients. Front. Med., 9.","DOI":"10.3389\/fmed.2022.834281"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"8565","DOI":"10.1038\/s41598-020-65417-y","article-title":"Using Artificial Intelligence and Novel Polynomials to Predict Subjective Refraction","volume":"10","author":"Rampat","year":"2020","journal-title":"Sci. Rep."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"pbae005","DOI":"10.1093\/pcmedi\/pbae005","article-title":"Accurate prediction of myopic progression and high myopia by machine learning","volume":"7","author":"Li","year":"2024","journal-title":"Precis. Clin. Med."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"e57644","DOI":"10.2196\/57644","article-title":"Machine Learning Approaches in High Myopia: Systematic Review and Meta-Analysis","volume":"27","author":"Zuo","year":"2025","journal-title":"J. Med. Internet Res."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1038\/s41746-023-00752-8","article-title":"Deep learning system to predict the 5-year risk of high myopia using fundus imaging in children","volume":"6","author":"Foo","year":"2023","journal-title":"NPJ Digit. Med."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Xie, Y., Zhai, Y., and Lu, G. (2025). Evolution of artificial intelligence in healthcare: A 30-year bibliometric study. Front. Med., 11.","DOI":"10.3389\/fmed.2024.1505692"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Ampong, J., Agyekum, S., Eisenbarth, W., Andoh, A.K.A., Osei Duah Junior, I., Amankwah, G., Agbeshie, G.K., Acquah, E.A., Afari, C., and Assan, E. (2025). Artificial intelligence applications in refractive error management: A systematic review and meta-analysis. PLoS Digit. Health, 4.","DOI":"10.1371\/journal.pdig.0000904"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"367","DOI":"10.1097\/00006324-199706000-00019","article-title":"Power vectors: An application of Fourier analysis to the description and statistical analysis of refractive error","volume":"74","author":"Thibos","year":"1997","journal-title":"Optom. Vis. Sci."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"e081554","DOI":"10.1136\/bmj-2024-081554","article-title":"FUTURE-AI: International consensus guideline for trustworthy and deployable artificial intelligence in healthcare","volume":"388","author":"Lekadir","year":"2025","journal-title":"BMJ"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Wong, K.K.L., Han, Y., Cai, Y., Ouyang, W., Du, H., and Liu, C. (2025). From Trust in Automation to Trust in AI in Healthcare: A 30-Year Longitudinal Review and an Interdisciplinary Framework. Bioengineering, 12.","DOI":"10.3390\/bioengineering12101070"}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/2\/331\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,20]],"date-time":"2026-01-20T09:41:03Z","timestamp":1768902063000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/2\/331"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,20]]},"references-count":20,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2026,1]]}},"alternative-id":["diagnostics16020331"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16020331","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,20]]}}}