RESUMEN
PURPOSE: Epidemiological changes in retinopathy of prematurity (ROP) depend on neonatal care, neonatal mortality, and the ability to carefully titrate and monitor oxygen. We evaluate whether an artificial intelligence (AI) algorithm for assessing ROP severity in babies can be used to evaluate changes in disease epidemiology in babies from South India over a 5-year period. DESIGN: Retrospective cohort study. PARTICIPANTS: Babies (3093) screened for ROP at neonatal care units (NCUs) across the Aravind Eye Care System (AECS) in South India. METHODS: Images and clinical data were collected as part of routine tele-ROP screening at the AECS in India over 2 time periods: August 2015 to October 2017 and March 2019 to December 2020. All babies in the original cohort were matched 1:3 by birthweight (BW) and gestational age (GA) with babies in the later cohort. We compared the proportion of eyes with moderate (type 2) or treatment-requiring (TR) ROP, and an AI-derived ROP vascular severity score (from retinal fundus images) at the initial tele-retinal screening exam for all babies in a district, VSS), in the 2 time periods. MAIN OUTCOME MEASURES: Differences in the proportions of type 2 or worse and TR-ROP cases, and VSS between time periods. RESULTS: Among BW and GA matched babies, the proportion [95% confidence interval {CI}] of babies with type 2 or worse and TR-ROP decreased from 60.9% [53.8%-67.7%] to 17.1% [14.0%-20.5%] (P < 0.001) and 16.8% [11.9%-22.7%] to 5.1% [3.4%-7.3%] (P < 0.001), over the 2 time periods. Similarly, the median [interquartile range] VSS in the population decreased from 2.9 [1.2] to 2.4 [1.8] (P < 0.001). CONCLUSIONS: In South India, over a 5-year period, the proportion of babies developing moderate to severe ROP has dropped significantly for babies at similar demographic risk, strongly suggesting improvements in primary prevention of ROP. These results suggest that AI-based assessment of ROP severity may be a useful epidemiologic tool to evaluate temporal changes in ROP epidemiology. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found after the references.
Asunto(s)
Retinopatía de la Prematuridad , Telemedicina , Recién Nacido , Lactante , Humanos , Retinopatía de la Prematuridad/diagnóstico , Retinopatía de la Prematuridad/epidemiología , Estudios Retrospectivos , Inteligencia Artificial , Factores de Riesgo , Edad Gestacional , Peso al Nacer , Telemedicina/métodos , Tamizaje Neonatal/métodosRESUMEN
PURPOSE: The International Classification of Retinopathy of Prematurity is a consensus statement that creates a standard nomenclature for classification of retinopathy of prematurity (ROP). It was initially published in 1984, expanded in 1987, and revisited in 2005. This article presents a third revision, the International Classification of Retinopathy of Prematurity, Third Edition (ICROP3), which is now required because of challenges such as: (1) concerns about subjectivity in critical elements of disease classification; (2) innovations in ophthalmic imaging; (3) novel pharmacologic therapies (e.g., anti-vascular endothelial growth factor agents) with unique regression and reactivation features after treatment compared with ablative therapies; and (4) recognition that patterns of ROP in some regions of the world do not fit neatly into the current classification system. DESIGN: Review of evidence-based literature, along with expert consensus opinion. PARTICIPANTS: International ROP expert committee assembled in March 2019 representing 17 countries and comprising 14 pediatric ophthalmologists and 20 retinal specialists, as well as 12 women and 22 men. METHODS: The committee was initially divided into 3 subcommittees-acute phase, regression or reactivation, and imaging-each of which used iterative videoconferences and an online message board to identify key challenges and approaches. Subsequently, the entire committee used iterative videoconferences, 2 in-person multiday meetings, and an online message board to develop consensus on classification. MAIN OUTCOME MEASURES: Consensus statement. RESULTS: The ICROP3 retains current definitions such as zone (location of disease), stage (appearance of disease at the avascular-vascular junction), and circumferential extent of disease. Major updates in the ICROP3 include refined classification metrics (e.g., posterior zone II, notch, subcategorization of stage 5, and recognition that a continuous spectrum of vascular abnormality exists from normal to plus disease). Updates also include the definition of aggressive ROP to replace aggressive-posterior ROP because of increasing recognition that aggressive disease may occur in larger preterm infants and beyond the posterior retina, particularly in regions of the world with limited resources. ROP regression and reactivation are described in detail, with additional description of long-term sequelae. CONCLUSIONS: These principles may improve the quality and standardization of ROP care worldwide and may provide a foundation to improve research and clinical care.
