Medical hypothesis, discovery & innovation in optometry

Developing a compact machine learning–based predictor for detecting above-mild ocular surface disease index scores

Medical hypothesis, discovery & innovation in optometry, Vol. 7 No. 1 (2026), 23 May 2026 , Page 41-49
https://doi.org/10.51329/mehdioptometry244

Abstract

Background: The 12-item Ocular Surface Disease Index (OSDI) is widely used to assess dry eye disease (DED) severity; however, its length may reduce patient compliance and clinical efficiency. Given the expanding role of machine learning in ophthalmology, we aimed to develop and validate a compact OSDI predictor using supervised machine learning techniques to improve the efficiency and accuracy of DED assessment.
Methods: This retrospective study analyzed a dataset of complete 12-item OSDI questionnaires obtained from adult residents of the Gaza Strip. OSDI scores were recalculated using the standard scoring formula, with “not applicable” responses treated as missing and excluded from the denominator. Participants were categorized as having moderate-to-severe dry eye (OSDI > 22) or not. Three supervised machine learning models—decision tree, support vector machine, and logistic regression—were developed using Python. Binary feature-importance analysis was initially performed using the full 12-item OSDI, after which each model was retrained using only questionnaire items with a binary feature-importance value of 1. Model performance was evaluated using accuracy, sensitivity, specificity, and precision.
Results: Among 452 participants (mean [standard deviation] age, 32.0 [11.8] years; 52.9% male), 252 (55.8%) were classified as having moderate-to-severe dry eye, 200 (44.2%) were not. In the reduced-feature testing models, support vector machine model demonstrated the best overall performance, achieving 94.5% accuracy, 98.0% sensitivity, 90.0% specificity, and 92.6% precision. Logistic regression also showed strong performance, with 93.4% accuracy, 98.0% sensitivity, 87.5% specificity, and 90.9% precision. The decision tree model yielded lower testing accuracy (78.0%) and sensitivity (70.6%) but maintained relatively high specificity (87.5%) and precision (87.8%). Feature-importance analysis identified sensitivity to light, gritty sensation, computer or bank machine use, windy conditions, low-humidity environments, and air-conditioned places as informative predictors in the decision tree model. Both support vector machine and logistic regression models identified gritty sensation, painful or sore eyes, blurred vision, reading, watching television, and air-conditioned places as key predictors.
Conclusions: Supervised machine learning models, particularly support vector machine and logistic regression models, effectively classified moderate-to-severe dry eye using recalculated standard OSDI scores and reduced feature sets. The identified predictors underscore the importance of ocular discomfort, visual disturbance, sustained visual activities, and environmental triggers in DED symptom severity. These findings support the potential utility of machine learning-assisted tools for symptom-based DED screening and severity assessment. Further validation in independent clinical populations and integration with objective diagnostic measures are warranted.
Keywords:
  • dry eye disease
  • questionnaire
  • ocular surface disease index
  • OSDI-6
  • machine learning
  • AI (artificial intelligence)
  • learning
  • deep
  • machine
  • vision
  • sciences
  • optometrist

References

Stapleton F, Alves M, Bunya VY, Jalbert I, Lekhanont K, Malet F, Na KS, Schaumberg D, Uchino M, Vehof J, Viso E, Vitale S, Jones L. TFOS DEWS II Epidemiology Report. Ocul Surf. 2017 Jul;15(3):334-365. doi: 10.1016/j.jtos.2017.05.003. Epub 2017 Jul 20. PMID: 28736337.

Na KS, Han K, Park YG, Na C, Joo CK. Depression, Stress, Quality of Life, and Dry Eye Disease in Korean Women: A Population-Based Study. Cornea. 2015 Jul;34(7):733-8. doi: 10.1097/ICO.0000000000000464. PMID: 26002151.

Schiffman RM, Walt JG, Jacobsen G, Doyle JJ, Lebovics G, Sumner W. Utility assessment among patients with dry eye disease. Ophthalmology. 2003 Jul;110(7):1412-9. doi: 10.1016/S0161-6420(03)00462-7. PMID: 12867401.

