scholarly journals Evaluation of a deep learning image assessment system for detecting severe retinopathy of prematurity

2018 ◽  
Vol 103 (5) ◽  
pp. 580-584 ◽  
Author(s):  
Travis K Redd ◽  
John Peter Campbell ◽  
James M Brown ◽  
Sang Jin Kim ◽  
Susan Ostmo ◽  
...  
Keyword(s):  
Deep Learning ◽  
Retinopathy Of Prematurity ◽  
Predictive Value ◽  
Severity Score ◽  
Receiver Operating Curve ◽  
Fundus Photography ◽  
Image Analysis System ◽  
Reference Standard ◽  
Type 1 Rop

BackgroundPrior work has demonstrated the near-perfect accuracy of a deep learning retinal image analysis system for diagnosing plus disease in retinopathy of prematurity (ROP). Here we assess the screening potential of this scoring system by determining its ability to detect all components of ROP diagnosis.MethodsClinical examination and fundus photography were performed at seven participating centres. A deep learning system was trained to detect plus disease, generating a quantitative assessment of retinal vascular abnormality (the i-ROP plus score) on a 1–9 scale. Overall ROP disease category was established using a consensus reference standard diagnosis combining clinical and image-based diagnosis. Experts then ranked ordered a second data set of 100 posterior images according to overall ROP severity.Results4861 examinations from 870 infants were analysed. 155 examinations (3%) had a reference standard diagnosis of type 1 ROP. The i-ROP deep learning (DL) vascular severity score had an area under the receiver operating curve of 0.960 for detecting type 1 ROP. Establishing a threshold i-ROP DL score of 3 conferred 94% sensitivity, 79% specificity, 13% positive predictive value and 99.7% negative predictive value for type 1 ROP. There was strong correlation between expert rank ordering of overall ROP severity and the i-ROP DL vascular severity score (Spearman correlation coefficient=0.93; p<0.0001).ConclusionThe i-ROP DL system accurately identifies diagnostic categories and overall disease severity in an automated fashion, after being trained only on posterior pole vascular morphology. These data provide proof of concept that a deep learning screening platform could improve objectivity of ROP diagnosis and accessibility of screening.

Optimised retinopathy of prematurity screening guideline in China based on a 5-year cohort study

2020 ◽  
pp. bjophthalmol-2020-316401
Author(s):  
Qian Yang ◽  
Xiaohong Zhou ◽  
Yingqin Ni ◽  
Haidong Shan ◽  
Wenjing Shi ◽  
...  
Keyword(s):  
Cohort Study ◽  
Retinopathy Of Prematurity ◽  
Neonatal Intensive Care ◽  
Predictive Performance ◽  
Screening Guideline ◽  
Postmenstrual Age ◽  
The Mean ◽  
Type 1 Rop ◽  
Rop Screening

PurposesTo develop an optimised retinopathy of prematurity (ROP) screening guideline by adjusting the screening schedule and thresholds of gestational age (GA) and birth weight (BW).MethodsA multicentre retrospective cohort study was conducted based on data from four tertiary neonatal intensive care units in Shanghai, China. The medical records of enrolled infants, born from 2012 to 2016 who underwent ROP examinations, were collected and analysed. The incidence and risk factors for ROP were analysed in all infants. Postnatal age (PNA) and postmenstrual age (PMA) of infants, detected to diagnose ROP for the first time, were compared with the present examination schedule. The predictive performance of screening models was evaluated by internally validating sensitivity and specificity.ResultsOf the 5606 eligible infants, ROP was diagnosed in 892 (15.9%) infants; 63 (1.1%) of them received treatment. The mean GA of ROP patients was 29.4±2.4 weeks, and the mean BW was 1260±330 g. Greater prematurity was associated with an older PNA at which ROP developed. The minimum PMA and PNA at which diagnosis of treatable ROP occurred were 32.43 and 3 weeks, respectively. The optimised criteria (GA <32 weeks or BW <1600 g) correctly predicted 98.4% type 1 ROP infants, reducing the infants requiring examinations by 43.2% when internally validated.ConclusionsThe incidence of type 1 ROP and the mean GA and BW of ROP infants have decreased in China. The suggested screening threshold and schedule may be reliably used to guide the modification of ROP screening guideline and decrease medical costs.