Asunto(s)
Retina/diagnóstico por imagen , Retinopatía de la Prematuridad/clasificación , Diagnóstico por Imagen , Progresión de la Enfermedad , Edad Gestacional , Humanos , Recién Nacido , Retinopatía de la Prematuridad/diagnósticoRESUMEN
A spiropyran-containing triazole-phosphatidylcholine (SPTPC) was synthesized through a copper-catalyzed azide alkyne cyclo-addition (CuAAC) reaction. In water, SPTPCs self-assembled and a spontaneous spiropyran-to-merocyanine (SP-to-MC) isomerization occurred, resulting in coexistence of liposomes and fibers, and switching from the spiropyran (SP) to the merocyanine (MC) isomeric structure induced a reversible transition between these molecular assemblies. Study of the self-assembly of SPTPCs and photo-induced liposome-fiber assembly-transition revealed that the presence of MC enabled additional inter-membrane interaction during self-assembly and that the MC-stacking effect was the driving force for the assembly-transition. Exposure to UV light induced switching from SP to MC, where the planar structure of MC and the confinement of MC led to enhanced MC-stacking. The effect of MC-stacking was both advantageous and disadvantageous: MC-stacking perturbed the hydrophobic phase in the bilayer membrane and facilitated the liposome-to-fiber transition, otherwise the MC-stacking retarded switching of MC to SP, and caused an incomplete recovery of MC to SP during fiber-to-liposome recovery, thus a fatigue of SP was induced by MC-stacking during the liposome-to-fiber transition cycle. To decrease the intermolecular interactions and suppress MC-stacking, photo-inert triazole-phosphatidylcholine (TPC) was incorporated to prepare two-component TPC/SPTPC-liposomes, which exhibited better recovery kinetics. The photo-adaptive behavior of TPC/SPTPC-liposomes confirmed the disturbance of bilayer membranes by inter-membrane MC-stacking and the formation of MCTPC-enriched phases in the bilayer membrane.
Asunto(s)
Benzopiranos/química , Indoles/química , Liposomas/química , Nitrocompuestos/química , Fosfatidilcolinas/química , Procesos Fotoquímicos , Triazoles/química , Alquinos/química , Azidas/química , Catálisis , Cobre/química , Membrana Dobles de Lípidos/química , Transición de Fase , Rayos UltravioletaRESUMEN
A series of functional nanogels were synthesized by a step-growth mechanism that involved diisocyanate addition to a modest stoichiometric excess of multi-thiols. Nanogels with sizes less than 10 nm were obtained as room temperature liquids with residual thiol groups used to attach methacrylate functionality. Depending on nanogel structure, bulk nanogel properties varied widely, as did the properties of the nanogel-derived and nanogel-modified polymers. Photopolymerization of the reactive nanogels in combination with a dimethacrylate monomer showed dramatically enhanced reaction rate and conversion compared with the dimethacrylate homopolymer. Polymerization shrinkage/ stress as well as mechanical properties of the polymer networks were controlled by changing the ratio of nanogels and dimethacrylate monomers used in formulations. Thus, this study shows the potential of step-growth nanogels for beneficial changes in resin reactivity and application-based performance.