Wolffsohn JS, Arita R, Chalmers R, Djalilian A, Dogru M, Dumbleton K, Gupta PK, Karpecki P, Lazreg S, Pult H, Sullivan BD, Tomlinson A, Tong L, Villani E, Yoon KC, Jones L, Craig JP. TFOS DEWS II Diagnostic Methodology report. Ocul Surf. 2017 Jul;15(3):539-574. doi: 10.1016/j.jtos.2017.05.001. Epub 2017 Jul 20. PMID: 28736342.

Sheppard J, Shen Lee B, Periman LM. Dry eye disease: identification and therapeutic strategies for primary care clinicians and clinical specialists. Ann Med. 2023 Dec;55(1):241-252. doi: 10.1080/07853890.2022.2157477. PMID: 36576348; PMCID: PMC9809411.

Barabino S, Labetoulle M, Rolando M, Messmer EM. Understanding Symptoms and Quality of Life in Patients With Dry Eye Syndrome. Ocul Surf. 2016 Jul;14(3):365-76. doi: 10.1016/j.jtos.2016.04.005. Epub 2016 May 17. PMID: 27224876.

Nichols KK, Mousavi M (2023). 'Chapter 2 - Clinical Assessments of Dry Eye Disease: Tear Film and Ocular Surface Health. In: Dry eye disease 2023 (pp. 15-23)'. Elsevier. doi: 10.1016/B978-0-323-82753-9.00002-3.

Wolffsohn JS (2023). 'Chapter 1 - Questionnaire Design and Use to Assess Dry Eye Disease. In: Dry Eye Disease 2023 Jan 1 (pp. 1-14)'. Elsevier. doi: 10.1016/B978-0-323-82753-9.00006-0.

Midorikawa-Inomata A, Inomata T, Nojiri S, Nakamura M, Iwagami M, Fujimoto K, Okumura Y, Iwata N, Eguchi A, Hasegawa H, Kinouchi H, Murakami A, Kobayashi H. Reliability and validity of the Japanese version of the Ocular Surface Disease Index for dry eye disease. BMJ Open. 2019 Nov 25;9(11):e033940. doi: 10.1136/bmjopen-2019-033940. PMID: 31772113; PMCID: PMC6886996.

Barber L, Khodai O, Croley T, Lievens C, Montaquila S, Ziemanski J, McCart M, Lunacsek O, Burk C, Patel V. Dry eye symptoms and impact on vision-related function across International Task Force guidelines severity levels in the United States. BMC Ophthalmol. 2018 Sep 29;18(1):260. doi: 10.1186/s12886-018-0919-7. PMID: 30268117; PMCID: PMC6162873.

Herbaut A, Liang H, Rabut G, Trinh L, Kessal K, Baudouin C, Labbé A. Impact of Dry Eye Disease on Vision Quality: An Optical Quality Analysis System Study. Transl Vis Sci Technol. 2018 Jul 12;7(4):5. doi: 10.1167/tvst.7.4.5. PMID: 30009091; PMCID: PMC6042522.

Schiffman RM, Christianson MD, Jacobsen G, Hirsch JD, Reis BL. Reliability and validity of the Ocular Surface Disease Index. Arch Ophthalmol. 2000 May;118(5):615-21. doi: 10.1001/archopht.118.5.615. PMID: 10815152.

Pult H, Wolffsohn JS. The development and evaluation of the new Ocular Surface Disease Index-6. Ocul Surf. 2019 Oct;17(4):817-821. doi: 10.1016/j.jtos.2019.08.008. Epub 2019 Aug 20. PMID: 31442595.

Qin G, Moutari S, Chen J, Xu L, He W, He X, Pazo EE, Yu S. Validity of OSDI-6 questionnaire in a Chinese adult population. Sci Rep. 2024 Jun 21;14(1):14320. doi: 10.1038/s41598-024-64953-1. PMID: 38906923; PMCID: PMC11192897.

Facchin A, Boccardo L. Psychometric properties and diagnostic performance of three dry eye questionnaires in Italian: OSDI, OSDI-6, and SPEED. Optom Vis Sci. 2024 Sep 1;101(9):579-588. doi: 10.1097/OPX.0000000000002184. PMID: 39514738.

