Comprehensive Retinal Image Analysis for Aggressive Posterior Retinopathy of Prematurity

Development and validation of an artificial intelligence based screening tool for detection of retinopathy of prematurity in a South Indian population

Purpose

The primary objective of this study was to develop and validate an AI algorithm as a screening tool for the detection of retinopathy of prematurity (ROP).

Participants

Images were collected from infants enrolled in the KIDROP tele-ROP screening program.

Methods

We developed a deep learning (DL) algorithm with 227,326 wide-field images from multiple camera systems obtained from the KIDROP tele-ROP screening program in India over an 11-year period. 37,477 temporal retina images were utilized with the dataset split into train (n = 25,982, 69.33%), validation (n = 4,006, 10.69%), and an independent test set (n = 7,489, 19.98%). The algorithm consists of a binary classifier that distinguishes between the presence of ROP (Stages 1-3) and the absence of ROP. The image labels were retrieved from the daily registers of the tele-ROP program. They consist of per-eye diagnoses provided by trained ROP graders based on all images captured during the screening session. Infants requiring treatment and a proportion of those not requiring urgent referral had an additional confirmatory diagnosis from an ROP specialist.

Results

Of the 7,489 temporal images analyzed in the test set, 2,249 (30.0%) images showed the presence of ROP. The sensitivity and specificity to detect ROP was 91.46% (95% CI: 90.23%-92.59%) and 91.22% (95% CI: 90.42%-91.97%), respectively, while the positive predictive value (PPV) was 81.72% (95% CI: 80.37%-83.00%), negative predictive value (NPV) was 96.14% (95% CI: 95.60%-96.61%) and the AUROC was 0.970.

Conclusion

The novel ROP screening algorithm demonstrated high sensitivity and specificity in detecting the presence of ROP. A prospective clinical validation in a real-world tele-ROP platform is under consideration. It has the potential to lower the number of screening sessions required to be conducted by a specialist for a high-risk preterm infant thus significantly improving workflow efficiency.

A Novel Method to Improve Inter-Clinician Variation in the Diagnosis of Retinopathy of Prematurity Using Machine Learning

PURPOSE

Inter-clinician variation could cause uncertainty in disease management. This is reported to be high in Retinopathy of Prematurity (ROP), a potentially blinding retinal disease affecting premature infants. Machine learning has the potential to quantify the differences in decision-making between ROP specialists and trainees and may improve the accuracy of diagnosis.

METHODS

An anonymized survey of ROP images was administered to the expert(s) and the trainee(s) using a study-designed user interface. The results were analyzed for repeatability as well as to identify the level of agreement in the classification. "Ground truth" was prepared for each individual and a unique classifier was built for each individual using the same. The classifier allowed the identification of the most important features used by each individual.

RESULTS

Correlation and disagreement between the expert and the trainees were visualized using the Dipstick™ diagram. Intra-clinician repeatability and reclassification statistics were assessed for all. The repeatability was 88.4% and 86.2% for two trainees and 92.1% for the expert, respectively. Commonly used features differed for the expert and the trainees and accounted for the variability.

CONCLUSION

This novel, automated algorithm quantifies the differences using machine learning techniques. This will help audit the training process by objectively measuring differences between experts and trainees.

TRANSLATIONAL RELEVANCE

Training for image-based ROP diagnosis can be more objectively performed using this novel, machine learning-based automated image analyzer and classifier.

Imaging the pediatric retina: An overview

Recent decade has seen a shift in the causes of childhood blinding diseases from anterior segment to retinal disease in both developed and developing countries. The common retinal disorders are retinopathy of prematurity and vitreoretinal infections in neonates, congenital anomalies in infants, and vascular retinopathies including type 1 diabetes, tumors, and inherited retinal diseases in children (up to 12 years). Retinal imaging helps in diagnosis, management, follow up and prognostication in all these disorders. These imaging modalities include fundus photography, fluorescein angiography, ultrasonography, retinal vascular and structural studies, and electrodiagnosis. Over the decades there has been tremendous advances both in design (compact, multifunctional, tele-consult capable) and technology (wide- and ultra-wide field and noninvasive retinal angiography). These new advances have application in most of the pediatric retinal diseases though at most times the designs of new devices have remained confined to use in adults. Poor patient cooperation and insufficient attention span in children demand careful crafting of the devices. The newer attempts of hand-held retinal diagnostic devices are welcome additions in this direction. While much has been done, there is still much to do in the coming years. One of the compelling and immediate needs is the pediatric version of optical coherence tomography angiography. These needs and demands would increase many folds in future. A sound policy could be the simultaneous development of adult and pediatric version of all ophthalmic diagnostic devices, coupled with capacity building of trained medical personnel.

