Evaluation of Mooren's corneal ulcer by anterior segment optical coherence tomography.

BACKGROUND: This study evaluates the use of anterior segment optical coherence tomography (AS-OCT) to identify focal changes and inform surgical plans in eyes with Mooren's ulcer. METHODS: A total of 18 eyes of 17 patients with Mooren's ulcer were examined prospectively using the AS-OCT system. RESULTS: Optical hyperreflectivity noted on AS-OCT images was in accordance with corneal ulceration, neovascularization, fibrovascular membranes, the junction of the native stromal bed, and the overlying lamellar corneal grafts. Focal corneal ectasia was observed in 13 eyes with a decrease in corneal thickness to ≤0.39 mm. There was a cut-off value of 0.39 mm in corneal thickness between the eyes with and without focal corneal ectasia in the thinned corneal area (Fisher = 0.383, χ2 = 14.873, P = 0.000). Based on the AS-OCT findings, six eyes were subjected to an individualized lamellar corneal graft. The thickness of the residual cornea after surgery was 47 ± 34 µm less than the presumed healthy corneal thickness before surgery (t = 3.376, P = 0.02). A small corneal perforation covered by a pseudopterygium in Mooren's ulcer was found through AS-OCT but undetectable by slit-lamp biomicroscopy. CONCLUSIONS: AS-OCT is a valuable non-contact technique for monitoring corneal thinning in Mooren's ulcer, and assisting surgical design. A decrease in peripheral corneal thickness to ≤0.39 mm may cause focal corneal ectasia.

Differentiation of Active Corneal Infections from Healed Scars Using Deep Learning.

PURPOSE: To develop and evaluate an automated, portable algorithm to differentiate active corneal ulcers from healed scars using only external photographs. DESIGN: A convolutional neural network was trained and tested using photographs of corneal ulcers and scars. PARTICIPANTS: De-identified photographs of corneal ulcers were obtained from the Steroids for Corneal Ulcers Trial (SCUT), Mycotic Ulcer Treatment Trial (MUTT), and Byers Eye Institute at Stanford University. METHODS: Photographs of corneal ulcers (n = 1313) and scars (n = 1132) from the SCUT and MUTT were used to train a convolutional neural network (CNN). The CNN was tested on 2 different patient populations from eye clinics in India (n = 200) and the Byers Eye Institute at Stanford University (n = 101). Accuracy was evaluated against gold standard clinical classifications. Feature importances for the trained model were visualized using gradient-weighted class activation mapping. MAIN OUTCOME MEASURES: Accuracy of the CNN was assessed via F1 score. The area under the receiver operating characteristic (ROC) curve (AUC) was used to measure the precision-recall trade-off. RESULTS: The CNN correctly classified 115 of 123 active ulcers and 65 of 77 scars in patients with corneal ulcer from India (F1 score, 92.0% [95% confidence interval (CI), 88.2%-95.8%]; sensitivity, 93.5% [95% CI, 89.1%-97.9%]; specificity, 84.42% [95% CI, 79.42%-89.42%]; ROC: AUC, 0.9731). The CNN correctly classified 43 of 55 active ulcers and 42 of 46 scars in patients with corneal ulcers from Northern California (F1 score, 84.3% [95% CI, 77.2%-91.4%]; sensitivity, 78.2% [95% CI, 67.3%-89.1%]; specificity, 91.3% [95% CI, 85.8%-96.8%]; ROC: AUC, 0.9474). The CNN visualizations correlated with clinically relevant features such as corneal infiltrate, hypopyon, and conjunctival injection. CONCLUSIONS: The CNN classified corneal ulcers and scars with high accuracy and generalized to patient populations outside of its training data. The CNN focused on clinically relevant features when it made a diagnosis. The CNN demonstrated potential as an inexpensive diagnostic approach that may aid triage in communities with limited access to eye care.

Ultrasound-Enhanced Transcorneal Drug Delivery for Treatment of Fungal Keratitis.

