Potential applications of artificial intelligence in image analysis in cornea diseases: a review

Tey, Kai Yuan; Cheong, Ezekiel Ze Ken; Ang, Marcus

doi:10.1186/s40662-024-00376-3

Cited by 6 publications

(1 citation statement)

References 161 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The management of KCN could be revolutionized by further application of artificial intelligence in the diagnosis, grading, and early detection of KCN. Artificial intelligence capabilities in management approach considering the currently available imaging modalities [ 76 , 77 ] and treatment options [ 78 ] would best benefit the patient. Table 1 summarizes novel applications of AI in KCN management [ 79 - 94 ] .…”

Section: Resultsmentioning

confidence: 99%

Keratoconus: imaging modalities and management

Alqudah

2024

Med Hypothesis Discov Innov Ophthalmol

View full text Add to dashboard Cite

Background: Keratoconus (KCN) is characterized by gradual thinning and steepening of the cornea, which can lead to significant vision problems owing to high astigmatism, corneal scarring, or even corneal perforation. The detection of KCN in its early stages is crucial for effective treatment. In this review, we describe current advances in the diagnosis and treatment of KCN. Methods: This narrative review focuses on recent advancements in the diagnosis and treatment of KCN, especially evolving approaches and strategies. To ensure the inclusion of the most recent literature, relevant publications discussing advanced imaging techniques and treatment options for KCN were extensively gathered from the PubMed/MEDLINE and Google Scholar databases. The following index terms and keywords were used for the online search: keratoconus, diagnosis of keratoconus, advances in the diagnosis of keratoconus, topography or tomography, anterior segment optical coherence tomography, treatment of keratoconus, advances in the treatment of keratoconus, collagen crosslinking, intrastromal ring, keratoplasty, and new techniques in keratoconus. Results: Various screening methods such as corneal topography, tomography, anterior segment optical coherence tomography, and assessment of corneal biomechanics have been developed to identify KCN in its early stages. After diagnosis, KCN management focuses on preventing disease progression. Corneal collagen crosslinking is a minimally invasive treatment that can slow or stop the progression of the condition. Recent research has also explored the use of copper sulfate eye drops (IVMED-80) as a noninvasive treatment to prevent the progression of KCN. Current treatment options for visual improvement include scleral lenses, intracorneal ring segments, corneal allogeneic intrastromal ring segments, and deep anterior lamellar keratoplasty. Recently, novel alternative procedures, such as isolated Bowman layer transplantation, either as a corneal stromal inlay or onlay, have demonstrated encouraging outcomes. Artificial intelligence has gained acceptance for providing best practices for the diagnosis and management of KCN, and the science of its application is contentiously debated; however, it may not have been sufficiently developed. Conclusions: Early detection and advancements in screening methods using current imaging modalities have improved diagnosis of KCN. Improvement in the accuracy of current screening or diagnostic tests and comparison of their validities are achievable by well-designed, large-scale, prospective studies. The safety and effectiveness of emerging treatments for KCN are currently being investigated. There is an ongoing need for studies to track progress and evaluate clinicians’ knowledge and practices in treating patients with KCN. Artificial intelligence capabilities in management approach considering the currently available imaging modalities and treatment options would best benefit the patient.

show abstract

Section: Resultsmentioning

confidence: 99%

Keratoconus: imaging modalities and management

Alqudah

2024

Med Hypothesis Discov Innov Ophthalmol

View full text Add to dashboard Cite

show abstract

Current applications of artificial intelligence for Fuchs endothelial corneal dystrophy: a systematic review

Liu,

Kandakji,

Stupnicki

et al. 2024

Preprint

View full text Add to dashboard Cite

Purpose: Fuchs Endothelial Corneal Dystrophy (FECD) is a common, age-related cause of visual impairment. This systematic review synthesizes evidence from the literature on Artificial Intelligence (AI) models developed for the diagnosis and management of FECD. Methods: We conducted a systematic literature search in MEDLINE, PubMed, Web of Science, and Scopus from January 1, 2000, to June 31, 2024. Full-text studies utilizing AI for various clinical contexts of FECD management were included. Data extraction covered model development, predicted outcomes, validation, and model performance metrics. We graded included studies using the Quality Assessment of Diagnostic Accuracies Studies 2 tool. This review adheres to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) recommendations. Results: Nineteen studies were analyzed. Primary AI algorithms applied in FECD diagnosis and management included neural network architectures specialized for computer vision, utilized on confocal or specular microscopy images, or anterior segment optical coherence tomography images. AI was employed in diverse clinical contexts, such as assessing corneal endothelium and edema, and predicting post-corneal transplantation graft detachment and survival. Despite many studies reporting promising model performance, a notable limitation was that only 3 studies performed external validation. Bias introduced by patient selection processes and experimental designs was evident in the included studies. Conclusions: Despite the potential of AI algorithms to enhance FECD diagnosis and prognostication, further work is required to evaluate their real-world applicability and clinical utility. Translational Relevance: This review offers critical insights for researchers, clinicians, and policymakers, aiding their understanding of existing AI research in FECD management and guiding future health service strategies.

show abstract