Current Application of Digital Diagnosing Systems for Retinopathy of Prematurity

BACKGROUND AND OBJECTIVE

Retinopathy of prematurity (ROP), a proliferative vascular eye disease, is one of the leading causes of blindness in childhood and prevails in premature infants with low-birth-weight. The recent progress in digital image analysis offers novel strategies for ROP diagnosis. This paper provides a comprehensive review on the development of digital diagnosing systems for ROP to software researchers. It may also be adopted as a guide to ophthalmologists for selecting the most suitable diagnostic software in the clinical setting, particularly for the remote ophthalmic support.

METHODS

We review the latest literatures concerning the application of digital diagnosing systems for ROP. The diagnosing systems are analyzed and categorized. Articles published between 1998 and 2020 were screened with the two searching engines Pubmed and Google Scholar.

RESULTS

Telemedicine is a method of remote image interpretation that can provide medical service to remote regions, and yet requires training to local operators. On the basis of image collection in telemedicine, computer-based image analytical systems for ROP were later developed. So far, the aforementioned systems have been mainly developed by virtue of classic machine learning, deep learning (DL) and multiple machine learning. During the past two decades, various computer-aided systems for ROP based on classic machine learning (e.g. RISA, ROPtool, CAIER) became available and have achieved satisfactory performance. Further, automated systems for ROP diagnosis based on DL are developed for clinical applications and exhibit high accuracy. Moreover, multiple instance learning is another method to establish an automated system for ROP detection besides DL, which, however, warrants further investigation in future.

CONCLUSION

At present, the incorporation of computer-based image analysis with telemedicine potentially enables the detection, supervision and in-time treatment of ROP for the preterm babies.

Retinal immaturity at first screening and retinopathy of prematurity: Image-based validation of 1202 eyes of premature infants to predict disease progression

Purpose:

To use the extent of retinal immaturity at the first visit to predict progression to any stage and treatment-requiring retinopathy of prematurity (ROP).

Methods:

Retrospective, multicenter, nonrandomized, observational, clinical, validation study. In all, 601 Asian Indian preterm infants born < 2000 g and/or < 34 weeks of gestation completing ROP screening with RetCam images taken during each visit were included. A total of 1202 eyes of these infants were classified into three groups based on the retinal immaturity at the first screening visit into “mild” (Group 1), vessels reaching the posterior boundary of zone 3; “moderate” (Group 2), vessels entering zone 2 anterior; and “severe” (Group 3), vessels in zone 1 or zone 2 posterior. RetCam images at each subsequent visit were evaluated and the proportion of eyes that progressed to Type 1 or Type 2 ROP was correlated with the degree of retinal immaturity.

Results:

Of the 958 eyes in Group 1, 200 eyes in Group 2, and 44 eyes in Group 3, any stage ROP developed in 15% of eyes in Group 1, 46.5% of eyes in Group 2, and 100% of eyes in Group 3 (P < 0.001). Sixteen of 128 eyes (12.5%), 12 of 72 (16.6%), and 28 of 44 of eyes (63.6%) in Groups 1, 2, and 3, respectively, required treatment (P < 0.001).

Conclusion:

Retinal immaturity at first screening visit predicts Type 1 and Type 2 ROP. “Severe” immaturity is more likely to progress to “treatment-requiring” disease. This could be a useful tool for prognostication, counseling, and scheduling follow-up.

A computer-aided diagnosis system for plus disease in retinopathy of prematurity with structure adaptive segmentation and vessel based features

Retinopathy of Prematurity (ROP) is a blinding disease affecting the retina of low birth-weight preterm infants. Accurate diagnosis of ROP is essential to identify treatment-requiring ROP, which would help to prevent childhood blindness. Plus disease, which characterizes abnormal twisting, widening and branching of the blood vessels, is a significant symptom of treatment requiring ROP. In this paper, we have developed and evaluated a computer-based analysis system for objective assessment of plus disease in ROP, which best mimics the clinical method of disease diagnosis by identifying unique vessel based features. The proposed system consists of an initial segmentation stage, which will efficiently extract blood vessels of varying width and length by utilizing structure adaptive filtering, connectivity analysis and image fusion. The paper proposes the usage of additional retinal features namely leaf node count and vessel density, to portray the abnormal growth and branching of the blood vessels and to complement the commonly used features namely tortuosity and width. The test results show a better classification of plus disease in terms of sensitivity (95%) and specificity (93%), emphasizing the superiority of the proposed segmentation algorithm and vessel-based features. An additional advantage of the proposed system is that the process of selection of relevant vessels for feature extraction is fully automated, which makes the system highly useful to the non-physician graders, owing to the unavailability of a sufficient number of ROP specialists.