scholarly journals Incidence and Characteristics of Infants with Retinopathy of Prematurity in Croatia

2018 ◽  
Vol 1 (1) ◽  
pp. 35-45
Author(s):  
Mirjana Vucinovic ◽  
Ljubo Znaor ◽  
Ana Vucinovic ◽  
Vesna Capkun ◽  
Julijana Bandic
Keyword(s):  
Risk Factors ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Neonatal Intensive Care Unit ◽  
Gestational Age ◽  
Retinopathy Of Prematurity ◽  
Small For Gestational Age ◽  
Neonatal Intensive Care ◽  
Type 1 Rop

PURPOSE: To study the incidence of retinopathy of prematurity (ROP) in a neonatal intensive care unit in Croatia and obtain information on risk factors associated with ROP. There have been limited studies on ROP in Croatia where the screening for ROP and its treatment is still insufficient and not introduced in many intensive care units. MATERIAL AND METHODS: This retrospective study included 247 premature infants admitted to the neonatal intensive care unit of University Hospital Split, over a 5-year period between January 2012, and December 2016. In this paper the relationship between clinical risk factors and the development of ROP was analyzed. RESULTS: The overall incidence for ROP was 23,9 % (59 infants), for Type 1 ROP was 9,3% (23 infants); for Type 2 ROP was 14,6% (36 infants). Median gestational age (GA) and birthweight (BW) were significantly lower among infants with ROP versus those without ROP (29: 23-34 vs. 31: 23-34,p<0,001 and 1,180:630-2,000 vs. 1485:590-2000, p<0,001 respectively). Multivariate analysis showed that only BW (p=0,029) and small for gestational age (SGA) (p=0,045) predicted the development of ROP. CONCLUSION: Birth weight and small for gestational age were the most significant risk factors for developing ROP. In comparison with studies from highly developed countries, infants with a much wider range of gestational age and birth weights are developing Type 1 ROP. 

Systemic conbercept pharmacokinetics and VEGF pharmacodynamics following intravitreal injections of conbercept in patients with retinopathy of prematurity

2021 ◽  
pp. bjophthalmol-2021-319131
Author(s):  
Yong Cheng ◽  
Shuang Sun ◽  
Xun Deng ◽  
Xuemei Zhu ◽  
Dandan Linghu ◽  
...  
Keyword(s):  
Retinopathy Of Prematurity ◽  
Intravitreal Injections ◽  
Vascular Endothelial ◽  
Randomised Study ◽  
Limit Of Quantitation ◽  
Significant Difference ◽  
Endothelial Growth Factor ◽  
Type 1 Rop

BackgroundData on serum vascular endothelial growth factor (VEGF) and drug levels in patients with retinopathy of prematurity (ROP) following intravitreal injections of conbercept (IVC) are lacking.MethodsMulticentre, prospective, non-randomised study of patients with aggressive posterior retinopathy of prematurity (APROP) or type 1 ROP who had not received other treatment. All infants received therapy in both eyes plus intravitreal IVC 0.25 mg/0.025 mL in one eye and had at least 6 months of follow-up. Blood samples were collected before and 1 week and 4 weeks after IVC. The main outcome measures were serum conbercept and VEGF concentrations.ResultsForty infants with APROP or type 1 ROP were enrolled. The mean serum VEGF at baseline and 1 week and 4 weeks after a total of 0.25 mg of IVC was 953.35±311.90 pg/mL, 303.46±181.89 pg/mL and 883.12±303.89 pg/mL, respectively. Serum VEGF 1 week after IVC was significantly lower (p<0.05) than baseline, and at 4 weeks after IVC, it was significantly higher (p<0.05) than at 1 week. There was no significant difference (p>0.05) between baseline and 4 weeks. Serum conbercept was below the limit of quantitation (BLOQ) at baseline and 4 weeks and was 19.81±7.60 ng/mL at 1 week.ConclusionSerum VEGF 1 week after IVC was significantly lower than baseline but returned to baseline at 4 weeks. Serum conbercept increased at 1 week and was BLOQ at 4 weeks.