PURPOSE: Transcorneal drug delivery is hindered by ocular physical and biochemical properties, such as tear production, the epithelial layer of the cornea, and blinking. The aim of this study was to determine whether ultrasound can be applied to increase the transcorneal drug delivery of natamycin used in the treatment of fungal keratitis without dangerously overheating the surrounding ocular tissues. METHODS: To verify the safety of various sets of ultrasound parameters, modeling studies were conducted using OnScale, an ultrasonic wave modeling software. Ultrasound parameters determined optimal for ocular tissue safety were used in a laboratory setting in a jacketed Franz diffusion cell setup. Histological images of the cross-section of the corneas used in experiments were examined for cell damage under a microscope. RESULTS: Increases in transcorneal drug delivery were seen in every treatment parameter combination when compared with the sham treatment. The highest increase was 4.0 times for 5 minutes of pulsed ultrasound at a 25% duty cycle and a frequency of 400 kHz and an intensity of 0.5 W/cm 2 with statistical significance ( P < 0.001). Histological analysis revealed structural damage only in the corneal epithelium, with most damage being at the epithelial surface. CONCLUSIONS: This study suggests that ultrasound is a safe, effective, and minimally invasive treatment method for enhancing the transcorneal drug delivery of natamycin. Further research is needed into the long-term effects of ultrasound parameters used in this study on human ocular tissues.

Treatment of spontaneous corneal perforation secondary to undiagnosed Sjögren's syndrome using regenerating agent and autologous serum eye drops.

PURPOSE: The aim of this study was to report a case of sterile corneal ulcer leading to perforation, which was treated effectively with autologous serum eye drops, topical regenerative agent (poly-carboxymethylglucose sulfate), steroids, and systemic immunosuppression in a patient with undiagnosed primary Sjögren's syndrome. METHODS: A 74-year-old female presented with a month's history of gradually worsening blurry vision in her left eye. Ophthalmic examination revealed a central descemetocele with excessive corneal stromal melting and absence of signs of infection. A bandage contact lens was applied for tectonic support along with topical corticosteroid and antibiotic drops. Autoimmune screen disclosed a diagnosis of Sjögren's syndrome, and the patient was commenced on systemic immunosuppression. Forty-eight hours after presentation, the patient developed a localized corneal perforation, presenting with a flat anterior chamber. RESULTS: Urgent amniotic membrane transplantation was arranged while topical dexamethasone, moxifloxacin, and autologous serum eye drops were administered. After 24 h of intensive topical treatment, a significant reforming of the anterior chamber and subsequent gradual regeneration of the corneal stroma were noted, thus postponing amniotic grafting. The patient remained under close monitoring, showing progressive clinical improvement. Regenerating agent eye drops (Cacicol20®) were also applied over the next month, with careful and slow tapering of topical dexamethasone. Further improvement of corneal thickness was observed, and visual acuity increased to 20/80. CONCLUSION: This case report demonstrates the successful medical treatment of an autoimmune-related sterile corneal perforation without surgical intervention, highlighting the fact that early diagnosis and rigorous medical treatment with autologous serum and regenerating agent eye drops can effectively aid tissue regeneration and favorable visual rehabilitation.

Endothelial Biopsy for the Diagnosis and Management of Culture-Negative Retrocorneal Fungal Keratitis With the Assistance of Optical Coherence Tomography Imaging.

PURPOSE: To report a case of a culture-negative deep fungal corneal infection that was diagnosed after histopathology of an anterior segment optical coherence tomography-guided endothelial biopsy. METHODS: A 22-year-old woman with history of contact lens wear and concomitant topical steroid use presented with a mid-stromal corneal infiltrate that failed to respond to oral acyclovir and topical fortified antibiotics. Although cornea stains, cultures, and confocal microscopy showed negative results, there was high clinical suspicion for fungal keratitis. After 2 months on topical natamycin, oral voriconazole, and serial intrastromal and intracameral voriconazole injections, the infiltrate enlarged and deepened. Imaging with anterior segment optical coherence tomography revealed that the infection had progressed to an endothelial plaque. RESULTS: Diagnostic endothelial biopsy was performed in the operating room. Cultures showed again negative results, whereas histopathology of the removed specimen revealed fungal elements. The postoperative edema at the site of the biopsy resolved over the course of 4 weeks, and a posterior stromal scar formed. Serial intrastromal and intracameral voriconazole injections were continued for the first postoperative month. At the 1-year and the 3-year follow-up examinations, the patient's vision was 20/20 without recurrence. CONCLUSIONS: Intraoperative scraping of the endothelial plaque and histopathologic evaluation of the specimen proved to be of utmost importance for definitive diagnosis and resolution of the culture-negative deep fungal infection in this case. This young patient's cornea was retained and vision remains excellent.