Tak A. Artificial Intelligence and Machine Learning in Diagnostics and Treatment Planning. Journal of Artificial Intelligence & Cloud Computing. Scientific Research and Community/Journal of Artificial Intelligence & Cloud Computing-162. 2023; 2(1):1-6. doi: 10.47363/JAICC/2023(2)150.

Balyen L, Peto T. Promising Artificial Intelligence-Machine Learning-Deep Learning Algorithms in Ophthalmology. Asia Pac J Ophthalmol (Phila). 2019 May-Jun;8(3):264-272. doi: 10.22608/APO.2018479. Epub 2019 May 31. PMID: 31149787.

Vyas A, Raman S, Sen S, Ramasamy K, Rajalakshmi R, Mohan V, Raman R. Machine Learning-Based Diagnosis and Ranking of Risk Factors for Diabetic Retinopathy in Population-Based Studies from South India. Diagnostics (Basel). 2023 Jun 16;13(12):2084. doi: 10.3390/diagnostics13122084. PMID: 37370980; PMCID: PMC10297706.

Tsao HY, Chan PY, Su EC. Predicting diabetic retinopathy and identifying interpretable biomedical features using machine learning algorithms. BMC Bioinformatics. 2018 Aug 13;19(Suppl 9):283. doi: 10.1186/s12859-018-2277-0. PMID: 30367589; PMCID: PMC6101083.

Amparo F, Schaumberg DA, Dana R. Comparison of Two Questionnaires for Dry Eye Symptom Assessment: The Ocular Surface Disease Index and the Symptom Assessment in Dry Eye. Ophthalmology. 2015 Jul;122(7):1498-503. doi: 10.1016/j.ophtha.2015.02.037. Epub 2015 Apr 8. PMID: 25863420; PMCID: PMC4485570.

Jin K, Ye J. Artificial intelligence and deep learning in ophthalmology: Current status and future perspectives. Adv Ophthalmol Pract Res. 2022 Aug 24;2(3):100078. doi: 10.1016/j.aopr.2022.100078. PMID: 37846285; PMCID: PMC10577833.

Ting DSW, Pasquale LR, Peng L, Campbell JP, Lee AY, Raman R, Tan GSW, Schmetterer L, Keane PA, Wong TY. Artificial intelligence and deep learning in ophthalmology. Br J Ophthalmol. 2019 Feb;103(2):167-175. doi: 10.1136/bjophthalmol-2018-313173. Epub 2018 Oct 25. PMID: 30361278; PMCID: PMC6362807.

Krishnan A, Dutta A, Srivastava A, Konda N, Prakasam RK. Artificial Intelligence in Optometry: Current and Future Perspectives. Clin Optom (Auckl). 2025 Mar 12;17:83-114. doi: 10.2147/OPTO.S494911. PMID: 40094103; PMCID: PMC11910921.

Aljarousha M, Badarudin NE, Che Azemin MZ, Aljeesh Y, Amer A, Abdul Rahim MAS. The validity and reliability of the Arabic version of the ocular surface disease index (OSDI) questionnaire in a sample of the Gazan population: a study from Palestine. Int Ophthalmol. 2023 Apr;43(4):1303-1316. doi: 10.1007/s10792-022-02528-7. Epub 2022 Sep 26. PMID: 36156181.

Aljarousha M, Alghamdi WM, Che Azemin MZ, Mahmud M. Epidemiology of ocular surface symptoms and their association with stress levels among the Gazan population in crowded shelters during the 2023 Israel war. Cont Lens Anterior Eye. 2025 Jul;48(4):102403. doi: 10.1016/j.clae.2025.102403. Epub 2025 Mar 17. PMID: 40102140.

Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D, Brucher M, Perrot M, Duchesnay E. Scikit-learn: Machine learning in Python. the Journal of machine Learning research. 2011 Nov 1;12(85):2825-30. (Machine Learning Open Source Software Paper). Link.

Anteneh A, Belachew M, Birhanu S, Meseret M. Predicting cardiovascular disease among diabetic patients in Ethiopia using machine learning models: evidence from Ethiopian public health Institute data (2024/2025). BMC Public Health. 2025 Nov 24;25(1):4114. doi: 10.1186/s12889-025-24850-2. PMID: 41286785; PMCID: PMC12642091.