Effect of Intravitreal Aflibercept on Central Retinal Arterial Blood Flow in Type 1 Retinopathy of Prematurity

2017 ◽  
Vol 27 (6) ◽  
pp. 751-755 ◽  
Author(s):  
Emine A. Sukgen ◽  
Gökhan Söker ◽  
Yusuf Koçluk ◽  
Bozkurt Gülek
Keyword(s):  
Blood Flow ◽  
Retinopathy Of Prematurity ◽  
Color Doppler ◽  
Color Doppler Imaging ◽  
Doppler Imaging ◽  
Intravitreal Aflibercept ◽  
Retinal Artery ◽  
The Mean ◽  
Type 1 Rop

Purpose To evaluate the blood flow changes of the central retinal artery measured with color Doppler imaging (CDI) in infants receiving intravitreal aflibercept (IVA) for treatment of type 1 retinopathy of prematurity (ROP). Methods Patients with type 1 ROP were assessed prospectively by CDI following IVA. Color Doppler imaging was used to measure the peak systolic velocity, end diastolic velocity (EDV), pulsatility index (PI), and resistivity index (RI) of the central retinal artery (CRA) before IVA injection and 1 hour, 1 week, and 1 month after injection. Results A total of 29 eyes of 15 infants were included in this study. The mean gestational age at birth was 28.62 ± 2.48 weeks and the mean birthweight was 1,198.62 ± 348.99 g. All treated eyes showed complete regression of ROP and peripheral retinal vascularization continued. Measurements of EDV-CRA, RI-CRA, and PI-CRA showed significant changes after IVA treatment. Conclusions This study showed that IVA is an effective treatment for type 1 ROP. After IVA treatment, vascular resistance increases, ocular blood flow decreases, and changes in hemodynamic parameters of CRA may remain for a month. Further studies are needed to evaluate the effect of anti-vascular endothelial growth factor agents on ocular hemodynamics in infants with ROP.

scholarly journals Asymmetric Outcomes of Type 1 Retinopathy of Prematurity after Bilateral Intravitreal Ranibizumab Treatment

2017 ◽  
Vol 2017 ◽  
pp. 1-8
Author(s):  
Qiujing Huang ◽  
Qi Zhang ◽  
Yu Xu ◽  
Xunda Ji ◽  
Ping Fei ◽  
...  
Keyword(s):  
Retinal Detachment ◽  
Retinopathy Of Prematurity ◽  
Vitreous Hemorrhage ◽  
High Rate ◽  
Intravitreal Ranibizumab ◽  
Outcome Group ◽  
Significant Difference ◽  
Type 1 Rop ◽  
Ranibizumab Treatment

Purpose. To present cases with retinopathy of prematurity (ROP), who were treated with intravitreal injection of ranibizumab (IVR) and had unpredictable asymmetric outcomes. Methods. A retrospective review was performed in infants with type 1 ROP and had bilateral IVR (0.25 mg/0.025 mL) as initial treatment. Patients were classified into the asymmetric outcome group and the symmetric outcome group. Results. Eighty-four patients (168 eyes) were included. There were 18 eyes of 9 patients (10.7%) in the asymmetric outcome group and 150 eyes of 75 patients (89.3%) in the symmetric outcome group. In the symmetric outcome group, 86 eyes (57.3%) had ROP regression, 60 eyes (40%) had reactivation requiring laser treatment, and 4 eyes (2.7%) progressed to retinal detachment requiring vitrectomy. In the asymmetric outcome group, one of the eyes of the 9 patients had ROP regression with/without reactivation after IVR, while the contralateral eyes had negative response, including remarkable posterior fibrosis, partial or total retinal detachment, and vitreous hemorrhage. There was statistically significant difference between the birth weight of the two groups. Conclusion. Contralateral eyes with ROP can take a different clinical course after ranibizumab treatment. High rate of reactivation after IVR is another concern that ophthalmologists should pay attention to.

scholarly journals Time to regain birth weight - a marker to predict the severity of retinopathy of prematurity?

