Automated diagnosis and staging of Fuchs' endothelial cell corneal dystrophy using deep learning

Applications of Imaging Technologies in Fuchs Endothelial Corneal Dystrophy: A Narrative Literature Review

Fuchs endothelial corneal dystrophy (FECD) is a complex genetic disorder characterized by the slow and progressive degeneration of corneal endothelial cells. Thus, it may result in corneal endothelial decompensation and irreversible corneal edema. Moreover, FECD is associated with alterations in all corneal layers, such as thickening of the Descemet membrane, stromal scarring, subepithelial fibrosis, and the formation of epithelial bullae. Hence, anterior segment imaging devices that enable precise measurement of functional and anatomical changes in the cornea are essential for the management of FECD. In this review, the authors will introduce studies on the application of various imaging modalities, such as anterior segment optical coherence tomography, Scheimpflug corneal tomography, specular microscopy, in vitro confocal microscopy, and retroillumination photography, in the diagnosis and monitoring of FECD and discuss the results of these studies. The application of novel technologies, including image processing technology and artificial intelligence, that are expected to further enhance the accuracy, precision, and speed of the imaging technologies will also be discussed.

Comparative Analysis of Tomographic Indicators Forecasting Decompensation in Fuchs Endothelial Corneal Dystrophy

PURPOSE

To compare the performance of 3 commercially available tomographers (the Pentacam Scheimpflug camera, the swept-source optical coherence tomography Casia, and the blue light slit-scanning tomographer Precisio) in the identification of patterns associated with Fuchs endothelial corneal dystrophy (FECD) decompensation.

METHODS

This was a clinic-based cross-sectional imaging study. Pachymetry maps and posterior surface elevation maps were acquired with the 3 devices from 61 eyes affected by FECD. The maps were graded according to the evidence of tomographic patterns predictive of FECD decompensation (loss of parallel isopachs, displacement of the thinnest point, and focal posterior depression) by 2 blind cornea specialists.

RESULTS

The loss of parallel isopachs was significantly less frequently evident in Pentacam pachymetry maps [8%, 95% confidence interval (CI) (3%, 18%)] compared with both the Casia [31%, 95% CI (20%, 44%), P = 0.01] and Precisio devices [24%, 95% CI (15%, 37%), P = 0.05]. The displacement of the thinnest point was graded as most evident in a significantly higher proportion of Precisio pachymetry maps [43%, 95% CI (31%, 55%)] compared with both the Pentacam [13%, 95% CI (6%, 24%), P = 0.001] and Casia devices [21%, 95% CI (12%, 33%), P = 0.03]. There were no significant differences in the identification of focal posterior depression on posterior elevation maps across the 3 devices.

CONCLUSIONS

Identification of patterns predictive of FECD prognosis on pachymetry and posterior elevation maps is possible with different devices. However, their evidence varies across tomographers, and the results from different devices are not interchangeable.

Three-Dimensional Assessment of Descemet Membrane Reflectivity by Optical Coherence Tomography in Fuchs Endothelial Corneal Dystrophy

PURPOSE

This study aimed to evaluate Descemet membrane reflectivity using anterior segment optical coherence tomography (AS-OCT) in eyes with Fuchs endothelial corneal dystrophy (FECD).

METHODS

We retrospectively assessed 144 eyes of 88 consecutive participants (41 FECD, 15 pseudophakic bullous keratopathies [BKs], and 32 healthy controls, 63.5 ± 16.5 years). FECD was graded 0 to 3 based on the guttae areas using specular microscopy. The sum of AS-OCT reflectivity of the 3-dimensional volume from 10 μm thickness from the endothelial surface of the cornea and residual stromal area was calculated as D sum (endo) and D sum (stroma) in the central area of 3- and 6-mm diameters, respectively. The D ES ratio was defined as the ratio of D sum (endo) to D sum (stroma). The percentage of the guttae area in the specular images was calculated using MATLAB. D sum (endo) and D ES ratio were compared among FECD, BK, and healthy controls.