Pediatric Use of Recombinant Human Nerve Growth Factor 20 µg/mL Eye Drops (Cenegermin) for Bilateral Neurotrophic Keratopathy in Congenital Corneal Anesthesia.

PURPOSE: This study aimed to present the efficacy and safety of cenegermin eye drop (Oxervate; Dompè Farmaceutici, Milan, Italy) treatment in a pediatric patient affected by neurotrophic keratopathy (NK) with Goldenhar syndrome. METHODS: This case reports an infant presenting ulceration and a small central opacity in the cornea of the right and left eyes, respectively. The NK bilaterally worsened despite the use of therapeutic contact lenses and temporary partial tarsorrhaphy. Magnetic resonance imaging showed absence and hypoplasia of the right and left trigeminal nerves, respectively. Cenegermin eye drops were administered 1 drop/each eye, 6 times daily for 8 weeks to promote corneal healing. RESULTS: Complete healing was achieved in both eyes after treatment. During the 16-month follow-up period, no epithelial defect, recurrence, or complications were noticed, whereas corneal opacities progressively became clearer, although insignificant improvements in corneal sensitivity or in the reflex tearing were observed. CONCLUSIONS: Cenegermin was effective in treating NK in an infant with Goldenhar syndrome.

A deep learning approach in diagnosing fungal keratitis based on corneal photographs.

Fungal keratitis (FK) is the most devastating and vision-threatening microbial keratitis, but clinical diagnosis a great challenge. This study aimed to develop and verify a deep learning (DL)-based corneal photograph model for diagnosing FK. Corneal photos of laboratory-confirmed microbial keratitis were consecutively collected from a single referral center. A DL framework with DenseNet architecture was used to automatically recognize FK from the photo. The diagnoses of FK via corneal photograph for comparing DL-based models were made in the Expert and NCS-Oph group through a majority decision of three non-corneal specialty ophthalmologist and three corneal specialists, respectively. The average percentage of sensitivity, specificity, positive predictive value, and negative predictive value was approximately 71, 68, 60, and 78. The sensitivity was higher than that of the NCS-Oph (52%, P < .01), whereas the specificity was lower than that of the NCS-Oph (83%, P < .01). The average accuracy of around 70% was comparable with that of the NCS-Oph. Therefore, the sensitive DL-based diagnostic model is a promising tool for improving first-line medical care at rural area in early identification of FK.

Anterior segment optical coherence tomography imaging in peripheral ulcerative keratitis, a corneal structural description.

BACKGROUND: Anterior segment optical coherence tomography (AS OCT) is a helpful tool used to diagnose and manage many corneal conditions, but its use has not been reported in case of peripheral ulcerative keratitis (PUK). The aim of this study is to describe AS OCT findings in cases of PUK. METHODS: Retrospective observational case series of six eyes presenting with a PUK and proven systemic vasculitis. Clinical course, slit lamp photographs, and AS OCT findings were the main outcomes. RESULTS: The AS OCT findings were found to correlate with the ocular disease's level of activity. In the acute stage, an absence of corneal epithelium, a scrambled appearance of the anterior stroma and a heterogeneous stromal reflectivity were observed. During the reduction of disease level activity, an irregular hyporeflective epithelium, a smoother anterior stroma, and a homogenous hyperreflective stroma were seen. At the healed stage, a filling of the corneal defect by a hyporeflective thick epithelium, the persistence of the hyperreflective underlying stroma, and a demarcation line were observed. The mean total corneal thickness at last follow-up was significantly thicker (509 ± 147 µm) compared with the mean corneal thickness at onset (408 ± 131 µm; P = 0.03). CONCLUSIONS: AS OCT provides an assessment of structural changes occurring in PUK, useful for its diagnosis and monitoring.