RESUMEN
We present a strategy for directly and efficiently polymerizing aqueous dispersions of reactive nanogels into covalently crosslinked polymer networks with properties that are determined by the initial chemical and physical nanogel structure. This technique can extend the range of achievable properties and architectures for networks formed in solution, particularly in water where monomer selection for direct polymerization and the final network properties are quite limited. Nanogels were initially obtained from a solution polymerization of a hydrophilic monomethacrylate and either a hydrophilic PEG-based dimethacrylate or a more hydrophobic urethane dimethacrylate, which produced globular particles with diameters of 10-15 nm with remarkably low polydispersity in some cases. Networks derived from a single type of nanogel or a blend of nanogels with different chemistries when dispersed in water gelled within minutes when exposed to low intensity UV light. Modifying the nanogel structure changes both covalent and non-covalent secondary interactions in the crosslinked networks and reveals critical design criteria for the development of networks from highly internally branched, nanoscale prepolymer precursors.
Asunto(s)
Nanopartículas/química , Polietilenglicoles/química , Polietileneimina/química , Interacciones Hidrofóbicas e Hidrofílicas , Nanogeles , Nanopartículas/ultraestructura , Polietilenglicoles/síntesis química , Polietileneimina/síntesis química , Polimerizacion/efectos de la radiación , Polímeros/síntesis química , Polímeros/química , Rayos Ultravioleta , Agua/químicaRESUMEN
PURPOSE: To evaluate the demographic profile of premature infants presenting with stage 4B and stage 5 retinopathy of prematurity (ROP) at a tertiary referral center in South India. METHODS: This was a retrospective review including all premature infants with stage 4B and 5 ROP between January 1, 2015 and December 31, 2022. Parameters included the newborns born at the tertiary care nursery of various institutes, gestational age, birth weight, age at presentation to the hospital, risk factors, screening details, neonatal intensive care unit details, reason for consultation, and timing of referral to the center. RESULTS: Two hundred twenty eyes of 110 premature infants were included. Of 110 infants, 6 were born within the same city and 104 were from other districts or states. Mean birth weight was 1,125 ± 360 g and mean gestational age at birth was 28 ± 2 weeks. Mean age was 42 ± 82 weeks and median age at presentation was 17 weeks (range: 2.86 to 591 weeks). Male-to-female infant ratio was 1.34:1. Fifty (45.4%) infants had bilateral stage 5 ROP, 17 (15.4%) had stage 5 in one eye and stage 4B in the other eye, 15 (13.6%) had bilateral stage 4B and 23 (20.9%) had stage 4B in one eye and stage 4A/stage 3 in the other eye. Five (4.5%) had stage 5 in one eye and vitreous hemorrhage/stage 4A in the other eye. Among those with bilateral stage 5 ROP, 90% were from neighboring districts/states. Fifty-two (47.27%) infants underwent vitreoretinal surgery. Of 110 infants, 28 (25.45%) were self-referred (late presentation due to family-related issues), 80 (72.73%) were referred by ophthalmologists either after a few sessions of late screening or for further management, 1 (0.91%) was referred through telescreening, and 1 was referred from pediatricians directly to the hospital. CONCLUSIONS: This study highlights the importance of awareness of the disease and screening of premature infants. Lack of these two factors leads to late presentation of these infants with advanced stages and serious implications. [J Pediatr Ophthalmol Strabismus. 20XX;X(X):XX-XX.].
RESUMEN
Artificial Intelligence (AI) is a revolutionary technology that has the potential to develop into a widely implemented system that could reduce the dependence on qualified professionals/experts for screening the large at-risk population, especially in the Indian scenario. Deep learning involves learning without being explicitly told what to focus on and utilizes several layers of artificial neural networks (ANNs) to create a robust algorithm that is capable of high-complexity tasks. Convolutional neural networks (CNNs) are a subset of ANNs that are particularly useful for image processing as well as cognitive tasks. Training of these algorithms involves inputting raw human-labeled data, which are then processed through the algorithm's multiple layers and allow CNN to develop their own learning of image features. AI systems must be validated using different population datasets since the performance of the AI system would vary according to the population. Indian datasets have been used in AI-based risk model that could predict whether an infant would develop treatment-requiring retinopathy of prematurity (ROP). AI also served as an epidemiological tool by objectively showing that a higher ROP severity was in Neonatal intensive care units (NICUs) that did not have the resources to monitor and titrate oxygen. There are rising concerns about the medicolegal aspect of AI implementation as well as discussion on the possibilities of catastrophic life-threatening diseases like retinoblastoma and lipemia retinalis being missed by AI. Computer-based systems have the advantage over humans in not being susceptible to biases or fatigue. This is especially relevant in a country like India with an increased rate of ROP and a preexisting strained doctor-to-preterm child ratio. Many AI algorithms can perform in a way comparable to or exceeding human experts, and this opens possibilities for future large-scale prospective studies.
