Development of a valid and reliable pterygium surgery assessment scale for ophthalmology residents.

BACKGROUND: Microsurgery training has become an important part of ophthalmology teaching and one of the main topics of examination. Accurate and effective evaluation of microsurgery skills is vital for the training and teaching of residents. In this study, we aimed to establish a pterygium surgery assessment scale for use by ophthalmic residents and evaluate its reliability and validity. METHODS: Based on a literature search, experienced pterygium surgeons developed the preliminary scale according to the standard surgical procedure. The preliminary scale and a questionnaire were sent to teaching and research experts in the field for feedback. Face and content validity and reliability of the scale were determined by rounds of modifications based on expert feedback. For construct validity, existing assessment scales were obtained and a range of factors were tested. RESULTS: Nineteen expert surgeons completed the questionnaire and modifications were made until all surgeons agreed on the final scale. Good construct validity was found by evaluation against 257 existing scales. For reliability, 280 evaluation scales were completed. Inter- and intra-rater reliability analysis both found Intraclass Correlation Coefficient (ICC) > 0.8 for all items and total scores. CONCLUSION: The pterygium surgery assessment scale developed in this study has good reliability and validity, and is an effective measurement tool for the evaluation of ophthalmology residents' pterygium surgical skills.

A deep transfer learning framework for the automated assessment of corneal inflammation on in vivo confocal microscopy images.

PURPOSE: Infiltration of activated dendritic cells and inflammatory cells in cornea represents an important marker for defining corneal inflammation. Deep transfer learning has presented a promising potential and is gaining more importance in computer assisted diagnosis. This study aimed to develop deep transfer learning models for automatic detection of activated dendritic cells and inflammatory cells using in vivo confocal microscopy images. METHODS: A total of 3453 images was used to train the models. External validation was performed on an independent test set of 558 images. A ground-truth label was assigned to each image by a panel of cornea specialists. We constructed a deep transfer learning network that consisted of a pre-trained network and an adaptation layer. In this work, five pre-trained networks were considered, namely VGG-16, ResNet-101, Inception V3, Xception, and Inception-ResNet V2. The performance of each transfer network was evaluated by calculating the area under the curve (AUC) of receiver operating characteristic, accuracy, sensitivity, specificity, and G mean. RESULTS: The best performance was achieved by Inception-ResNet V2 transfer model. In the validation set, the best transfer system achieved an AUC of 0.9646 (P<0.001) in identifying activated dendritic cells (accuracy, 0.9319; sensitivity, 0.8171; specificity, 0.9517; and G mean, 0.8872), and 0.9901 (P<0.001) in identifying inflammatory cells (accuracy, 0.9767; sensitivity, 0.9174; specificity, 0.9931; and G mean, 0.9545). CONCLUSIONS: The deep transfer learning models provide a completely automated analysis of corneal inflammatory cellular components with high accuracy. The implementation of such models would greatly benefit the management of corneal diseases and reduce workloads for ophthalmologists.

Spectral domain anterior segment optical coherence tomography assessment of pterygium and pinguecula.

PURPOSE: To study the morphological patterns of pterygia and pingueculae using high-resolution anterior segment spectral domain optical coherence tomography (SD-OCT). METHODS: Prospective cross-sectional study of 25 eyes presented with pterygia and pingueculae was conducted, and the eyes were examined by anterior segment SD-OCT. RESULTS: We examined 25 eyes, including 13 eyes with primary pterygia, six eyes with recurrent pterygia, one case with a pseudopterygium and five eyes with pingueculae. Primary pterygia revealed elevation of the corneal epithelium by a wedge-shaped mass of tissue separating the corneal epithelium from the underlying Bowman's membrane, which became wavy and interrupted. We found satellite masses of pterygium tissue advanced under the epithelium beyond the clinically seen pterygium margins. In recurrent pterygia, we detected that the central tip of the pterygium was more advanced and creeping beneath the basal corneal epithelium than the primary pterygium. In pseudopterygium, the SD-OCT images showed that the overgrowing membrane was not really attached to the underlying cornea. In cases of pingueculae, SD-OCT revealed a wedge-shaped mass that was nearly similar in pattern to that of the pterygia but stopped at the limbal region. Immediately after removal of pterygia, we noticed many remnants of the pterygia masses over the corneal stroma in spite of the clinically clear appearance of cornea. CONCLUSIONS: SD-OCT provided us with high-resolution images of the pterygium and the pinguecula and showed clearly the anatomical relationship between the corneal tissues and these lesions. The use of this new modality of imaging may help to decrease the current recurrence rates after pterygium excision through using the anterior segment SD-OCT in the evaluation of these lesions.