The SUSTech-SYSU dataset for automatically segmenting and classifying corneal ulcers.

Corneal ulcer is a common ophthalmic symptom. Segmentation algorithms are needed to identify and quantify corneal ulcers from ocular staining images. Developments of such algorithms have been obstructed by a lack of high quality datasets (the ocular staining images and the corresponding gold-standard ulcer segmentation labels), especially for supervised learning based segmentation algorithms. In such context, we prepare a dataset containing 712 ocular staining images and the associated segmentation labels of flaky corneal ulcers. In addition to segmentation labels for flaky corneal ulcers, we also provide each image with three-fold class labels: firstly, each image has a label in terms of its general ulcer pattern; secondly, each image has a label in terms of its specific ulcer pattern; thirdly, each image has a label indicating its ulcer severity degree. This dataset not only provides an excellent opportunity for investigating the accuracy and reliability of different segmentation and classification algorithms for corneal ulcers, but also advances the development of new supervised learning based algorithms especially those in the deep learning framework.

Automatic Corneal Ulcer Segmentation Combining Gaussian Mixture Modeling and Otsu Method.

In this paper, we proposed and validated a novel and accurate pipeline for automatically segmenting flaky corneal ulcer areas from fluorescein staining images. The ulcer area was segmented within the cornea by employing a joint method of Otsu and Gaussian Mixture Modeling (GMM). In the GMM based segmentation, the total number of Gaussians was determined intelligently using an information theory based algorithm. And the fluorescein staining images were processed in the HSV color model rather than the original RGB color model, aiming to improve the segmentation results' robustness and accuracy. In the Otsu based segmentation, the images were processed in the grayscale space with Gamma correction being conducted before the Otsu binarization. Afterwards, morphological operations and median filtering were employed to further improve the Otsu segmentation result. The GMM and Otsu segmentation results were then intersected, for which post-processing was conducted by identifying and filling holes through a fast algorithm using priority queues of pixels. The proposed pipeline has been validated on a total of 150 clinical images. Accurate ulcer segmentation results have been obtained, with the mean Dice Similarity Coefficient (DSC) being 0.88 when comparing the automatic segmentation result with the manually-delineated gold standard. For images in the RGB color space, the mean DSC was 0.83, being much lower than that of the images in the HSV color space.

In Vivo Confocal Microscopy Cellular Features of Host and Organism in Bacterial, Fungal, and Acanthamoeba Keratitis.

PURPOSE: To determine cellular features of fungal (FK), Acanthamoeba (AK), and bacterial keratitis (BK) using HRT3 in vivo confocal microscopy (IVCM). DESIGN: Prospective observational cross-sectional study. METHODS: Eligible participants were adults with microbiologically positive FK, AK, or BK, of size ≥ 3 mm, attending Aravind Eye Hospital from February 2012 to February 2013. Exclusion criteria were descemetocele or perforation. At presentation, IVCM imaging was performed, then corneal scrapes were obtained for culture/light microscopy. An experienced grader (masked to microbiology/clinical features) assessed IVCM images for presence/absence of normal keratocyte-like morphology, stellate interconnected cells with/without visible nuclei, dendritiform cells (DFCs), inflammatory cells in a honeycomb distribution, and organism features. Statistical significance was assessed by logistic regression, adjusted for age, sex, ulcer size, and symptom duration. Main outcome measures were presence/absence of IVCM features in FK, AK, BK. RESULTS: A total of 183 participants had FK, 18 AK, 17 BK. Acanthamoeba appeared as bright spots (16/18, 89%), double-walled cysts (15/18, 83%), or signet rings (3/18, 17%), and often formed clusters after topical steroid use (univariable odds ratio [OR] 9.98, 95% confidence interval [CI] 1.02-97.96, P = .048). BK was associated with bullae in anterior stroma (OR 9.99, 95% CI: 3.11-32.06, P < .001). Honeycomb distribution of anterior stromal inflammatory cells was associated with FK (univariable OR 2.74, 95% CI: 1.01-7.40, P = .047). Aspergillus ulcers were associated with stromal DFCs (OR 11.05, 95% CI: 1.49-82.13, P = .019) and Fusarium ulcers with stellate appearance of interconnected cell processes with nuclei (OR 0.24, 95% CI: 0.09-0.65, P = .005). CONCLUSION: Specific cellular and structural features observed using IVCM in microbial keratitis may be associated with organism.

