Multi-Scale Convolutional Neural Network for Accurate Corneal Segmentation in Early Detection of Fungal Keratitis

Mayya, Veena; S, Sowmya Kamath; Kulkarni, Uma; Hazarika, Manali; Barua, Prabal Datta; Acharya, U. Rajendra

doi:10.3390/jof7100850

Cited by 16 publications

(11 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is essential to note that a DL model learning from the impressions of experts may reduce interference from uncharacteristic manifestations of several BK patients. Nevertheless, many prior DL models accepted experts’ impressions as a gold standard [ 21 , 22 , 23 ]. Moreover, comparing the performance between ensemble models and their constituent DL models is objective and fair when using the same image database.…”

Section: Discussionmentioning

confidence: 99%

Deep Learning Approach in Image Diagnosis of Pseudomonas Keratitis

et al. 2022

View full text Add to dashboard Cite

This investigation aimed to explore deep learning (DL) models’ potential for diagnosing Pseudomonas keratitis using external eye images. In the retrospective research, the images of bacterial keratitis (BK, n = 929), classified as Pseudomonas (n = 618) and non-Pseudomonas (n = 311) keratitis, were collected. Eight DL algorithms, including ResNet50, DenseNet121, ResNeXt50, SE-ResNet50, and EfficientNets B0 to B3, were adopted as backbone models to train and obtain the best ensemble 2-, 3-, 4-, and 5-DL models. Five-fold cross-validation was used to determine the ability of single and ensemble models to diagnose Pseudomonas keratitis. The EfficientNet B2 model had the highest accuracy (71.2%) of the eight single-DL models, while the best ensemble 4-DL model showed the highest accuracy (72.1%) among the ensemble models. However, no statistical difference was shown in the area under the receiver operating characteristic curve and diagnostic accuracy among these single-DL models and among the four best ensemble models. As a proof of concept, the DL approach, via external eye photos, could assist in identifying Pseudomonas keratitis from BK patients. All the best ensemble models can enhance the performance of constituent DL models in diagnosing Pseudomonas keratitis, but the enhancement effect appears to be limited.

show abstract

Section: Discussionmentioning

confidence: 99%

Deep Learning Approach in Image Diagnosis of Pseudomonas Keratitis

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Various reports have described the use of artificial intelligence deep learning to identify the presence of infectious keratitis, or determine type of infection based on imaging [45][46][47][48][49][50][51].…”

Section: Diagnosticsmentioning

confidence: 99%

Therapy for contact lens-related ulcers

Jeang

Tuli

2022

Current Opinion in Ophthalmology

View full text Add to dashboard Cite

Purpose of reviewThe current review covers the current literature and practice patterns of antimicrobial therapy for contact lens-related microbial keratitis (CLMK). Although the majority of corneal ulcers are bacterial, fungus and acanthamoeba are substantial contributors in CLMK and are harder to treat due to the lack of commercially available topical medications and low efficacy of available topical therapy.Recent findingsTopical antimicrobials remain the mainstay of therapy for corneal ulcers. Fluoroquinolones may be used as monotherapy for small, peripheral bacterial ulcers. Antibiotic resistance is a persistent problem. Fungal ulcers are less responsive to topical medications and adjunct oral or intrastromal antifungal medications may be helpful. Acanthamoeba keratitis continues to remain a therapeutic challenge but newer antifungal and antiparasitic agents may be helpful adjuncts. Other novel and innovative therapies are being studied currently and show promise.SummaryContact lens-associated microbial keratitis is a significant health issue that can cause vision loss. Treatment remains a challenge but many promising diagnostics and procedures are in the pipeline and offer hope.

show abstract

“…CBMIR systems aim to solve this issue by supporting query by example (QBE) interfaces, which can return the closest matched images from a huge database, given a query image. CBMIR also helps doctors by providing functionalities like identification of similar patient characteristics through automated image analysis and feature extraction [8,9]. CBMIR techniques have been adapted by researchers to address various clinical tasks across imaging modalities.…”

Section: Related Workmentioning

confidence: 99%