Asunto(s)
Inteligencia Artificial , Retinopatía de la Prematuridad , Humanos , Retinopatía de la Prematuridad/diagnóstico , Recién Nacido , Redes Neurales de la Computación , Manejo de la Enfermedad , Algoritmos , India/epidemiologíaRESUMEN
OBJECTIVES: To analyse the profile and implication of genetic testing in a cohort of retinoblastoma (RB) patients and their families conducted on a single day during World Retinoblastoma Awareness Week 2017. METHODS: Retrospective analysis of blood samples were collected from 411 subjects, including 113 probands at a camp organised for RB awareness and were analysed for RB1 mutations by Sanger sequencing and Multiplex Ligation-dependent Probe Amplification (MLPA). If germline mutations were detected, the parents and siblings of the proband were tested for the same mutation. RESULTS: Germline RB1 mutations were identified in 61/113(54%) probands with a mutation detection rate of 96% (47/49) and 22% (14/64) for bilateral and unilateral RB, respectively. Ten novel pathogenic mutations were identified. Splice mutation was most common (31%) followed by nonsense mutation (26%). The mean age at RB diagnosis was significantly lower in patients having germline RB1 mutation (mean 10.7 months ±2.5) compared to those without (mean 27.2 months ±6.5) (p = <0.0001). Parental transmission of the mutant allele was detected in 15/61(25%) cases of which 11(18%) parents were unaffected indicating incomplete penetrance. The origin of the variant allele was both paternal (n = 7) and maternal (n = 4) wherein 5 were bilateral and 6 unilateral. CONCLUSIONS: The detection of a germline mutation impacts the proband and family members due to its implications on change in prognosis, frequency of subsequent evaluations, screening for ocular and non-ocular cancers, and surveillance of family and future progeny.
Asunto(s)
Pruebas Genéticas , Mutación de Línea Germinal , Neoplasias de la Retina , Proteínas de Unión a Retinoblastoma , Retinoblastoma , Humanos , Retinoblastoma/genética , Retinoblastoma/sangre , Masculino , Femenino , Estudios Retrospectivos , Neoplasias de la Retina/genética , Neoplasias de la Retina/diagnóstico , Neoplasias de la Retina/sangre , India/epidemiología , Lactante , Pruebas Genéticas/métodos , Proteínas de Unión a Retinoblastoma/genética , Preescolar , Ubiquitina-Proteína Ligasas/genética , Linaje , Análisis Mutacional de ADNRESUMEN
Importance: Retinopathy of prematurity (ROP) is a leading cause of blindness in children, with significant disparities in outcomes between high-income and low-income countries, due in part to insufficient access to ROP screening. Objective: To evaluate how well autonomous artificial intelligence (AI)-based ROP screening can detect more-than-mild ROP (mtmROP) and type 1 ROP. Design, Setting, and Participants: This diagnostic study evaluated the performance of an AI algorithm, trained and calibrated using 2530 examinations from 843 infants in the Imaging and Informatics in Retinopathy of Prematurity (i-ROP) study, on 2 external datasets (6245 examinations from 1545 infants in the Stanford University Network for Diagnosis of ROP [SUNDROP] and 5635 examinations from 2699 infants in the Aravind Eye Care Systems [AECS] telemedicine programs). Data were taken from 11 and 48 neonatal care units in the US and India, respectively. Data were collected from January 2012 to July 2021, and data were analyzed from July to December 2023. Exposures: An imaging processing pipeline was created using deep learning to autonomously identify mtmROP and type 1 ROP in eye examinations performed via telemedicine. Main Outcomes and Measures: The area under the receiver operating characteristics curve (AUROC) as well as sensitivity and specificity for detection of mtmROP and type 1 ROP at the eye examination and patient levels. Results: The prevalence of mtmROP and type 1 ROP were 5.9% (91 of 1545) and 1.2% (18 of 1545), respectively, in the SUNDROP dataset and 6.2% (168 of 2699) and 2.5% (68 of 2699) in the AECS dataset. Examination-level AUROCs for mtmROP and type 1 ROP were 0.896 and 0.985, respectively, in the SUNDROP dataset and 0.920 and 0.982 in the AECS dataset. At the cross-sectional examination level, mtmROP detection had high sensitivity (SUNDROP: mtmROP, 83.5%; 95% CI, 76.6-87.7; type 1 ROP, 82.2%; 95% CI, 81.2-83.1; AECS: mtmROP, 80.8%; 95% CI, 76.2-84.9; type 1 ROP, 87.8%; 95% CI, 86.8-88.7). At the patient level, all infants who developed type 1 ROP screened positive (SUNDROP: 100%; 95% CI, 81.4-100; AECS: 100%; 95% CI, 94.7-100) prior to diagnosis. Conclusions and Relevance: Where and when ROP telemedicine programs can be implemented, autonomous ROP screening may be an effective force multiplier for secondary prevention of ROP.