Successful management of bilateral periorbital necrotising fasciitis with ocular involvement.

A 4-year-old girl had suffered from multiple pustules and severely swollen eyelids for a week which led to difficulties in opening her eyes. The pustules burst 2 days prior to admission, causing blood and pus to come out. As a result, the evaluation of eye movement, visual acuity and intraocular pressure was hindered. Eyelids were spasming, hyperaemic, warm and of soft consistency on palpation. Crust and necrotic tissues were also observed. The patient had a history of fever with cough and nasal congestion. CT scan revealed soft tissue thickening on bilateral anterolateral periorbital region. She also underwent debridement in both eyelids, followed by partial periosteal graft on the left eye due to corneal perforation. During follow-up after 1.5 months, she also underwent excision of the periosteal graft and synechiolysis, followed by cataract extraction of the left eye.

Factors Influencing the Diagnostic Accuracy of Laser-Scanning In Vivo Confocal Microscopy for Acanthamoeba Keratitis.

PURPOSE: To determine the factors that influence the sensitivity and specificity of laser-scanning in vivo confocal microscopy (IVCM) for diagnosing Acanthamoeba keratitis (AK). METHODS: This retrospective, controlled study included 28 eyes of 27 patients with AK and 34 eyes of 34 patients with bacterial keratitis (as the control group). All patients had undergone corneal imaging with a laser-scanning IVCM (Heidelberg Retina Tomograph 3 with the Rostock Cornea Module). The IVCM images were independently evaluated by 2 experienced and 2 inexperienced masked observers. Sensitivity and specificity of IVCM for diagnosing AK and the effects of various clinical and imaging parameters on the sensitivity were then investigated. RESULTS: Overall, IVCM had average sensitivity and specificity of 69.7% ± 2.5% and 97.1% ± 4.2% for experienced observers and 59.0% ± 7.6% and 92.7% ± 10.4% for inexperienced observers, respectively. However, the sensitivity did not show any significant association with the duration of disease, size of ulcer, depth of involvement, culture results, or cyst morphology. Although interobserver agreement was good (κ = 0.60, P < 0.001) for the experienced observers, it was only at a moderate level (κ = 0.48, P < 0.001) for the inexperienced observers. CONCLUSIONS: IVCM has a moderate sensitivity and a high specificity for diagnosis of AK. Although clinical parameters do not affect this diagnostic accuracy, a higher sensitivity is seen when images are interpreted by experienced observers.

Novel Image-Based Analysis for Reduction of Clinician-Dependent Variability in Measurement of the Corneal Ulcer Size.

PURPOSE: To assess variability in corneal ulcer measurements between ophthalmologists and reduce clinician-dependent variability using semiautomated segmentation of the ulcer from photographs. METHODS: Three ophthalmologists measured 50 patients' eyes for epithelial defects (EDs) and the stromal infiltrate (SI) size using slit-lamp (SL) calipers. SL photographs were obtained. An algorithm was developed for semiautomatic segmenting of the ED and SI in the photographs. Semiautomatic segmentation was repeated 3 times by different users (2 ophthalmologists and 1 trainee). Clinically significant variability was assessed with intraclass correlation coefficients (ICCs) and the percentage of pairwise measurements differing by ≥0.5 mm. Semiautomatic segmentation measurements were compared with manual delineation of the image by a corneal specialist (gold standard) using Dice similarity coefficients. RESULTS: Ophthalmologists' reliability in measurements by SL calipers had an ICC from 0.84 to 0.88 between examiners. Measurements by semiautomatic segmentation had an ICC from 0.96 to 0.98. SL measures of ulcers by clinical versus semiautomatic segmentation measures differed by ≥0.5 mm in 24% to 38% versus 8% to 28% (ED height); 30% to 52% versus 12% to 34% (ED width); 26% to 38% versus 10% to 32% (SI height); and 38% to 58% versus 14% to 34% (SI width), respectively. Average Dice similarity coefficients between manual and repeated semiautomatic segmentation ranged from 0.83 to 0.86 for the ED and 0.78 to 0.83 for the SI. CONCLUSIONS: Variability exists when measuring corneal ulcers, even among ophthalmologists. Photography and computerized methods for quantifying the ulcer size could reduce variability while remaining accurate and impact quantitative measurement endpoints.