Uddin S, Khan A, Hossain ME, Moni MA. Comparing different supervised machine learning algorithms for disease prediction. BMC Med Inform Decis Mak. 2019 Dec 21;19(1):281. doi: 10.1186/s12911-019-1004-8. PMID: 31864346; PMCID: PMC6925840.

Diallo R, Edalo C, Awe OO (2025). 'Machine Learning Evaluation of Imbalanced Health Data: A Comparative Analysis of Balanced Accuracy, MCC, and F1 Score'. In: Awe, O.O., A. Vance, E. (eds) Practical Statistical Learning and Data Science Methods: Case Studies from LISA 2020 Global Network, USA 2024 Dec 28 (pp. 283-312)'. STEAM-H: Science, Technology, Engineering, Agriculture, Mathematics & Health. Springer Nature Switzerland, Cham. Print ISBN: 978-3-031-72214-1; Online ISBN: 978-3-031-72215-8; doi: 10.1007/978-3-031-72215-8_12.

Miller KL, Walt JG, Mink DR, Satram-Hoang S, Wilson SE, Perry HD, Asbell PA, Pflugfelder SC. Minimal clinically important difference for the ocular surface disease index. Arch Ophthalmol. 2010 Jan;128(1):94-101. doi: 10.1001/archophthalmol.2009.356. PMID: 20065224.

Okumura Y, Inomata T, Iwata N, Sung J, Fujimoto K, Fujio K, Midorikawa-Inomata A, Miura M, Akasaki Y, Murakami A. A Review of Dry Eye Questionnaires: Measuring Patient-Reported Outcomes and Health-Related Quality of Life. Diagnostics (Basel). 2020 Aug 5;10(8):559. doi: 10.3390/diagnostics10080559. PMID: 32764273; PMCID: PMC7459853.

Katz G, Springs CL, Craven ER, Montecchi-Palmer M. Ocular surface disease in patients with glaucoma or ocular hypertension treated with either BAK-preserved latanoprost or BAK-free travoprost. Clin Ophthalmol. 2010 Nov 3;4:1253-61. doi: 10.2147/OPTH.S14113. PMID: 21151330; PMCID: PMC2993125.

Li Z, Wang L, Wu X, Jiang J, Qiang W, Xie H, Zhou H, Wu S, Shao Y, Chen W. Artificial intelligence in ophthalmology: The path to the real-world clinic. Cell Rep Med. 2023 Jul 18;4(7):101095. doi: 10.1016/j.xcrm.2023.101095. Epub 2023 Jun 28. PMID: 37385253; PMCID: PMC10394169.

Nam SM, Peterson TA, Butte AJ, Seo KY, Han HW. Explanatory Model of Dry Eye Disease Using Health and Nutrition Examinations: Machine Learning and Network-Based Factor Analysis From a National Survey. JMIR Med Inform. 2020 Feb 20;8(2):e16153. doi: 10.2196/16153. PMID: 32130150; PMCID: PMC7059080.

Graham AD, Kothapalli T, Wang J, Ding J, Tse V, Asbell PA, Yu SX, Lin MC. A machine learning approach to predicting dry eye-related signs, symptoms and diagnoses from meibography images. Heliyon. 2024 Aug 13;10(17):e36021. doi: 10.1016/j.heliyon.2024.e36021. PMID: 39286076; PMCID: PMC11403426.

Li S, Wang Y, Yu C, Li Q, Chang P, Wang D, Li Z, Zhao Y, Zhang H, Tang N, Guan W, Fu Y, Zhao YE. Unsupervised Learning Based on Meibography Enables Subtyping of Dry Eye Disease and Reveals Ocular Surface Features. Invest Ophthalmol Vis Sci. 2023 Oct 3;64(13):43. doi: 10.1167/iovs.64.13.43. PMID: 37883092; PMCID: PMC10615148.