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Ajay Anvekar ◽  
Sam Athikarisamy ◽  
Shripada Rao ◽  
Andy Gill ◽  
Elizabeth Nathan ◽  
...  
Keyword(s):  
Preterm Infants ◽  
Gestational Age ◽  
Retinopathy Of Prematurity ◽  
Oxygen Therapy ◽  
Control Groups ◽  
Gestational Age At Birth ◽  
And Control ◽  
Type 1 Rop ◽  
Age At Birth

Abstract Background Poor weight gain in the first few weeks of life has been studied as a predictor of retinopathy of prematurity (ROP). Our aim was to assess whether time taken to regain birthweight (BW) be used as an additional marker to identify infants with type 1 ROP. Methods In this retrospective study, preterm infants (< 27 weeks gestational age at birth) born during the period from 1/1/2010–31/12/2015 at a tertiary neonatal intensive care unit in Australia were included. Twenty-seven preterm infants with Type 1 ROP were identified. Controls (No ROP or ROP other than type 1) were matched with cases on gestational age at birth and BW (1:4 ratio). Data were collected from the database and medical records. Results The median (IQR) gestational age for Type 1 ROP and control groups were 24 (24–26) and 25 (24–26) weeks respectively and median (IQR) BW for Type 1 ROP and control groups were 675 (635–810) and 773 (666–884) grams respectively. Preterm infants with Type 1 ROP were more likely to be small for gestational age (SGA) (18.5% vs 3.7%, p = 0.015) and had increased weeks on oxygen therapy (median 11.9 vs 9.1, p = 0.028). Time to regain BW was longer in preterm infants with type 1 ROP than controls but did not reach statistical significance (median 9 vs 7 days, OR 1.08, 95% CI 1.00–1.17, p = 0.059) adjusted for SGA and duration of oxygen therapy. The area under the curve from the time to regain BW model with adjustment for SGA and duration of oxygen therapy was 0.73 (95% CI 0.62–0.83). Conclusion We hypothesize that time to regain BW has potential to aid prediction of Type 1 ROP and this warrants further investigation in a larger prospective study.

scholarly journals Severe Retinopathy of Prematurity Is Not Independently Associated With Worse Neurodevelopmental Outcomes in Preterm Neonates

2021 ◽  
Vol 9 ◽  
Author(s):  
Marie Altendahl ◽  
Myung Shin Sim ◽  
Artemiy Kokhanov ◽  
Bradley Gundlach ◽  
Irena Tsui ◽  
...  
Keyword(s):  
Health Insurance ◽  
Retinopathy Of Prematurity ◽  
Preterm Neonates ◽  
Neurodevelopmental Outcomes ◽  
Age Cohort ◽  
Male Sex ◽  
The Relationship ◽  
Type 1 Rop