RESULTS

D sum (endo) in FECD grade 3 was significantly higher than that in healthy control eyes, FECD patients with mild and moderate guttae, and BK (all P ≤ 0.040). The D ES ratio in FECD patients with mild to severe guttae (grade 1-3) was significantly higher than that in healthy control eyes and BK (all P ≤ 0.035). The percentage of the guttae area was significantly correlated with D sum (endo) (R = 0.488, P < 0.001 for 3 mm, R = 0.512, P < 0.001 for 6 mm) and D ES ratio (R = 0.450, P < 0.001 for 3 mm, R = 0.588, P < 0.001 for 6 mm).

CONCLUSIONS

Descemet membrane reflectivity in AS-OCT can be objective biomarkers for assessing guttae and FECD severity from early to end-stage FECD.

Systematic Review of the Diagnostic Criteria and Severity Classification for Fuchs Endothelial Corneal Dystrophy

PURPOSE

There are no defined diagnostic criteria and severity classification for Fuchs endothelial corneal dystrophy (FECD), which are required for objective standardized assessments. Therefore, we performed a systematic literature review of the current diagnosis and severity classification of FECD.

METHODS

We searched the Ovid MEDLINE and Web of Science databases for studies published until January 13, 2021. We excluded review articles, conference abstracts, editorials, case reports with <5 patients, and letters.

RESULTS

Among 468 articles identified, we excluded 173 and 165 articles in the first and second screenings, respectively. Among the 130 included articles, 61 (47%) and 99 (76%) mentioned the diagnostic criteria for FECD and described its severity classification, respectively. Regarding diagnosis, slitlamp microscope alone was the most frequently used device in 31 (51%) of 61 articles. Regarding diagnostic findings, corneal guttae alone was the most common parameter [adopted in 23 articles (38%)]. Regarding severity classification, slitlamp microscopes were used in 88 articles (89%). The original or modified Krachmer grading scale was used in 77 articles (78%), followed by Adami's classification in six (6%). Specular microscopes or Scheimpflug tomography were used in four articles (4%) and anterior segment optical coherence tomography in one (1%).

CONCLUSIONS

FECD is globally diagnosed by the corneal guttae using slitlamp examination, and its severity is predominantly determined by the original or modified Krachmer grading scale. Objective severity grading using Scheimpflug or anterior segment optical coherence tomography can be applied in the future innovative therapies such as cell injection therapy or novel small molecules.

Guidelines for the application of artificial intelligence in the diagnosis of anterior segment diseases (2023)

The landscape of ophthalmology has observed monumental shifts with the advent of artificial intelligence (AI) technologies. This article is devoted to elaborating on the nuanced application of AI in the diagnostic realm of anterior segment eye diseases, an area ripe with potential yet complex in its imaging characteristics. Historically, AI's entrenchment in ophthalmology was predominantly rooted in the posterior segment. However, the evolution of machine learning paradigms, particularly with the advent of deep learning methodologies, has reframed the focus. When combined with the exponential surge in available electronic image data pertaining to the anterior segment, AI's role in diagnosing corneal, conjunctival, lens, and eyelid pathologies has been solidified and has emerged from the realm of theoretical to practical. In light of this transformative potential, collaborations between the Ophthalmic Imaging and Intelligent Medicine Subcommittee of the China Medical Education Association and the Ophthalmology Committee of the International Translational Medicine Association have been instrumental. These eminent bodies mobilized a consortium of experts to dissect and assimilate advancements from both national and international quarters. Their mandate was not limited to AI's application in anterior segment pathologies like the cornea, conjunctiva, lens, and eyelids, but also ventured into deciphering the existing impediments and envisioning future trajectories. After iterative deliberations, the consensus synthesized herein serves as a touchstone, assisting ophthalmologists in optimally integrating AI into their diagnostic decisions and bolstering clinical research. Through this guideline, we aspire to offer a comprehensive framework, ensuring that clinical decisions are not merely informed but transformed by AI. By building upon existing literature yet maintaining the highest standards of originality, this document stands as a testament to both innovation and academic integrity, in line with the ethos of renowned journals such as Ophthalmology.

