Citation :

Subrina Akter, Rahasina Hossain, Kaniz Fatema Tanni, Samsun Nahar Wafia, Sumayea Benta Hasan, Abdullah Man, Nor Hidayati Abdul Aziz, "Subgroup-Aware Evaluation and Interpretability in Myopia Prediction using Deep ROC and XAI," International Journal of Engineering Trends and Technology (IJETT), vol. 74, no. 10, pp. 110-127, 2026. Crossref, https://doi.org/10.14445/22315381/IJETT-V74I10P109

Abstract

Machine Learning (ML) models are familiar for their great success in classification challenges within different domains. For medical diagnostics, those models have numerous applications as well. Basically, these ML models generate good predictions. However, sometimes, for clinicians, they are unable to depend on and use those models in a timely manner while making clinical decisions. Moreover, this factor arises because of the "black box" nature of these models. In medicine, it is a matter of concern. Because it is crucial to understand the base for predictions of patients’ safety and treatment assurance. Besides, recording key performance differences between risk groups is significant for informed clinical decision-making. In that case, traditional measures like the Receiver Operating Characteristic (ROC) curve and the Area Under the Curve (AUC) show some limitations. Hence, to defeat such matters, this study proposes a unified framework for experimenting with Myopia prediction (known as nearsightedness) and evaluating the results using some advanced techniques. This approach integrates Deep ROC analysis with Explainable AI (XAI) methods, where for XAI, Local Interpretable Model-agnostic Explanations (LIME) and Shapley Additive Explanations (SHAP) are applied. Here, the Deep ROC analysis is more refined and has clinical relevance, as the model performance is generated within subgroups defined by False Positive Rate (FPR). On the other hand, it allows for an in-depth understanding of model behavior in high-risk, medium-risk, and low-risk patient populations. Meanwhile, LIME and SHAP offer XAI value by explaining how individual attributes contribute to model output. While doing myopia prediction, this proposed evaluation structure is applied to the machine learning models, named Support Vector Machine (SVM), Random Forest (RF), Multi-Layer Perceptron (MLP), and Decision Tree (DT). Firstly, these models are trained and then evaluated to ensure robustness and reduce bias. After the evaluation, Random Forest (RF) outperforms the other three models across all subgroups, with an AUC of 98.92% in high-risk data 99.49% in low-risk data. The Brier score further confirmed the reliability of the model and improved its predictive power. The union of Deep ROC and XAI techniques significantly enhances the explainability and clinical utility of RF, with proof of how it forecasts for different risk levels. The approach is a leap forward towards the production of explainable, accurate, and clinically valuable AI models in medicine.

Keywords

Deep Roc Analysis, Explainable AI (XAI), Myopia prediction, Machine Learning, Model interpretabilit.

