PCR for the molecular diagnosis of mycotic keratitis

Class-Aware Attention Network for infectious keratitis diagnosis using corneal photographs

Infectious keratitis is one of the common ophthalmic diseases and also one of the main blinding eye diseases in China, hence rapid and accurate diagnosis and treatment for infectious keratitis are urgent to prevent the progression of the disease and limit the degree of corneal injury. Unfortunately, the traditional manual diagnosis accuracy is usually unsatisfactory due to the indistinguishable visual features. In this paper, we propose a novel end-to-end fully convolutional network, named Class-Aware Attention Network (CAA-Net), for automatically diagnosing infectious keratitis (normal, viral keratitis, fungal keratitis, and bacterial keratitis) using corneal photographs. In CAA-Net, a class-aware classification module is first trained to learn class-related discriminative features using separate branches for each class. Then, the learned class-aware discriminative features are fed into the main branch and fused with other feature maps using two attention strategies to assist the final multi-class classification performance. For the experiments, we have built a new corneal photograph dataset with 1886 images from 519 patients and conducted comprehensive experiments to verify the effectiveness of our proposed method. The code is available at https://github.com/SWF-hao/CAA-Net_Pytorch.

Mycotic Keratitis-A Global Threat from the Filamentous Fungi

Mycotic or fungal keratitis (FK) is a sight-threatening disease, caused by infection of the cornea by filamentous fungi or yeasts. In tropical, low and middle-income countries, it accounts for the majority of cases of microbial keratitis (MK). Filamentous fungi, in particular Fusarium spp., the aspergilli and dematiaceous fungi, are responsible for the greatest burden of disease. The predominant risk factor for filamentous fungal keratitis is trauma, typically with organic, plant-based material. In developed countries, contact lens wear and related products are frequently implicated as risk factors, and have been linked to global outbreaks of Fusarium keratitis in the recent past. In 2020, the incidence of FK was estimated to be over 1 million cases per year, and there is significant geographical variation; accounting for less than 1% of cases of MK in some European countries to over 80% in parts of south and south-east Asia. The proportion of MK cases is inversely correlated to distance from the equator and there is emerging evidence that the incidence of FK may be increasing. Diagnosing FK is challenging; accurate diagnosis relies on reliable microscopy and culture, aided by adjunctive tools such as in vivo confocal microscopy or PCR. Unfortunately, these facilities are infrequently available in areas most in need. Current topical antifungals are not very effective; infections can progress despite prompt treatment. Antifungal drops are often unavailable. When available, natamycin is usually first-line treatment. However, infections may progress to perforation in ~25% of cases. Future work needs to be directed at addressing these challenges and unmet needs. This review discusses the epidemiology, clinical features, diagnosis, management and aetiology of FK.

Deep learning-based automated diagnosis of fungal keratitis with in vivo confocal microscopy images

Background

The aim of this study was to develop an intelligent system based on a deep learning algorithm for automatically diagnosing fungal keratitis (FK) in in vivo confocal microscopy (IVCM) images.

Methods

A total of 2,088 IVCM images were included in the training dataset. The positive group consisted of 688 images with fungal hyphae, and the negative group included 1,400 images without fungal hyphae. A total of 535 images in the testing dataset were not included in the training dataset. Deep Residual Learning for Image Recognition (ResNet) was used to build the intelligent system for diagnosing FK automatically. The system was verified by external validation in the testing dataset using the area under the receiver operating characteristic curve (AUC), accuracy, specificity and sensitivity.

Results

In the testing dataset, 515 images were diagnosed correctly and 20 were misdiagnosed (including 6 with fungal hyphae and 14 without). The system achieved an AUC of 0.9875 with an accuracy of 0.9626 in detecting fungal hyphae. The sensitivity of the system was 0.9186, with a specificity of 0.9834. When 349 diabetic patients were included in the training dataset, 501 images were diagnosed correctly and thirty-four were misdiagnosed (including 4 with fungal hyphae and 30 without). The AUC of the system was 0.9769. The accuracy, specificity and sensitivity were 0.9364, 0.9889 and 0.8256, respectively.

Conclusions

The intelligent system based on a deep learning algorithm exhibited satisfactory diagnostic performance and effectively classified FK in various IVCM images. The context of this deep learning automated diagnostic system can be extended to other types of keratitis.

Fungal keratitis: Pathogenesis, diagnosis and prevention

As a kind of serious, potentially sight-threatening corneal infections with poor prognosis, fungal keratitis can bring a heavy economic burden to patients and seriously affect the quality of life, especially those in developing countries where fungal keratitis is more prevalent. Typical clinical features include immune rings, satellite lesions, pseudopods, hypha moss, hypopyon and endothelial plaques. The ideal therapeutic effects could not be achieved by current treatments for many reasons. Therefore, under the current status, understanding the pathogenesis, early diagnosis and prevention strategies might be of great importance. Here, in this review, we discuss the recent progresses that may advance our understanding of pathogenesis, early diagnosis and prevention of fungal keratitis.

