Bacterial keratitis (BK) is the most common type of infectious keratitis. The spectrum of pathogenic bacteria and their susceptibility to antibiotics varied with the different regions. A meta-analysis was conducted to review the global culture rate, distribution, current trends, and drug susceptibility of isolates from BK over the past 20 years (2000–2020). Four databases were searched, and published date was limited between 2000 and 2020. Main key words were “bacterial keratitis”, “culture results” and “drug resistance”. Forty-two studies from twenty-one countries (35 cities) were included for meta-analysis. The overall positive culture rate was 47% (95%CI, 42–52%). Gram-positive cocci were the major type of bacteria (62%), followed by Gram-negative bacilli (30%), Gram-positive bacilli (5%), and Gram-negative cocci (5%). Staphylococcus spp. (41.4%), Pseudomonas spp. (17.0%), Streptococcus spp. (13.1%), Corynebacterium spp. (6.6%) and Moraxella spp. (4.1%) were the most common bacterial organism. The antibiotic resistance pattern analysis revealed that most Gram-positive cocci were susceptive to aminoglycoside (86%), followed by fluoroquinolone (81%) and cephalosporin (79%). Gram-negative bacilli were most sensitive to cephalosporin (96%) and fluoroquinolones (96%), followed by aminoglycoside (92%). In Gram-positive cocci, the susceptibility trends of fluoroquinolones were decreasing since 2010. Clinics should pay attention to the changing trends of pathogen distribution and their drug resistance pattern and should diagnose and choose sensitive antibiotics based on local data.

Background: Infectious keratitis (IK) is an ocular emergency caused by a variety of microorganisms, including bacteria, fungi, viruses, and parasites. Culture-based methods were the gold standard for diagnosing IK, but difficult biopsy, delaying report, and low positive rate limited their clinical application. Objectives: This study aims to construct a deep-learning-based auxiliary diagnostic model for early IK diagnosis. Design: A retrospective study. Methods: IK patients with pathological diagnosis were enrolled and their slit-lamp photos were collected. Image augmentation, normalization, and histogram equalization were applied, and five image classification networks were implemented and compared. Model blending technique was used to combine the advantages of single model. The performance of combined model was validated by 10-fold cross-validation, receiver operating characteristic curves (ROC), confusion matrix, Gradient-wright class activation mapping (Grad-CAM) visualization, and t-distributed Stochastic Neighbor Embedding (t-SNE). Three experienced cornea specialists were invited and competed with the combined model on making clinical decisions. Results: Overall, 4830 slit-lamp images were collected from patients diagnosed with IK between June 2010 and May 2021, including 1490 (30.8%) bacterial keratitis (BK), 1670 (34.6%) fungal keratitis (FK), 600 (12.4%) herpes simplex keratitis (HSK), and 1070 (22.2%) Acanthamoeba keratitis (AK). KeratitisNet, the combination of ResNext101_32x16d and DenseNet169, reached the highest accuracy 77.08%. The accuracy of KeratitisNet for diagnosing BK, FK, AK, and HSK was 70.27%, 77.71%, 83.81%, and 79.31%, and AUC was 0.86, 0.91, 0.96, and 0.98, respectively. KeratitisNet was mainly confused in distinguishing BK and FK. There were 20% of BK cases mispredicted into FK and 16% of FK cases mispredicted into BK. In diagnosing each type of IK, the accuracy of model was significantly higher than that of human ophthalmologists ( p < 0.001). Conclusion: KeratitisNet demonstrates a good performance on clinical IK diagnosis and classification. Deep learning could provide an auxiliary diagnostic method to help clinicians suspect IK using different corneal manifestations.