Plus Disease in Retinopathy of Prematurity: Convolutional Neural Network Performance Using a Combined Neural Network and Feature Extraction Approach

Abstract

Retinopathy of prematurity (ROP), a leading cause of childhood blindness, is diagnosed by clinical ophthalmoscopic examinations or reading retinal images. Plus disease, defined as abnormal tortuosity and dilation of the posterior retinal blood vessels, is the most important feature to determine treatment-requiring ROP. We aimed to create a complete, publicly available and feature-extraction-based pipeline, I-ROP ASSIST, that achieves convolutional neural network (CNN)-like performance when diagnosing plus disease from retinal images. Methods: We developed two datasets containing 100 and 5512 posterior retinal images, respectively. After segmenting retinal vessels, we detected the vessel centerlines. Then, we extracted features relevant to ROP, including tortuosity and dilation measures, and used these features in the classifiers including logistic regression, support vector machine and neural networks to assess a severity score for the input. We tested our system with fivefold cross-validation and calculated the area under the curve (AUC) metric for each classifier and dataset. Results: For predicting plus versus not-plus categories, we achieved 99% and 94% AUC on the first and second datasets, respectively. For predicting pre-plus or worse versus normal categories, we achieved 99% and 88% AUC on the first and second datasets, respectively. The CNN method achieved 98% and 94% for predicting two categories on the second dataset. Conclusions: Our system combining automatic retinal vessel segmentation, tracing, feature extraction and classification is able to diagnose plus disease in ROP with CNNlike performance. Translational Relevance: The high performance of I-ROP ASSIST suggests potential applications in automated and objective diagnosis of plus disease.

Early Diagnosis and Quantitative Analysis of Stages in Retinopathy of Prematurity Based on Deep Convolutional Neural Networks

“…Many experts tried to have more explainable DCNNs by combining DCNNs with traditional feature extractions. Similar work had been done by Mao et al 23 and Yildiz et al 24 in the diagnosis of plus disease.…”

Vascular tortuosity quantification as an outcome metric of the oxygen-induced retinopathy model of ischemic retinopathy

“…A previously published algorithm for calculating plus disease on fundus images of infants with ROP was adapted to quantify vascular tortuosity of all segmented images and the associated coordinates of the optic disk center. (8) In summary, all unique pixels representing vessels from each segmentation were extracted to generate a graph of vessel segments. Point-based and vessel-based features such as integrated curvature, cumulative tortuosity index, and overall curvature were calculated from these vessel graphs using methods previously described by To investigate the validity of vascular tortuosity as a quantifiable feature of OIR, an initial pilot study using 20 randomly selected images (10 NOX and 10 OIR) from age-matched mice sacrificed at P12, P17, and P25 of a previously published dataset.…”

DIAROP: Automated Deep Learning-Based Diagnostic Tool for Retinopathy of Prematurity

“…The system obtained an accuracy of diagnosis of 83.33%, sensitivity and specificity of 100%, and 71.43%, correspondingly. Later the authors of [41] proposed a pipeline called "I-ROP ASSIST" to differentiate among healthy and plus ROP diseases. The authors segmented the images using U-Net CNN and extracted handcrafted features from these segmented images to train several machine learning classifiers.…”