Zhang Z, Lin X, Yu X, Fu Y, Chen X, Yang W, Dai Q. Meibomian Gland Density: An Effective Evaluation Index of Meibomian Gland Dysfunction Based on Deep Learning and Transfer Learning. J Clin Med. 2022 Apr 25;11(9):2396. doi: 10.3390/jcm11092396. PMID: 35566522; PMCID: PMC9099803.

Saha RK, Chowdhury AMM, Na KS, Hwang GD, Eom Y, Kim J, Jeon HG, Hwang HS, Chung E. Automated quantification of meibomian gland dropout in infrared meibography using deep learning. Ocul Surf. 2022 Oct;26:283-294. doi: 10.1016/j.jtos.2022.06.006. Epub 2022 Jun 24. PMID: 35753666.

Santos LF, Sánchez-Tena MÁ, Alvarez-Peregrina C, Martinez-Perez C. Artificial Intelligence-Driven Diagnostics in Eye Care: A Random Forest Approach for Data Classification and Predictive Modeling. Algorithms. 2025 Oct 15;18(10):647. doi: 10.3390/a18100647.

Aljarousha M, Alghamdi WM, Attaallah S, Alhoot MA. Ocular Surface Disease Index questionnaire in different languages. Med Hypothesis Discov Innov Ophthalmol. 2025 Feb 1;13(4):190-200. doi: 10.51329/mehdiophthal1510. PMID: 40065803; PMCID: PMC11890261.

Morthen MK, Magno MS, Utheim TP, Snieder H, Jansonius N, Hammond CJ, Vehof J. The vision-related burden of dry eye. Ocul Surf. 2022 Jan;23:207-215. doi: 10.1016/j.jtos.2021.10.007. Epub 2021 Oct 28. PMID: 34743866.

Craig JP, Nichols KK, Akpek EK, Caffery B, Dua HS, Joo CK, Liu Z, Nelson JD, Nichols JJ, Tsubota K, Stapleton F. TFOS DEWS II Definition and Classification Report. Ocul Surf. 2017 Jul;15(3):276-283. doi: 10.1016/j.jtos.2017.05.008. Epub 2017 Jul 20. PMID: 28736335.

Kalangara JP, Galor A, Levitt RC, Covington DB, McManus KT, Sarantopoulos CD, Felix ER. Characteristics of Ocular Pain Complaints in Patients With Idiopathic Dry Eye Symptoms. Eye Contact Lens. 2017 May;43(3):192-198. doi: 10.1097/ICL.0000000000000249. PMID: 26925537; PMCID: PMC5003761.

Koh S. Mechanisms of Visual Disturbance in Dry Eye. Cornea. 2016 Nov;35 Suppl 1:S83-S88. doi: 10.1097/ICO.0000000000000998. PMID: 27583799.

Patel S, Henderson R, Bradley L, Galloway B, Hunter L. Effect of visual display unit use on blink rate and tear stability. Optom Vis Sci. 1991 Nov;68(11):888-92. doi: 10.1097/00006324-199111000-00010. PMID: 1766652.

Al-Mohtaseb Z, Schachter S, Shen Lee B, Garlich J, Trattler W. The Relationship Between Dry Eye Disease and Digital Screen Use. Clin Ophthalmol. 2021 Sep 10;15:3811-3820. doi: 10.2147/OPTH.S321591. PMID: 34531649; PMCID: PMC8439964.

Gulati A, Sullivan R, Buring JE, Sullivan DA, Dana R, Schaumberg DA. Validation and repeatability of a short questionnaire for dry eye syndrome. Am J Ophthalmol. 2006 Jul;142(1):125-131. doi: 10.1016/j.ajo.2006.02.038. PMID: 16815260.

Wang Z, Keane PA, Chiang M, Cheung CY, Wong TY, Ting DS (2022). 'Artificial Intelligence and Deep Learning in Ophthalmology (pp 1519–1552)'. In: Lidströmer N, Ashrafian H. (eds) Artificial Intelligence in Medicine'. Springer, Cham. Print ISBN: 978-3-030-64572-4; Online ISBN: 978-3-030-64573-1; doi: 10.1007/978-3-030-64573-1_200.

  • Abstract Viewed: 0 times
  • Full Text PDF Downloaded: 0 times

Download Statastics

Open Journal Systems
Make a Submission