In vivo confocal microscopy appearance of Fusarium and Aspergillus species in fungal keratitis.

BACKGROUND: Clinical outcomes in fungal keratitis vary between Fusarium and Aspergillus spp, therefore distinguishing between species using morphological features such as filament branching angles, sporulation along filaments (adventitious sporulation) or dichotomous branching may be useful. In this study, we assessed these three features within Heidelberg Retina Tomograph 3 in vivo confocal microscopy (IVCM) images from culture-positive Fusarium and Aspergillus spp keratitis participants. METHODS: Prospective observational cohort study in Aravind Eye Hospital (February 2011-February 2012). Eligibility criteria: age ≥18 years, stromal infiltrate ≥3 mm diameter, Fusarium or Aspergillus spp culture-positive. EXCLUSION CRITERIA: previous/current herpetic keratitis, visual acuity <6/60 in fellow eye, >80% corneal thinning. IVCM was performed and images analysed for branch angle, presence/absence of adventitious sporulation or dichotomous branching by a grader masked to the microbiological diagnosis. RESULTS: 98 participants were included (106 eligible, 8 excluded as no measurable branch angles); 68 were positive for Fusarium spp, 30 for Aspergillus spp. Mean branch angle for Fusarium spp was 59.7° (95% CI 57.7° to 61.8°), and for Aspergillus spp was 63.3° (95% CI 60.8° to 65.8°), p=0.07. No adventitious sporulation was detected in Fusarium spp ulcers. Dichotomous branching was detected in 11 ulcers (7 Aspergillus spp, 4 Fusarium spp). CONCLUSIONS: There was very little difference in the branching angle of Fusarium and Aspergillus spp. Adventitious sporulation was not detected and dichotomous branching was infrequently seen. Although IVCM remains a valuable tool to detect fungal filaments in fungal keratitis, it cannot be used to distinguish Fusarium from Aspergillus spp and culture remains essential to determine fungal species.

Management of severe and refractory Mooren's ulcers with rituximab.

PURPOSE: Management of severe and refractory Mooren's ulcers is challenging as it encompasses tectonic surgical treatment and aggressive immunosuppressive therapies. Efficacy of rituximab in the management of severe Mooren's ulcers has never been reported. METHODS: Five patients (six eyes) from the Cornea and External Disorders department at the Rothschild Ophthalmologic Foundation (Paris, France) were treated for severe Mooren's ulcer unresponsive to conventional treatments between 2008 and 2016. Conventional treatment included topical steroid and ciclosporin 2%, high doses of systemic corticosteroids and/or cyclophosphamide and conjunctival resection with amniotic membrane graft. These patients received two infusions of 1000 mg of rituximab at 2 weeks interval. Epithelial healing, inflammation, additional surgery, systemic corticosteroids and rituximab-related side effects were reported. RESULTS: The mean follow-up was 46.8 months. Following rituximab treatment, we observed a complete healing of Mooren's ulcer within 2 weeks in all patients. Peripheral lamellar keratoplasty was associated when peripheral corneal perforation occurred (5/6 affected corneas). Systemic corticosteroids had been discontinued in all patients. Two recurrences occurred 13 and 53 months after the first rituximab infusion and where successfully treated with a new infusion. No rituximab-related adverse events were reported. CONCLUSIONS: Rituximab was effective in the management of severe Mooren's ulcers and could be an alternative to cyclophosphamide. Additional studies should assess the role of this biotherapy in the management of immunological corneal ulcer.

Orbital cellulitis and corneal ulcer due to Cedecea: First reported case and review of the literature.

Cedecea is a gram-negative bacterium from the family Enterobacteriaceae, rarely associated with human infection. We report the first case of an orbital cellulitis and corneal ulcer due to Cedecea in a patient who sustained a motor vehicle accident and was then found to have a retained wooden orbital foreign body.