ABSTRACT
Purpose: Early diagnosis and treatment of retinoblastoma are of paramount importance for a positive clinical outcome. The most common sign of retinoblastoma is leukocoria, or white pupil. Effective, easy-to-perform, community-based screening is needed to improve outcomes in lower-income regions. The EyeScreen (developed by Joshua Meyer from the University of Michigan) Android (Google LLC) smartphone application is an important step toward addressing this need. The purpose of this study was to examine the potential of the novel use of low-cost technologies-a cell phone application and machine learning-to identify leukocoria. Design: A cell phone application was developed and refined with the feedback from on-site, single-population use in Ethiopia. Application performance was evaluated in this technology validation study. Participants: One thousand four hundred fifty-seven participants were recruited from ophthalmology and pediatric clinics in Addis Ababa, Ethiopia. Methods: Photographs obtained with inexpensive Android smartphones running the EyeScreen Application were used to train an ImageNet (ResNet) machine learning model and to measure the performance of the app. Eighty percent of the images were used in training the model, and 20% were reserved for testing. Main Outcome Measures: Performance of the model was measured in terms of sensitivity, specificity, receiver operating characteristic (ROC) curve, and precision-recall curve. Results: Analyses of the participant images resulted in the following at the participant level: sensitivity, 87%; specificity, 73%; area under the ROC curve, 0.93; and area under the precision-recall curve, 0.77. Conclusions: EyeScreen has the potential to serve as an effective screening tool in the areas of the world most affected by delayed retinoblastoma diagnosis. The relatively high initial performance of the machine learning model with small training datasets in this early-phase study can serve as a proof of concept for future use of machine learning and artificial intelligence in ophthalmic applications.
ABSTRACT
INTRODUCTION: Providing adequate nutrition to preterm infants who are born to HIV-positive mothers is more challenging due to the mother's underlying health and nutrition status. The understanding of these issues and active participation of the mothers have a significant role in giving continuous care for HIV-exposed preterm infant. Hence, this study aimed to explore the experience of HIV-positive mothers' feeding practice of their preterm infants, and health workers to identify barriers and facilitators of feeding HIV-exposed preterm infants. METHODS: A phenomenological qualitative study design was conducted in Addis Ababa, Ethiopia, between May 1, 2016 and March 31, 2017. Mothers who gave birth to HIV-exposed preterm infants at the study sites' follow-up clinic were traced and invited by the healthcare providers to voluntarily participate in this study. Fifteen in-depth interviews with mothers of HIV-exposed preterm infants and seven key informant interviews with health professionals and policymakers were carried out. The interviews were transcribed and translated and then manually analyzed thematically. RESULTS: The health education given during antenatal care (ANC) did not consider the feeding practice needs for HIV-exposed preterm infants. Child health status, desire to have a healthy infant, financial constraints and family support were among the influential factors in the feeding practice of HIV-exposed preterm infants mentioned by the study participants. HIV-exposed preterm infant feeding procedure neither has a guideline nor is clearly mentioned in the national HIV guidelines. CONCLUSION: The desire to have a healthy infant was a major facilitator for feeding of HIV-exposed premature infants. However, financial constraints majorly limited the option to be only exclusive breastfeeding. This became even more problematic for the mother if the premature infant became ill and could not breastfeed well.