Artificial intelligence-assisted diagnosis of ocular surface diseases

With the rapid development of computer technology, the application of artificial intelligence (AI) in ophthalmology research has gained prominence in modern medicine. Artificial intelligence-related research in ophthalmology previously focused on the screening and diagnosis of fundus diseases, particularly diabetic retinopathy, age-related macular degeneration, and glaucoma. Since fundus images are relatively fixed, their standards are easy to unify. Artificial intelligence research related to ocular surface diseases has also increased. The main issue with research on ocular surface diseases is that the images involved are complex, with many modalities. Therefore, this review aims to summarize current artificial intelligence research and technologies used to diagnose ocular surface diseases such as pterygium, keratoconus, infectious keratitis, and dry eye to identify mature artificial intelligence models that are suitable for research of ocular surface diseases and potential algorithms that may be used in the future.

Deep Learning Model for the Detection of Corneal Edema Before Descemet Membrane Endothelial Keratoplasty on Optical Coherence Tomography Images

Purpose

Descemet membrane endothelial keratoplasty (DMEK) is the preferred method for treating corneal endothelial dysfunction, such as Fuchs endothelial corneal dystrophy (FECD). The surgical indication is based on the patients' symptoms and the presence of corneal edema. We developed an automated tool based on deep learning to detect edema in corneal optical coherence tomography images. This study aimed to evaluate this approach in edema detection before Descemet membrane endothelial keratoplasty surgery, for patients with or without FECD.

Methods

We used our previously described model allowing to classify each pixel in the corneal optical coherence tomography images as "normal" or "edema." We included 1992 images of normal and preoperative edematous corneas. We calculated the edema fraction (EF), defined as the ratio between the number of pixels labeled as "edema," and those representing the cornea for each patient. Differential central corneal thickness (DCCT), defined as the difference in central corneal thickness before and 6 months after surgery, was used to quantify preoperative edema. AUC of EF for the edema detection was calculated for Several DCCT thresholds and a value of 20 µm was selected to define significant edema as it provided the highest area under the curve value.

Results

The area under the curve of the receiver operating characteristic curve for EF for the detection of 20 µm of DCCT was 0.97 for all patients, 0.96 for Fuchs and normal only and 0.99 for non-FECD and normal patients. The optimal EF threshold was 0.143 for all patients and patients with FECD.

Conclusions

Our model is capable of objectively detecting minimal corneal edema before Descemet membrane endothelial keratoplasty surgery.

Translational Relevance

Deep learning can help to interpret optical coherence tomography scans and aid the surgeon in decision-making.

Everything real about unreal artificial intelligence in diabetic retinopathy and in ocular pathologies

Artificial Intelligence is a multidisciplinary field with the aim of building platforms that can make machines act, perceive, reason intelligently and whose goal is to automate activities that presently require human intelligence. From the cornea to the retina, artificial intelligence (AI) is expected to help ophthalmologists diagnose and treat ocular diseases. In ophthalmology, computerized analytics are being viewed as efficient and more objective ways to interpret the series of images and come to a conclusion. AI can be used to diagnose and grade diabetic retinopathy, glaucoma, age-related macular degeneration, cataracts, IOL power calculation, retinopathy of prematurity and keratoconus. This review article intends to discuss various aspects of artificial intelligence in ophthalmology.

Artificial intelligence and corneal diseases

PURPOSE OF REVIEW

Artificial intelligence has advanced rapidly in recent years and has provided powerful tools to aid with the diagnosis, management, and treatment of ophthalmic diseases. This article aims to review the most current clinical artificial intelligence applications in anterior segment diseases, with an emphasis on microbial keratitis, keratoconus, dry eye syndrome, and Fuchs endothelial dystrophy.

RECENT FINDINGS

Most current artificial intelligence approaches have focused on developing deep learning algorithms based on various imaging modalities. Algorithms have been developed to detect and differentiate microbial keratitis classes and quantify microbial keratitis features. Artificial intelligence may aid with early detection and staging of keratoconus. Many advances have been made to detect, segment, and quantify features of dry eye syndrome and Fuchs. There is significant variability in the reporting of methodology, patient population, and outcome metrics.

SUMMARY

Artificial intelligence shows great promise in detecting, diagnosing, grading, and measuring diseases. There is a need for standardization of reporting to improve the transparency, validity, and comparability of algorithms.

