Integration of Artificial Intelligence into the Approach for Diagnosis and Monitoring of Dry Eye Disease

Abstract

Dry eye disease (DED) is one of the most common diseases worldwide that can lead to a significant impairment of quality of life. The diagnosis and treatment of the disease are often challenging because of the lack of correlation between the signs and symptoms, limited reliability of diagnostic tests, and absence of established consensus on the diagnostic criteria. The advancement of machine learning, particularly deep learning technology, has enabled the application of artificial intelligence (AI) in various anterior segment disorders, including DED. Currently, many studies have reported promising results of AI-based algorithms for the accurate diagnosis of DED and precise and reliable assessment of data obtained by imaging devices for DED. Thus, the integration of AI into clinical approaches for DED can enhance diagnostic and therapeutic performance. In this review, in addition to a brief summary of the application of AI in anterior segment diseases, we will provide an overview of studies regarding the application of AI in DED and discuss the recent advances in the integration of AI into the clinical approach for DED.

AI-Based Advanced Approaches and Dry Eye Disease Detection Based on Multi-Source Evidence: Cases, Applications, Issues, and Future Directions

“…The Generative Adversarial Network (GAN) is an efficient approach for medical image enhancements and quality improvements by performing a max-min game between two subnetworks in the architecture [69][70][71][72] . The network of generator and discriminator is responsible for maximizing the loss of embedded image and minimizing the reward respectively.…”

“…GAN-based networks have already been used for DED fields, the most application falls on meibomian gland images. Khan et al [71] proposed a GAN framework based on meibomian gland infrared images to solve the image irregularity issues, including improper light focusing and positioning, specular reflection, vague boundaries between intragland and intergland, and eyelid ectropion. Results show that GAN-based methods show a great performance of average Pompeiu-Hausdorff distance, and Aggregated Jaccard Index for Pearson correlation grading, meiboscoring, and Bland-Altman analysis.…”

“…Results show that GAN-based methods show a great performance of average Pompeiu-Hausdorff distance, and Aggregated Jaccard Index for Pearson correlation grading, meiboscoring, and Bland-Altman analysis. It contributes a significant value for MG image quality enhancements [71] . However, the evidence of DED detection involves various types of images and videos.…”

“…Meibomian gland images, captured using the Oculus Keratograph 5M (K5M; Oculus, Wetzlar, Germany) [105] , serve as crucial diagnostic evidence for MGD [35] . The identification of MGD involves recognizing variations in shape and severity levels within the acini [71] . García-Marqués et al [106] suggested that MGD detection based on meibomian gland images is a more comprehensive and powerful approach compared to single metrics [106] .…”

Corneal Imaging Techniques for Dry Eye Disease

“…• Developing new algorithms and models to improve the accuracy and reliability of corneal measurements [96,97].…”