Big data and artificial intelligence in ophthalmology - clinical application and future exploration

Tae‐im Kim, Darren Shu, Jeng Ting, Yi‐Ting Hsieh, Tyler Hyungtaek Rim, Darren Jeng, Yong Zi Tan, Syed Muhammad, Ali Imran, Waqas Muhammad, Muhammad Waqas Saleem, Abida Hameed, Rizwan Hussain, Seung Naqvi, Xinxing Guo, Jian Zhang, Xing Liu, Robert T. Chang, Mingguang He, Sunil K. Jalamangala Shivananjaiah, Sneha Kumari, Iyad Majid, Sophia Wang, Jingwen Wang, Jinhong Wang, Dan Chen, Xingdi Wu, Zhe Xu, Xuewen Yu, Siting Sheng, Xueqi Lin, Xiang Chen, Jian Wu, Haochao Ying, Wen Xu, Yeo Kyoung Won, Hyebin Lee, Youngjun Kim, Gyule Han, Tae‐Young Chung, Yong Man Ro, Dong Hui Lim, Kaiqun Liu, Shaopeng Liu, Xiao Jian Tan, Wangting Li, Ling Wang, Xinnan Li, Xiaoyu Xu, Yue Fu, Xiaoning Liu, Jiaming Hong, Haotian Lin, Hui Yang, Tanvi Verma, Liyuan Jin, Jun Zhou, Jia Huang, Mingrui Tan, Benjamin Chen, Ming Choong, Ting Fang Tan, Fei Gao, Xinxing Xu, Daniel Ting, Yong Liu, Yun Bai, Jing Li, Lianjun Shi, Qin Jiang, Biao Yan, Zhenhua Wang, David F. Gibson, Thai Tran, Vidhur Raveendran, Clémence Bonnet, Nathan Siu, M. Vinet, Theo Stoddard-Bennett, Corey Arnold, Sophie X. Deng, William Speier, Jodhbir S. Mehta, Usman Ali, Muhammad Owais, Rizwan Ali Naqvi, Ali Hussain, Seung Hwan Lee, M Arsalan, J Choi, T Mahmood, K Park, A Khawaja, T Khan, K NaveedFrontiers Research Topics 2023 · DOI · OpenAlex

Abstract

Frontiers is a pioneering approach to the world of academia, radically improving the way scholarly research is managed. The grand vision of Frontiers is a world where all people have an equal opportunity to seek, share and generate knowledge. Frontiers provides immediate and permanent online open access to all its publications, but this alone is not enough to realize our grand goals.

Introduction

Ophthalmology is one of the most data-intensive fields in medicine, relying heavily on high-resolution imaging—such as fundus photography and Optical Coherence Tomography (OCT)—alongside detailed Electronic Health Records (EHRs). Despite this wealth of data, clinical practice faces significant bottlenecks, including human variability in diagnosis, the static nature of current AI models, and the legal complexities of data privacy.

This research collection, published in Frontiers in Medicine, explores the integration of Big Data and Artificial Intelligence (AI) to overcome these hurdles. AI is evolving beyond simple image recognition to become a comprehensive clinical assistant. By fusing imaging with clinical text and employing advanced techniques like latent diffusion and continual learning, AI can now match or exceed human experts in detecting blinding diseases while addressing real-world challenges like data privacy and model efficiency.

Methodological Innovations and Datasets

The research highlights several novel architectural approaches designed to improve diagnostic accuracy and operational efficiency:

  • Multimodal Fusion: A standout architecture that combines structured EHR data (e.g., intraocular pressure) with unstructured clinical free-text notes processed via Natural Language Processing (NLP).
  • Feature Preserving Mesh Networks (FPM-Net): A shallow, high-performance network (2.45 million parameters) developed to solve the "vanishing gradient" problem in segmenting tiny retinal vessels more efficiently than standard deep models.
  • Latent Diffusion Augmentation: For rare conditions like Limbal Stem Cell Deficiency, researchers utilized Stable Diffusion models to generate synthetic training images, significantly boosting model robustness.
  • Privacy-Preserving Continual Learning: To allow models to learn from new datasets without "catastrophic forgetting" or violating privacy, the authors evaluated "Brain-Inspired Replay" (BIR) and "Efficient Feature Transformations" (EFT).
Clinical data collection and processing form the foundation of AI-driven ophthalmic diagnostics.