Asunto(s)
Retinopatía de la Prematuridad , Recién Nacido , Lactante , Niño , Humanos , Retinopatía de la Prematuridad/diagnóstico , Inteligencia Artificial , Estudios Transversales , Edad Gestacional , Recien Nacido PrematuroRESUMEN
BACKGROUND: Intravitreal anti-vascular endothelial growth factor (IVA) injection is known to cause contraction of fibrovascular proliferation (FVP), when present in severe retinopathy of prematurity (ROP). AIM: To assess the structural outcomes of IVA injection in the treatment of severe posterior ROP with significant FVP. METHODS: It was a retrospective study in which 36 eyes of 18 preterm babies who developed > 4 clock hours of FVP in zone I or posterior zone II, were treated with either intravitreal 0.625 mg bevacizumab or intravitreal 0.2 mg of ranibizumab. Favorable structural outcome included resolution of plus disease and FVP without the development of tractional retinal detachment. Secondary outcome measure included either full retinal maturation at follow-up or development of recurrent disease requiring additional treatment. Adverse outcomes included progression to retinal detachment. RESULTS: The mean gestational age of the 18 preterm babies was 30 wk (range 27-36), and mean birth weight was 1319 g (range 650-1980 g). Mean post-menstrual age (PMA) at the time of primary treatment was 35.5 wk (range 31-41 wk). All eyes showed regression of plus disease and FVP. 5 eyes of 3 babies showed reactivation of disease and were treated with repeat IVA (n = 2 eyes) or peripheral laser photocoagulation (n = 3 eyes) respectively. 16 out of 36 (44%) reached retinal vascular maturation at final follow up at 5 years. CONCLUSION: There was good resolution of severe posterior ROP with FVP with IVA, with retinal maturity of 44% at 5 year follow-up and a reactivation rate of 13.8%. When the IVA injection is given prior to 37 wk PMA, while disease is in phase 2, it is less likely to cause contracture of pre-existing FVP.
RESUMEN
INTRODUCTION: Pediatric rhegmatogenous retinal detachments, especially those presenting at birth or soon afterward, have a high likelihood of syndromic associations that can be confirmed by genetic testing. MATERIALS AND METHODS: A 5-month-old child was found to have high myopia in the right eye (RE) with highly tessellated fundus, opalescent vitreous, and peripheral thinning. Left eye had a shallow retinal detachment for which he underwent belt buckling. The baby had an occipital skin tag. A provisional diagnosis of Stickler syndrome was made. RESULTS: On 1-month follow-up, left eye retina was attached and 360° laser barrage was done. Fluorescein angiography was done which revealed peripheral avascular retina in both eyes. MRI and genetic testing were suggestive of syndromic association. Genetic testing revealed pathogenic mutation in COL 18A1 suggestive of Knobloch syndrome in the baby, and both parents were found to be carriers of the same mutation. However, brain MRI showed features not pathognomonic of Knobloch syndrome. CONCLUSION: Although Knobloch syndrome is associated with vitreoretinal degeneration and high risk of retinal detachment, there seems to be no recommendation for prophylaxis in the other eye and therefore we preferred to observe the RE closely. A unique feature noted in our case was the peripheral avascular zone (PAZ). The PAZ could be contributed by multiple factors such as high myopia, or due to endostatin deficiency (which is a derivative of collagen XVIII) or an underlying WNT signalling abnormality.