References

1. Shahab S. Band et al., “Application of Explainable Artificial Intelligence in Medical Health: A Systematic Review of Interpretability Methods,” Informatics in Medicine Unlocked, vol. 40, pp. 1-20, 2023.
   [CrossRef] [Google Scholar] [Publisher Link] 
2. Amir Masoud Rahmani et al., “Machine Learning (ML) in Medicine: Review, Applications, and Challenges,” Mathematics, vol. 9, no. 22, pp. 1-52, 2021.
   [CrossRef] [Google Scholar] [Publisher Link]
3. André M. Carrington et al., “A new Concordant Partial AUC and Partial c Statistic for Imbalanced Data in the Evaluation of Machine Learning Algorithms,” BMC Medical Informatics and Decision Making, vol. 20, no. 4, pp. 1-12, 2020.
   [CrossRef] [Google Scholar] [Publisher Link]
4. Lior Rabkin, Ilan Cohen, and Gonen Singer, “Resource Allocation in Ordinal Classification Problems: A Prescriptive Framework Utilizing Machine Learning and Mathematical Programming,” Engineering Applications of Artificial Intelligence, vol. 132, 2024.
   [CrossRef] [Google Scholar] [Publisher Link]
5. Elie Dolgin, “The Myopia Boom,” Nature, vol. 519, no. 7543, pp. 276-278, 2015.
   [CrossRef] [Google Scholar] [Publisher Link] 
6. Brien A. Holden et al., “Global Prevalence of Myopia and High Myopia and Temporal Trends from 2000 through 2050,” Ophthalmology, vol. 123, no. 5, pp. 1036-1042, 2016. 
   [CrossRef] [Google Scholar] [Publisher Link]
7. Lik Thai Lim et al., “Impact of Parental History of Myopia on the Development of Myopia in Mainland China School-Aged Children,” Therapeutic Advances in Ophthalmology, vol. 6, pp. 31-35, 2014.
   [CrossRef] [Google Scholar] [Publisher Link]
8. L.L.K. Lin et al., “Prevalence of Myopia in Taiwanese Schoolchildren: 1983 to 2000,” Annals-Academy of Medicine Singapore, vol. 33, no. 1, pp. 27-33, 2004.
   [CrossRef] [Google Scholar] [Publisher Link]
9. Mingguang He et al., “Effect of Time Spent Outdoors at School on the Development of Myopia Among Children in China: A Randomized Clinical Trial,” Jama Network, vol. 314, no. 11, pp. 1142-1148, 2015.
   [CrossRef] [Google Scholar] [Publisher Link]
10. Ian G Morgan, Kyoko Ohno-Matsui and Seang-Mei Saw, “Myopia,” The Lancet, vol. 379, no. 9827, pp. 1739-1748, 2012.
    [Google Scholar] [Publisher Link]
11. Chen-Wei Pan, Dharani Ramamurthy, and Seang-Mei Saw, “Worldwide Prevalence and Risk Factors for Myopia,” Ophthalmic and Physiological Optics, vol. 32, no. 1, pp. 3-16, 2012.
    [CrossRef] [Google Scholar] [Publisher Link]
12. Richard F. Spaide, Kyoko Ohno-Matsui, and Lawrence A. Yannuzzi, Pathologic Myopia, 1^st ed., Springer New York, NY, 2014. 
    [CrossRef] [Google Scholar] [Publisher Link]
13. Ian G. Morgan et al., “The Epidemics of Myopia: Aetiology and Prevention,” Progress in Retinal and Eye Research, vol. 62, pp. 134-149, 2018.
    [CrossRef] [Google Scholar] [Publisher Link]
14. Annechien E.G. Haarman et al., “The Complications of Myopia: A Review and Meta-Analysis,” Investigative Ophthalmology and Visual Science, vol. 61, no. 4, pp. 1-19, 2020.
    [CrossRef] [Google Scholar] [Publisher Link]
15. Pavan Kumar Verkicharla, Priyanka Kammari, and Anthony Vipin Das, “Myopia Progression Varies with Age and Severity of Myopia,” PLoS One, vol. 15, no. 11, pp. 1-9, 2020.
    [CrossRef] [Google Scholar] [Publisher Link]
16. Namra Rauf, Syed Omer Gilani, and Asim Waris, “Automatic Detection of Pathological Myopia using Machine Learning,” Scientific Reports, vol. 11, no. 1, pp. 1-9. 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
17. Shi-Ming Li et al., “Machine Learning to Determine Risk Factors for Myopia Progression in Primary School Children: The Anyang Childhood Eye Study,” Ophthalmology and Therapy, vol. 11, no. 2, pp. 573-585, 2022. 
    [CrossRef] [Google Scholar] [Publisher Link]
18. Jianxia Fang et al., “Machine Learning for Predicting the Treatment Effect of Orthokeratology in Children,” Frontiers in Pediatrics, vol. 10, pp. 1-11, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
19. Yuzhuo Fan et al., “Machine Learning Algorithm Improves Accuracy of Ortho-K Lens Fitting in Vision Shaping Treatment,” Contact Lens and Anterior Eye, vol. 45, no. 3, 2022.
    [CrossRef] [Google Scholar] [Publisher Link]
20. Tien-En Tan et al., “Retinal Photograph-based Deep Learning Algorithms for Myopia and a Blockchain Platform to Facilitate Artificial Intelligence Medical Research: A Retrospective Multicohort Study,” The Lancet Digit Health, vol. 3, no. 5, pp. 317-329, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
21. Ruben Hemelings et al., “Pathological Myopia Classification with Simultaneous Lesion Segmentation using Deep Learning,” Computer Methods and Programs in Biomedicine, vol. 199, pp. 1-18, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
22. Worood Esam Noori Noori, and A.S. Albahri, “Towards Trustworthy Myopia Detection: Integration Methodology of Deep Learning Approach, XAI Visualization, and User Interface System,” Applied Data Science and Analysis, vol. 2023, pp. 1-15, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
23. Jai Prashar, and Nicole Tay, “Performance of Artificial Intelligence for the Detection of Pathological Myopia from Colour Fundus Images: A Systematic Review and Meta-Analysis,” Eye, vol. 38, no. 2, pp. 303-314, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