These methods were validated using a mix of classic benchmarks (DRIVE, CHASE-DB1, STARE) and massive real-world clinical cohorts, including data from over 3,400 glaucoma patients at Stanford University and 210,473 fundus photographs for multi-disease classification.

Key Insights and Results

The findings suggest that clinical text acts as a vital biomarker; models that "read" clinical notes are significantly better at predicting near-term disease progression than those relying on imaging alone. Key results include:

  • Glaucoma Prediction: A multimodal model achieved an AUC of 0.899 in predicting whether a patient would require surgery within 12 months. For identifying glaucomatous optic neuropathy, AI achieved an AUC of 0.994, matching top-tier specialists.
  • Infection Classification: A deep learning system for distinguishing bacterial from fungal keratitis reached 87.8% accuracy, a significant improvement over the 81.1% baseline.
  • Vessel Segmentation: The FPM-Net achieved 96.92% accuracy on the DRIVE database while remaining computationally light.
  • Cataract Outcomes: Models using preoperative OCT scans achieved a mean absolute error of only 0.1194 logMAR in predicting post-operative vision.

Limitations and Future Directions

While promising, several hurdles remain for the next generation of ophthalmic AI. Many models are trained on specific ethnic populations, raising questions about generalizability across different demographics. Furthermore, AI performance still degrades with poor-quality or "noisy" images common in real-world settings.

Future work will focus on the "Black Box" problem—improving the explainability of multimodal fusion models—and moving these tools into mobile applications for remote screening and longitudinal disease tracking using 3D OCT data.

Summary updated

Abstract

it is a pioneering approach to the world of academia, radically improving the way scholarly research is managed. The grand vision of Frontiers is a world where all people have an equal opportunity to seek, share and generate knowledge. Frontiers provides immediate and permanent online open access to all its publications, but this alone is not enough to realize our grand goals.
BibTeX
@inproceedings{kim2023big,
  author = {Tae‐im Kim and Darren Shu and Jeng Ting and Yi‐Ting Hsieh and Tyler Hyungtaek Rim and Darren Jeng and Yong Zi Tan and Darren Shu Jeng Ting and Syed Muhammad and Ali Imran and Waqas Muhammad and Muhammad Waqas Saleem and Abida Hameed and Rizwan Hussain and Seung Naqvi and Xinxing Guo and Jian Zhang and Xing Liu and Robert T. Chang and Mingguang He and Sunil K. Jalamangala Shivananjaiah and Sneha Kumari and Iyad Majid and Sophia Wang and Jingwen Wang and Jinhong Wang and Dan Chen and Xingdi Wu and Zhe Xu and Xuewen Yu and Siting Sheng and Xueqi Lin and Xiang Chen and Jian Wu and Haochao Ying and Wen Xu and Yeo Kyoung Won and Hyebin Lee and Youngjun Kim and Gyule Han and Tae‐Young Chung and Yong Man Ro and Dong Hui Lim and Kaiqun Liu and Shaopeng Liu and Xiao Jian Tan and Wangting Li and Ling Wang and Xinnan Li and Xiaoyu Xu and Yue Fu and Xiaoning Liu and Jiaming Hong and Haotian Lin and Hui Yang and Tanvi Verma and Liyuan Jin and Jun Zhou and Jia Huang and Mingrui Tan and Benjamin Chen and Ming Choong and Ting Fang Tan and Fei Gao and Xinxing Xu and Daniel Ting and Yong Liu and Yun Bai and Jing Li and Lianjun Shi and Qin Jiang and Biao Yan and Zhenhua Wang and David F. Gibson and Thai Tran and Vidhur Raveendran and Clémence Bonnet and Nathan Siu and M. Vinet and Theo Stoddard-Bennett and Corey Arnold and Sophie X. Deng and William Speier and Jodhbir S. Mehta and Darren Shu Jeng Ting and Usman Ali and Muhammad Owais and Rizwan Ali Naqvi and Ali Hussain and Seung Hwan Lee and Naqvi and Seung Lee and M Owais and M Arsalan and J Choi and T Mahmood and K Park and A Khawaja and T Khan and K Naveed},
  title = {{Big data and artificial intelligence in ophthalmology - clinical application and future exploration}},
  booktitle = {Frontiers Research Topics},
  year = {2023},
  doi = {10.3389/978-2-8325-4171-5},
}
Figure