Global Prevalence of Fuchs Endothelial Corneal Dystrophy (FECD) in Adult Population: A Systematic Review and Meta-Analysis

Purpose. To evaluate the global prevalence of Fuchs endothelial corneal dystrophy (FECD). Design. Systematic review and meta-analysis. Methods. A systematic electronic literature search was conducted on PubMed/MedLine, Cochrane Library, and Google Scholar, in order to select papers analysing the prevalence rate of FECD. Two authors independently conducted the electronic search. After removal of duplicates, title and abstract screening, and full-text analysis, data from selected articles were archived in a customized Excel spreadsheet. Risk of bias assessment was performed using the Joanna Briggs Institute Prevalence Critical Appraisal Tool. Meta-analysis was conducted using R (version 1.4.1106, package “meta”). Results. A total of 6660 eligible articles were retrieved from the initial electronic search. Only 4 original works were included in the qualitative and quantitative analysis. Of the 4746 patients included in this meta-analysis (i.e., 2232 male (M) and 2322 female (F)), we retrieved 269 FECD cases (81 M; 188 F), with a pooled prevalence estimates of 7.33% (95% CI: 4.08–12.8%). Statistically significant gender-related differences in the prevalence of FECD emerged by the analysis (OR: 2.22; 95% CI: 1.66–2.96, $p=0.0016$ ). While the total number of people aged >30 years with FECD is nowadays estimated at nearly 300 million, an increase of 41.7% in the number of FECD-affected patients is expected by 2050, when the overall figure is supposed to rise up to 415 million. Conclusion. This study provides a reliable figure of the present and future epidemiological burden of FECD globally, which might be useful for the design of FECD screening, treatment, rehabilitation, and related public health strategies.

Deep Learning Automated Diagnosis and Quantitative Classification of Cataract Type and Severity

PURPOSE

To develop and evaluate deep learning models to perform automated diagnosis and quantitative classification of age-related cataract, including all three anatomical types, from anterior segment photographs.

DESIGN

Application of deep learning models to Age-Related Eye Disease Study (AREDS) dataset.

PARTICIPANTS

18,999 photographs (6,333 triplets) from longitudinal follow-up of 1,137 eyes (576 AREDS participants).

METHODS

Deep learning models were trained to detect and quantify nuclear cataract (NS; scale 0.9-7.1) from 45-degree slit-lamp photographs and cortical (CLO; scale 0-100%) and posterior subcapsular (PSC; scale 0-100%) cataract from retroillumination photographs. Model performance was compared with that of 14 ophthalmologists and 24 medical students. The ground truth labels were from reading center grading.

MAIN OUTCOME MEASURES

Mean squared error (MSE).

RESULTS

On the full test set, mean MSE values for the deep learning models were: 0.23 (SD 0.01) for NS, 13.1 (SD 1.6) for CLO, and 16.6 (SD 2.4) for PSC. On a subset of the test set (substantially enriched for positive cases of CLO and PSC), for NS, mean MSE for the models was 0.23 (SD 0.02), compared to 0.98 (SD 0.23; p=0.000001) for the ophthalmologists, and 1.24 (SD 0.33; p=0.000005) for the medical students. For CLO, mean MSE values were 53.5 (SD 14.8), compared to 134.9 (SD 89.9; p=0.003) and 422.0 (SD 944.4; p=0.0007), respectively. For PSC, mean MSE values were 171.9 (SD 38.9), compared to 176.8 (SD 98.0; p=0.67) and 395.2 (SD 632.5; p=0.18), respectively. In external validation on the Singapore Malay Eye Study (sampled to reflect the distribution of cataract severity in AREDS), MSE was 1.27 for NS and 25.5 for PSC.

CONCLUSIONS

A deep learning framework was able to perform automated and quantitative classification of cataract severity for all three types of age-related cataract. For the two most common types (NS and CLO), the accuracy was significantly superior to that of ophthalmologists; for the least common type (PSC), the accuracy was similar. The framework may have wide potential applications in both clinical and research domains. In the future, such approaches may increase the accessibility of cataract assessment globally. The code and models are publicly available at https://XXX.