Purpose: To evaluate the relationship between retinopathy of prematurity (ROP) severity and neurodevelopmental outcomes in premature neonates at 0–36 months corrected age.Methods: A retrospective chart review was performed on 228 neonates screened for ROP at the UCLA Mattel Children's Hospital between 2011 and 2018. Demographic information, clinical outcomes, ROP severity (no ROP, type 1 ROP, type 2 ROP), and Bayley-III neurodevelopmental scores were collected. Infants were grouped into corrected age cohorts (0–12, 12–24, and 24–36 months) to assess neurodevelopmental outcomes with increasing age. Within each age cohort, ANOVA and Chi-Square testing were used to detect differences in birth characteristics and neurodevelopmental scores between infants with type 1 ROP, type 2 ROP, or no ROP. Univariable analyses assessed the relationship between ROP severity and neurodevelopmental outcomes within each age cohort. A multivariable analysis was then performed to determine if ROP severity remained significantly associated with worse neurodevelopmental scores after controlling for birth weight (BW), intraventricular hemorrhage grade (IVH), health insurance type, male sex, and age at Bayley testing.Results: Without controlling for factors associated with prematurity, neonates with type 1 ROP had poorer cognition (p = 0.001) and motor (p = 0.006) scores at ages 0–12 months and poorer cognition (p = 0.01), language (p = 0.04) and motor (p = 0.04) scores at ages 12–24 months than infants without ROP, but no significant differences were detected at ages 24–36 months. After adjusting for BW, IVH, insurance type, male sex, and age at Bayley testing, ROP severity was no longer associated with worse neurodevelopmental scores in any domain.Conclusion: This study emphasizes that poorer neurodevelopmental outcomes in preterm neonates are most likely related to lower birthweight, associated co-morbidities of prematurity, and socioeconomic factors such as health insurance, not severity of ROP itself.

scholarly journals Diagnostic Accuracy of Direct Ophthalmoscopy and Non-Mydriatic Retinal Photography by Trained Optometrists for Screening of Diabetic Retinopathy

2020 ◽  
Vol 36 (2) ◽  
Author(s):  
Muhammad Saleh Memon ◽  
Shahid Ahsan ◽  
Muhammad Fahadullah ◽  
Khalida Parveen ◽  
Sumaira Salim ◽  
...  
Keyword(s):  
Diabetic Retinopathy ◽  
Predictive Value ◽  
Fundus Photography ◽  
Slit Lamp ◽  
Reference Standard ◽  
Cross Sectional ◽  
Fundus Camera ◽  
Retinal Photography ◽  
Direct Ophthalmoscopy

Purpose:  To determine the reliability of direct ophthalmoscopy and Non-Mydriatic fundus photography for screening of Diabetic Retinopathy by optometrist. Study Design:  Observational, cross sectional. Place and Duration of Study:  Al-Ibrahim eye hospital, Karachi from October to December 2018. Material and Methods:  All individuals with type 2 diabetes of≥ 40 years of age were screened for diabetic retinopathy (DR) by two trained optometrists and an ophthalmologist. First Optometrist used Non Mydriatic Fundus Camera (NMFC) and second optometrist used direct ophthalmoscopy (DO) after dilating the pupils. Final examination was done by the Ophthalmologist with slit lamp using Volk fundus lens which was considered as reference standard. Every investigator was kept unaware of the findings of others. Results:  A total of 698 eyes of 349 respondents were screened. Ophthalmologist could not make decision by bio microscopy in 44 (6.3%) individuals as compared to 128 (18.3%) by 1st optometrist by NMFC and 142 (20.3%) by 2nd optometrist with DO. Diabetic retinopathy (DR) was diagnosed in 1493 (21.4%) with slit lamp, in 1172 (16.8%) with NMFC and in 1333 (19.1%) with DO. Sensitivity of NMFC was 76% and that of DO was 64.8%. Specificity of NMFC was 97.45% and that of DO was 96.63%.  Positive predictive value (PPV) of NMFC was 89.33% and that of DO was 84.3% Negative predictive value (NPV) of NMFC was 93.33% and that of DO was 90.7%. Conclusion:  NMFC is recommended tool for DR screening; but DO by well-trained optometrist can be reliable where neither ophthalmologist nor NMFC is available.

scholarly journals Incidence, timing and risk factors of type 1 retinopathy of prematurity in a North American cohort

2020 ◽  
pp. bjophthalmol-2020-317467
Author(s):  
Yinxi Yu ◽  
Lauren A Tomlinson ◽  
Gil Binenbaum ◽  
Gui-shuang Ying
Keyword(s):  
Risk Factors ◽  
Premature Infants ◽  
Retinopathy Of Prematurity ◽  
Characteristic Curve ◽  
Time Windows ◽  
Secondary Analysis ◽  
Prenatal Factors ◽  
The Mean ◽  
Type 1 Rop