Asunto(s)
Miopía , Desprendimiento de Retina , Recién Nacido , Lactante , Masculino , Humanos , Niño , Desprendimiento de Retina/diagnóstico , Desprendimiento de Retina/genética , Retina , Fondo de Ojo , Miopía/genéticaRESUMEN
PURPOSE: The aim was to study the characteristics of recurrence patterns in the form of scar recurrence, new lesions, and vitreous seeds which is necessary in anticipating future events for the management of retinoblastoma (RB). METHODS: This retrospective analytical observational study was conducted in a tertiary care hospital in South India; we included 64 eyes of 45 patients having RB from January 2019 to July 2020. The inclusion criterion was treatment-naïve patients with > 12 months of follow-up period. Recurrence patterns were defined as Pattern 1a and Pattern 1b: local and diffuse dissemination of vitreous seeds, respectively. Pattern 2: Scar recurrences: these are new tumor growths over chemoreduced lesions. Pattern 3: New lesions: local dissemination of subretinal seeds leading to new lesions elsewhere in the retina. RESULTS: A noncomparative analysis of 64 eyes of 45 patients having 108 lesions was studied; of which 28/45 (62.22%) were male and 17/45 (37.78%) were female. The mean time of presentation since the first clinical sign was 40 days (range: 10-180). The most common sign at presentation was leukocoria 42/64 (65.6%), followed by squint 4/64 (6.34%). Nineteen patients (42.22%) had bilateral RB, while 26 patients (57.78%) had unilateral RB. Primary enucleation was done for 19/26 eyes with advanced unilateral disease. Out of the total 32 eyes with subretinal tumor seeds at presentation, 17/32 eyes had a recurrence in the form of new lesions (Pattern 3) and 22/32 eyes had scar recurrence (Pattern 2). All of these 32 eyes were salvaged by local tumor consolidation methods. Recurrence due to vitreous seed dissemination was found in 18/64 eyes, in which diffuse dissemination (Pattern 1b) was present in 8/18 eyes (44.4%); all required enucleation even after local and systemic chemotherapeutic measures. Rest 10/18 eyes with local vitreous seeds (Pattern 1a) were cured at the end of the follow-up. Globe salvage was more with Pattern 1a rather than Pattern 1b even after additional intravitreal chemotherapy. CONCLUSION: All eyes with Patterns 2 and 3 were salvaged at the end of follow-up with local tumor consolidation methods, while the globe salvage rate with Pattern 1 was poor even with multiple doses of intravitreal chemotherapy. The rate of successful treatment for managing these recurrence patterns depends on early identification by regular follow-ups with detailed retina examination.
Asunto(s)
Neoplasias de la Retina , Retinoblastoma , Humanos , Femenino , Masculino , Retinoblastoma/diagnóstico , Retinoblastoma/epidemiología , Retinoblastoma/terapia , Cicatriz , Estudios Retrospectivos , Retina , Neoplasias de la Retina/diagnóstico , Neoplasias de la Retina/epidemiología , Neoplasias de la Retina/terapiaRESUMEN
We report an unusual presentation of a 10-month-old girl with left eye (LE) redness and watering. Evaluation showed an iris vascular lesion and lens opacity in her LE. Child underwent USG B-scan and ultrasound biomicroscopy, by which an extensive mass lesion arising from iris and ciliary body with absent calcification was revealed. Following extensive evaluation, child underwent cataract extraction and trans-scleral total excision of the mass lesion. Histopathology proved it as juvenile xanthogranuloma (JXG) with vascular proliferation. JXG is a rare benign self-limiting dermatologic disorder affecting mainly infants and small children. Ocular lesions are the most common extracutaneous manifestation. Cataract in JXG is less frequently reported. This case is reported due to its rarity and as it presented solely as an intraocular lesion with combined diffuse infiltration into ciliary body and cataract which is unusual. Early recognition and systematic approach helped in sight saving and organ salvaging.
