Automated diabetic retinopathy grading and lesion detection based on the modified R-FCN object-detection algorithm

In this work, we develop a computer-aided retinal image screening system that can perform automated diabetic retinopathy (DR) grading and DR lesion detection in retinal fundus images. We propose a modified object-detection method for this task via a region-based fully convolutional network (R-FCN)....

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IET computer vision Jg. 14; H. 1; S. 1 - 8
Hauptverfasser: Wang, Jialiang, Luo, Jianxu, Liu, Bin, Feng, Rui, Lu, Lina, Zou, Haidong
Format: Journal Article
Sprache:Englisch
Veröffentlicht: The Institution of Engineering and Technology 01.02.2020
Wiley
Schlagworte:
biomedical optical imaging
blood vessels
computer-aided retinal image screening system
diseases
DR-grading model
eye
feature extraction
feature pyramid network
hospital data
image classification
image segmentation
lesion detection
medical image processing
Messidor data
modified object-detection method
modified R-FCN lesion-detection model
modified R-FCN object-detection algorithm
modified region proposal network
object detection
region-based fully convolutional network
Research Article
retinal fundus images
Shanghai Eye Hospital
hospital data
image classification
modified R-FCN lesion-detection model
blood vessels
lesion detection
diseases
object detection
DR-grading model
modified region proposal network
eye
Shanghai Eye Hospital
retinal fundus images
computer-aided retinal image screening system
feature extraction
image segmentation
feature pyramid network
region-based fully convolutional network
Messidor data
modified R-FCN object-detection algorithm
modified object-detection method
medical image processing
biomedical optical imaging
ISSN:1751-9632, 1751-9640, 1751-9640
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In this work, we develop a computer-aided retinal image screening system that can perform automated diabetic retinopathy (DR) grading and DR lesion detection in retinal fundus images. We propose a modified object-detection method for this task via a region-based fully convolutional network (R-FCN). A feature pyramid network and a modified region proposal network are applied to enhance the detection of small objects. The DR-grading model based on the modified R-FCN is evaluated on the Messidor data set and images provided by the Shanghai Eye Hospital. High sensitivity of 99.39% and specificity of 99.93% are obtained on the hospital data. Moreover, high sensitivity of 92.59% and specificity of 96.20% are obtained on the Messidor data set. The modified R-FCN lesion-detection model is validated on the hospital data set and achieves a 92.15% mean average precision. The proposed R-FCN can efficiently accomplish DR grading and lesion detection with high accuracy.
ISSN:1751-9632
1751-9640
1751-9640
DOI:10.1049/iet-cvi.2018.5508