Search

Purpose To compare the performance of a novel convolutional neural network (CNN) classifier and human graders in detecting angle closure in EyeCam (Clarity Medical Systems, Pleasanton, California, USA) goniophotographs. Design Retrospective cross-sectional study. Methods Subjects from the Chinese American Eye Study underwent EyeCam goniophotography in 4 angle quadrants. A CNN classifier based on the ResNet-50 architecture was trained to detect angle closure, defined as inability to visualize the pigmented trabecular meshwork, using reference labels by a single experienced glaucoma specialist. The performance of the CNN classifier was assessed using an independent test dataset and reference labels by the single glaucoma specialist or a panel of 3 glaucoma specialists. This performance was compared to that of 9 human graders with a range of clinical experience. Outcome measures included area under the receiver operating characteristic curve (AUC) metrics and Cohen kappa coefficients in the binary classification of open or closed angle. Results The CNN classifier was developed using 29,706 open and 2,929 closed angle images. The independent test dataset was composed of 600 open and 400 closed angle images. The CNN classifier achieved excellent performance based on single-grader (AUC = 0.969) and consensus (AUC = 0.952) labels. The agreement between the CNN classifier and consensus labels (κ = 0.746) surpassed that of all non-reference human graders (κ = 0.578-0.702). Human grader agreement with consensus labels improved with clinical experience (P = 0.03). Conclusion A CNN classifier can effectively detect angle closure in goniophotographs with performance comparable to that of an experienced glaucoma specialist. This provides an automated method to support remote detection of patients at risk for primary angle closure glaucoma.

Simple Summary Breastfeeding has been shown to lower the risk of oncogenesis in many pediatric cancers, with longer periods of breastfeeding having the most protective effect. However, an association has not yet been determined for the consequence or benefit of breastfeeding in retinoblastoma (RB), the most common intraocular malignancy of childhood affecting 8000 children worldwide each year. Herein, we aimed to understand the role of breastfeeding in the severity of development of nonhereditary RB, specifically its relationship to age at diagnosis, ocular prognosis, and extraocular involvement. Our analysis of 344 patients indicated that neither breastfeeding nor formula feeding was associated with differences in age at diagnosis, ocular prognosis, or extraocular involvement. More research elucidating the factors affecting the development of RB is warranted both to understand the pathophysiology of tumor development and to develop clinical recommendations for preventive care. Abstract The protective effects of breastfeeding on various childhood malignancies have been established but an association has not yet been determined for retinoblastoma (RB). We aimed to further investigate the role of breastfeeding in the severity of nonhereditary RB development, assessing relationship to (1) age at diagnosis, (2) ocular prognosis, measured by International Intraocular RB Classification (IIRC) or Intraocular Classification of RB (ICRB) group and success of eye salvage, and (3) extraocular involvement. Analyses were performed on a global dataset subgroup of 344 RB patients whose legal guardian(s) consented to answer a neonatal questionnaire. Patients with undetermined or mixed feeding history, family history of RB, or sporadic bilateral RB were excluded. There was no statistically significant difference between breastfed and formula-fed groups in (1) age at diagnosis (p = 0.20), (2) ocular prognosis measures of IIRC/ICRB group (p = 0.62) and success of eye salvage (p = 0.16), or (3) extraocular involvement shown by International Retinoblastoma Staging System (IRSS) at presentation (p = 0.74), lymph node involvement (p = 0.20), and distant metastases (p = 0.37). This study suggests that breastfeeding neither impacts the sporadic development nor is associated with a decrease in the severity of nonhereditary RB as measured by age at diagnosis, stage of disease, ocular prognosis, and extraocular spread. A further exploration into the impact of diet on children who develop RB is warranted.