RESUMEN
Retinopathy of prematurity (ROP) is a leading cause of childhood blindness. Not only do the epidemiologic determinants and distributions of patients with ROP vary worldwide, but clinical differences have also been described. The Third Edition of the International Classification of ROP (ICROP3) acknowledges that aggressive ROP (AROP) can occur in larger preterm infants and involve areas of the more anterior retina, particularly in low-resource settings with unmonitored oxygen supplementation. As sub-specialty training programs are underway to address an epidemic of ROP in sub-Saharan Africa, recognizing characteristic retinal pathology in preterm infants exposed to unmonitored supplemental oxygen is important to proper diagnosis and treatment. This paper describes specific features associated with various ROP presentations: oxygen-induced retinopathy in animal models, traditional ROP seen in high-income countries with modern oxygen management, and ROP related to excessive oxygen supplementation in low- and middle-income countries: oxygen-associated ROP (OA-ROP).
RESUMEN
Importance: Retinopathy of prematurity (ROP) telemedicine screening programs have been found to be effective, but they rely on widefield digital fundus imaging (WDFI) cameras, which are expensive, making them less accessible in low- to middle-income countries. Cheaper, smartphone-based fundus imaging (SBFI) systems have been described, but these have a narrower field of view (FOV) and have not been tested in a real-world, operational telemedicine setting. Objective: To assess the efficacy of SBFI systems compared with WDFI when used by technicians for ROP screening with both artificial intelligence (AI) and human graders. Design, Setting, and Participants: This prospective cross-sectional comparison study took place as a single-center ROP teleophthalmology program in India from January 2021 to April 2022. Premature infants who met normal ROP screening criteria and enrolled in the teleophthalmology screening program were included. Those who had already been treated for ROP were excluded. Exposures: All participants had WDFI images and from 1 of 2 SBFI devices, the Make-In-India (MII) Retcam or Keeler Monocular Indirect Ophthalmoscope (MIO) devices. Two masked readers evaluated zone, stage, plus, and vascular severity scores (VSS, from 1-9) in all images. Smartphone images were then stratified by patient into training (70%), validation (10%), and test (20%) data sets and used to train a ResNet18 deep learning architecture for binary classification of normal vs preplus or plus disease, which was then used for patient-level predictions of referral warranted (RW)- and treatment requiring (TR)-ROP. Main Outcome and Measures: Sensitivity and specificity of detection of RW-ROP, and TR-ROP by both human graders and an AI system and area under the receiver operating characteristic curve (AUC) of grader-assigned VSS. Sensitivity and specificity were compared between the 2 SBFI systems using Pearson χ2testing. Results: A total of 156 infants (312 eyes; mean [SD] gestational age, 33.0 [3.0] weeks; 75 [48%] female) were included with paired examinations. Sensitivity and specificity were not found to be statistically different between the 2 SBFI systems. Human graders were effective with SBFI at detecting TR-ROP with a sensitivity of 100% and specificity of 83.49%. The AUCs with grader-assigned VSS only were 0.95 (95% CI, 0.91-0.99) and 0.96 (95% CI, 0.93-0.99) for RW-ROP and TR-ROP, respectively. For the AI system, the sensitivity of detecting TR-ROP sensitivity was 100% with specificity of 58.6%, and RW-ROP sensitivity was 80.0% with specificity of 59.3%. Conclusions and Relevance: In this cross-sectional study, 2 different SBFI systems used by technicians in an ROP screening program were highly sensitive for TR-ROP. SBFI systems with AI may be a cost-effective method to improve the global capacity for ROP screening.