Separate Detection of Stromal and Epithelial Corneal Edema on Optical Coherence Tomography Using a Deep Learning Pipeline and Transfer Learning

The accurate detection of corneal edema has become a topic of growing interest with the generalization of endothelial keratoplasty. Despite recent advances in deep learning for corneal edema detection, the problem of minimal edema remains challenging. Using transfer learning and a limited training set of 11 images, we built a model to segment the corneal epithelium, which is part of a three-model pipeline to detect corneal edema. A second and a third model are used to detect edema on the stroma alone and on the epithelium. A validation set of 233 images from 30 patients consisting of three groups (Normal, Minimal Edema and important Edema) was used to compare the results of our new pipeline to our previous model. The mean edema fraction (EF), defined as the number of pixels detected as edema divided by the total number of pixels of the cornea, was calculated for each image. With our previous model, the mean EF was not statistically different between the Normal and Minimal Edema groups (p = 0.24). With the current pipeline, the mean EF was higher in the Minimal Edema group compared to the Normal group (p < 0.01). The described pipeline constitutes an adjustable framework for the detection of corneal edema based on optical coherence tomography and yields better performances in cases of minimal or localized edema.

Artificial Intelligence in Cornea, Refractive Surgery, and Cataract: Basic Principles, Clinical Applications, and Future Directions

ABSTRACT

Corneal diseases, uncorrected refractive errors, and cataract represent the major causes of blindness globally. The number of refractive surgeries, either cornea- or lens-based, is also on the rise as the demand for perfect vision continues to increase. With the recent advancement and potential promises of artificial intelligence (AI) technologies demonstrated in the realm of ophthalmology, particularly retinal diseases and glaucoma, AI researchers and clinicians are now channeling their focus toward the less explored ophthalmic areas related to the anterior segment of the eye. Conditions that rely on anterior segment imaging modalities, including slit-lamp photography, anterior segment optical coherence tomography, corneal tomography, in vivo confocal microscopy and/or optical biometers, are the most commonly explored areas. These include infectious keratitis, keratoconus, corneal grafts, ocular surface pathologies, preoperative screening before refractive surgery, intraocular lens calculation, and automated refraction, among others. In this review, we aimed to provide a comprehensive update on the utilization of AI in anterior segment diseases, with particular emphasis on the recent advancement in the past few years. In addition, we demystify some of the basic principles and terminologies related to AI, particularly machine learning and deep learning, to help improve the understanding, research and clinical implementation of these AI technologies among the ophthalmologists and vision scientists. As we march toward the era of digital health, guidelines such as CONSORT-AI, SPIRIT-AI, and STARD-AI will play crucial roles in guiding and standardizing the conduct and reporting of AI-related trials, ultimately promoting their potential for clinical translation.

Digital Image Processing and Development of Machine Learning Models for the Discrimination of Corneal Pathology: An Experimental Model

Machine learning (ML) has an impressive capacity to learn and analyze a large volume of data. This study aimed to train different algorithms to discriminate between healthy and pathologic corneal images by evaluating digitally processed spectral-domain optical coherence tomography (SD-OCT) corneal images. A set of 22 SD-OCT images belonging to a random set of corneal pathologies was compared to 71 healthy corneas (control group). A binary classification method was applied where three approaches of ML were explored. Once all images were analyzed, representative areas from every digital image were also extracted, processed and analyzed for a statistical feature comparison between healthy and pathologic corneas. The best performance was obtained from transfer learning—support vector machine (TL-SVM) (AUC = 0.94, SPE 88%, SEN 100%) and transfer learning—random forest (TL- RF) method (AUC = 0.92, SPE 84%, SEN 100%), followed by convolutional neural network (CNN) (AUC = 0.84, SPE 77%, SEN 91%) and random forest (AUC = 0.77, SPE 60%, SEN 95%). The highest diagnostic accuracy in classifying corneal images was achieved with the TL-SVM and the TL-RF models. In image classification, CNN was a strong predictor. This pilot experimental study developed a systematic mechanized system to discern pathologic from healthy corneas using a small sample.