Background/AimsEarly detection and timely treatment of type 1 retinopathy of prematurity (ROP) can reduce the risk of blindness. To evaluate the incidence, timing and risk factors of type 1 ROP in a large, broad-risk cohort of premature infants.MethodsSecondary analysis of data from the two Postnatal Growth and Retinopathy of Prematurity studies. Main outcomes are the incidence and timing of type 1 ROP.ResultsAmong 11 463 infants (mean birth weight (BW), 1095 g; mean gestational age (GA), 28 weeks), 677 (5.9%, 95% CI 5.5% to 6.3%) developed type 1 ROP. Rate of type 1 ROP decreased with larger GA (28.8% for GA ≤23 weeks, 0.2% for GA of 31–32 weeks) and no infants with GA >32 weeks developed type 1 ROP. Type 1 ROP was first diagnosed at a median postmenstrual age (PMA) of 36 weeks (range 30–46 weeks) or postnatal age (PNA) of 11 weeks (range 5–21 weeks). The mean PMA at diagnosis of type 1 ROP increased with GA (35 weeks for GA of 22–24 weeks, 41 weeks for GA of 29–30 weeks), but the mean PNA at diagnosis of type 1 ROP was similar (11–13 weeks) across GA of 22–29 weeks. GA and BW dominate the association (area under the receiver operating characteristic curve=0.87, 95% CI 0.86 to 0.88).ConclusionsType 1 ROP developed in about 6% of premature infants over wide time windows in terms of both PMA and PNA. BW and GA are the dominant risk factors for type 1 ROP, while other prenatal factors add minimal predictive power for type 1 ROP.

Refractive Error in Patients with Retinopathy of Prematurity after Laser Photocoagulation or Bevacizumab Monotherapy

2015 ◽  
Vol 234 (4) ◽  
pp. 211-217 ◽  
Author(s):  
Hsi-Kung Kuo ◽  
I-Ting Sun ◽  
Mei-Yung Chung ◽  
Yi-Hao Chen
Keyword(s):  
Refractive Error ◽  
Retinopathy Of Prematurity ◽  
Laser Photocoagulation ◽  
Group 4 ◽  
Group 3 ◽  
The Mean ◽  
Group 2 ◽  
Type 1 Rop ◽  
Group 1

Purpose: To evaluate the refractive development of premature infants with retinopathy of prematurity (ROP) after treatment with laser photocoagulation or intravitreal injection of bevacizumab (IVB). Methods: The medical records of patients with ROP treated between 2003 and 2012 who underwent yearly follow-ups were retrospectively reviewed. Patients with residual ROP abnormalities were excluded. The cycloplegic refraction at 3 years of age, assessed using an autorefractometer, was recorded. Results: In total, 54 eyes from 54 patients were enrolled. Patients were divided into 4 groups: group 1, including 14 eyes of 14 patients treated with laser therapy; group 2, 15 eyes of 15 patients treated with IVB; group 3, 13 eyes of 13 patients with non-type 1 ROP under conservative follow-up, and group 4, 12 eyes of 12 premature patients without ROP. The mean spherical equivalent at 3 years of age was -1.71 ± 1.27 dpt in group 1, -1.53 ± 2.20 dpt in group 2, 0.63 ± 1.37 dpt in group 3, and 0.41 ± 1.95 dpt in group 4. The mean refractive error differed significantly among the 4 groups (p < 0.001). Patients in groups 1 and 2 were more prone to myopia compared with those in groups 3 and 4. Furthermore, patients with type 1 ROP treated by laser photocoagulation (group 1) and those treated by IVB (group 2) had similar refraction (p = 1). Conclusions: The results of this study suggest that treatment-demanding ROP eyes are susceptible to more severe myopia with age compared with eyes without ROP or those with spontaneously regressed ROP. In addition, the myopic status between laser and IVB treatment did not differ statistically.

Sign in / Sign up

Export Citation Format

Share Document