Fungal keratitis: An overview of clinical and laboratory aspects

Mycotic keratitis or keratomycosis is a fungal infection with global distribution. The dominant aetiology of this disease varies based on geographical origin, socioeconomic status, and climatic condition. Generally, Aspergillus spp. and Fusarium spp. are common in tropical and subtropical regions and Candida spp. are dominant in temperate areas. Demonstration of fungal elements in microscopic examination besides the isolation of fungi in culture is the gold standard of laboratory diagnosis. As the culture is a time-consuming procedure, other approaches such as in vivo confocal microscopy which produces real-time imaging of corneal tissue and molecular techniques have been developed to facilitate rapid diagnosis of fungal keratitis. The first choice of treatment is topical natamycin, although topical amphotericin B is the best choice for Aspergillus and Candida keratitis. Regarding the diversity of fungal aetiology and the emergence of drug resistance in some genera and species, proper identification using molecular methods and antifungal susceptibility testing could provide useful data. Furthermore, as the better efficacy of combination therapy in comparison to monotherapy is reported, in vitro determination of interactions between various drugs seem informative. This review aims to provide a general and updated view on the aetiology, risk factors, epidemiology, clinical and laboratory diagnosis, and management of fungal keratitis.

Recent advances in diagnosis and management of Mycotic Keratitis

Mycotic keratitis is a major cause of corneal blindness, especially in tropical and subtropical countries. The prognosis is markedly worse compared to bacterial keratitis. Delayed diagnosis and scarcity of effective antifungal agents are the major factors for poor outcome. Over the last decade, considerable progress has been made to rapidly diagnose cases with mycotic keratitis and increase the efficacy of treatment. This review article discusses the recent advances in diagnosis and management of mycotic keratitis with a brief discussion on rare and emerging organisms. A MEDLINE search was carried out for articles in English language, with the keywords, mycotic keratitis, fungal keratitis, emerging or atypical fungal pathogens in mycotic keratitis, investigations in mycotic keratitis, polymerase chain reaction in mycotic keratitis, confocal microscopy, treatment of mycotic keratitis, newer therapy for mycotic keratitis. All relevant articles were included in this review. Considering the limited studies available on newer diagnostic and therapeutic modalities in mycotic keratitis, case series as well as case reports were also included if felt important.

Sensitivity and Specificity of Laser-Scanning In Vivo Confocal Microscopy for Filamentous Fungal Keratitis: Role of Observer Experience

PURPOSE

To determine sensitivity and specificity of laser-scanning in vivo confocal microscopy (LS-IVCM) for detection of filamentous fungi in patients with microbial keratitis and to evaluate the effect of observer's imaging experience on these parameters.

DESIGN

Retrospective reliability study.

METHODS

This study included 21 patients with filamentous fungal keratitis and 24 patients with bacterial keratitis (as controls). The etiology of infection was confirmed based on the response to specific therapy regardless of culture results. All patients had undergone full-thickness corneal imaging by a LS-IVCM (Heidelberg Retina Tomograph 3 with Rostock Cornea Module; Heidelberg Engineering, Heidelberg, Germany). The images were evaluated for the presence of fungal filaments by 2 experienced observers and 2 inexperienced observers. All observers were masked to the clinical and microbiologic data.

RESULTS

The mean number of images obtained per eye was 917 ± 353. The average sensitivity of LS-IVCM for detecting fungal filaments was 71.4% ± 0% for the experienced observers and 42.9% ± 6.7% for the inexperienced observers. The average specificity was 89.6% ± 3.0% and 87.5% ± 17.7% for these 2 groups of observers, respectively. Although there was a good agreement between the 2 experienced observers (κ = 0.77), the inexperienced observers showed only a moderate interobserver agreement (κ = 0.51). The LS-IVCM sensitivity was higher in patients with fungal infections who had positive culture or longer duration of the disease.

CONCLUSIONS

Although LS-IVCM has a high specificity for diagnosing filamentous fungal keratitis, its sensitivity is moderate and highly dependent on the level of the observer's experience and training with this imaging modality.

Fungal keratitis: study of increasing trend and common determinants

Fungal keratitis is one of the leading cause of ocular morbidity. Fungal keratitis possesses a clinical challenge due to its slow pathologic process, overlapping features, diagnostic difficulty, and potential complications. Its increasing trend can be attributed to the use of contact lens, non-judiciary corticosteroid, and vegetative trauma. Early diagnosis and treatment is the cornerstone for its effective control. Knowledge of pathological course and clinical characteristics of fungal keratitis will definitely add in early diagnosis and treatment, with reduction in ocular morbidity. This review article explores the risk factor of fungal keratitis, its clinical course and management strategy.

Keratitis due to the wood saprobic ascomycete, Auerswaldia lignicola (Family Botryosphaeriaceae), in a carpenter in India

Keratitis due to Auerswaldia lignicola in a 32-year-old Indian male carpenter is described. At presentation, the patient reported persistent pain and tearing (left eye) in spite of topical antimicrobial therapy for more than 3 weeks. Clinically, mycotic keratitis was suspected, and direct microscopy of corneal scrapings stained by lactophenol cotton blue and Gram stains revealed broad septate hyphae. Intensive topical antifungal therapy was then given for 15 days. The keratitis continued to progress, necessitating therapeutic penetrating keratoplasty. Following the keratoplasty, there was rapid reduction in inflammation and gradual quietening of the eye. Brown-black fungal colonies resembling Lasiodiplodia theobromae were isolated from corneal scrape and corneal button (post-surgery) material on Sabouraud glucose-neopeptone agar; however, sporulation did not occur, so the morphological identification could not be confirmed. Sequence analysis of the 18S rRNA region of extracted fungal genomic DNA yielded an identification of A. lignicola Ariyawansa, J.K. Liu & K.D. Hyde; the sequence data have been deposited in GenBank (A. lignicola strain DK/V4, accession number KC866317.1). Medical management of keratitis due to such rarely reported fungal species may be difficult, necessitating surgical procedures.