24. Andrzej Grzybowski et al., “A Review on the Epidemiology of Myopia in School Children Worldwide,” BMC Ophthalmology, vol. 20, no. 1, pp. 1-11, 2020.
    [CrossRef] [Google Scholar] [Publisher Link]
25. Mengru Ba, and Zhijie Li, “The Impact of Lifestyle Factors on Myopia Development: Insights and Recommendations,” AJO International, vol. 1, no. 1, pp. 1-21, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
26. Li Lu et al., “AI-Model for Identifying Pathologic Myopia based on Deep Learning Algorithms of Myopic Maculopathy Classification and “Plus” Lesion Detection in Fundus Images,” Frontiers in Cell and Developmental Biology, vol. 9, pp. 1-10, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
27. Li Lu et al., “Development of Deep Learning-based Detecting Systems for Pathologic Myopia using Retinal Fundus Images,” Communications Biology, vol. 4, no. 1, pp. 1-8, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
28. Chen Guo, Minzhong Yu, and Jing Li., “Prediction of Different Eye Diseases based on Fundus Photography via Deep Transfer Learning,” Journal of Clinical Medicine, vol. 10, no. 23, pp. 1-11, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
29. Ling-Ping Cen et al., “Automatic Detection of 39 Fundus Diseases and Conditions in Retinal Photographs using Deep Neural Networks,” Nature Communications, vol. 12, no. 1, pp. 1-13, 2021. 
    [CrossRef] [Google Scholar] [Publisher Link]
30. Xiaomeng Li et al., “Rotation-Oriented Collaborative Self-Supervised Learning for Retinal Disease Diagnosis,” IEEE Transactions on Medical Imaging, vol. 40, no. 9, pp. 2284-2294, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
31. Junjia Huang et al., “Myopia prediction for Children and Adolescents via Time-Aware Deep Learning,” Scientific Reports, vol. 13, no. 1, pp. 1-10, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
32. Yifan Wang et al., “Enhancing AI Transparency in IoT Intrusion Detection using Explainable AI Techniques,” Internet of Things, vol. 33, pp. 1-26, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
33. André M. Carrington et al., “Deep ROC Analysis and AUC as Balanced Average Accuracy, for Improved Classifier Selection, Audit and Explanation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 45, no. 1, pp. 329-341, 2022.
    [CrossRef] [Google Scholar] [Publisher Link]
34. Jingfeng Mu, Haoxi Zhong, and Mingjie Jiang, “Machine-Learning Models to Predict Myopia in Children and Adolescents,” Frontiers in Medicine, vol. 11, pp. 1-10, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
35. Xu Yang et al., “Prediction of Myopia in Adolescents through Machine Learning Methods,” International Journal of Environmental Research and Public Health, vol. 17, no. 2, pp. 1-17, 2020.
    [CrossRef] [Google Scholar] [Publisher Link]
36. José Carlos Fernández-Álvarez et al., “Using a Multilayer Perceptron in Intraocular Lens Power Calculation,” Journal of Cataract and Refractive Surgery, vol. 45, no. 12, pp. 1753-1761, 2019.
    [CrossRef] [Google Scholar] [Publisher Link]
37. Jun-Wei Chen et al., “Decision Tree Modeling to Predict Myopia Progression in Children Treated with Atropine: Toward Precision Ophthalmology,” Diagnostics, vol. 15, no. 16, pp. 1-13, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
38. Sajid Ali et al., “Explainable Artificial Intelligence (XAI): What we know and what is left to Attain Trustworthy Artificial Intelligence,” Information Fusion, vol. 99, pp. 1-52, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]
39. Turker Berk Donmez et al., “Explainable AI in Action: A Comparative Analysis of Hypertension Risk Factors using SHAP and LIME,” Neural Computing and Applications, vol. 37, no. 5, pp. 4053-4074, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
40. Iasonas Bakas, and Karolos J. Kontoleon, “ANNs and Game Theory SHAP Algorithms Quantifying the Contribution of Geometric and Temperature-Dependent Material Properties to the Thermal Response of Concrete Walls Exposed to Elevated Temperatures,” Construction and Building Materials, vol. 493, pp. 1-14, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
41. Ying Liu et al., “Clinical Decision Support Tool for Breast Cancer Recurrence Prediction using SHAP Value in Cooperative Game Theory,” Heliyon, vol. 10, no. 2, pp. 1-11, 2024.
    [CrossRef] [Google Scholar] [Publisher Link]
42. Haohan Zou et al., “Retinal Features as Predictive Indicators for High Myopia: Insights from Explainable Multi-Machine Learning Models,” Frontiers in Bioengineering and Biotechnology, vol. 13, pp. 1-9, 2025. 
    [CrossRef] [Google Scholar] [Publisher Link]
43. Qi Feng et al., “Interpretable Machine Learning Models for Predicting Childhood Myopia from School-based Screening Data,” Scientific Reports, vol. 15, no. 1, pp. 1-11, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
44. Isteaq Kabir Sifat et al., “Prediction of Myopia Among Undergraduate Students using Ensemble Machine Learning Techniques,” Health Science Reports, vol. 8, no. 5, pp. 1-12, 2025.
    [CrossRef] [Google Scholar] [Publisher Link]
45. George Papachristou, Myopia Study, Kaggle, 2025. [Online]. Available: https://www.kaggle.com/datasets/mscgeorges/myopia-study