Deep Retinal Image Understanding

Abstract

Abstract. This paper presents Deep Retinal Image Understanding (DRIU), a unified framework of retinal image analysis that provides both retinal vessel and optic disc segmentation. We make use of deep Convolutional Neural Networks (CNNs), which have proven revolutionary in other fields of computer vision such as object detection and image classification, and we bring their power to the study of eye fundus images. DRIU uses a base network architecture on which two set of specialized layers are trained to solve both the retinal vessel and optic disc segmentation. We present experimental validation, both qualitative and quantitative, in four public datasets for these tasks. In all of them, DRIU presents super-human performance, that is, it shows results more consistent with a gold standard than a second human annotator used as control.

A survey on deep learning in medical image analysis

“…Kisilev et al (2016) used a completely different approach and predicted categorical BI-RADS descriptors for breast lesions. In their work they focused on three descriptors used in mammography: shape, margin, and density, Fu et al (2016a) Blood vessel segmentation; extending the approach by Fu et al (2016b) by reformulating CRF as RNN Mahapatra et al (2016) Image quality assessment; classification output using CNN-based features combined with the output using saliency maps Maninis et al (2016) Segmentation of blood vessels and optic disk; VGG-19 network extended with specialized layers for each segmentation task Wu et al (2016) Blood vessel segmentation; patch-based CNN followed by mapping PCA solution of last layer feature maps to full segmentation Zilly et al (2017) Segmentation of the optic disk and the optic cup; simple CNN with filters sequentially learned using boosting Color fundus images: detection of abnormalities and diseases Chen et al (2015d) Glaucoma detection; end-to-end CNN, the input is a patch centered at the optic disk Abràmoff et al (2016) Diabetic retinopathy detection; end-to-end CNN, outperforms traditional method, evaluated on a public dataset Burlina et al (2016) Age-related macular degeneration detection; uses overfeat pretrained network for feature extraction van Grinsven et al (2016) Hemorrhage detection; CNN dynamically trained using selective data sampling to perform hard negative mining Gulshan et al (2016) Diabetic retinopathy detection; Inception network, performance comparable to a panel of seven certified ophthalmologists Hard exudate detection; end-to-end CNN combined with the outputs of traditional classifiers for detection of landmarks Worrall et al (2016) Retinopathy of prematurity detection; fine-tuned ImageNet trained GoogLeNet, feature map visualization to highlight disease Work in other imaging modalities Gao et al (2015) Cataract classification in slit lamp images; CNN followed by a set of recursive neural networks to extract higher order features Schlegl et al (2015) Fluid segmentation in OCT; weakly supervised CNN improved with semantic descriptors from clinical reports Blood vessel segmentation in OCT angiography; simple CNN, segmentation of several capillary networks where each have their own class label. The system was fed with the image data and region proposals and predicts the correct label for each descriptor (e.g.…”

Retinal Vessels Segmentation Techniques and Algorithms: A Survey

“…Conversely, other retinal anatomical structures have high contrast to other background tissues but with indistinct features in comparison with abnormal structures; optic disc and exudates lesions represent typical examples. All these challenges, in terms of medical image processing, make the classical segmentation techniques such as Sobel operators [26], Prewitt operators [27], gradient operators [28], and Robert and Krish differential operations [29] inefficient This challenge opens the room for a field of research specialized in detecting and segmenting thin (filamentary) retinal vascular structures, as in [18][19][20][21][22][23][24][25]. Secondly, Vessels identification in pathological retinal images faces a tension between accurate vascular structure extraction and false responses near pathologies (such as hard and soft exudates, hemorrhages, microaneuryms and cotton wool spots) and other nonvascular structures (such as optic disc and fovea region).…”

Deep learning with convolutional neural networks for decoding and visualization of EEG pathology

“…To understand class-specific spectral characteristics in the EEG recordings, we analyzed band powers in five frequency ranges: delta (0-4 Hz), theta (4-8 Hz), alpha (8)(9)(10)(11)(12)(13)(14), low beta (14)(15)(16)(17)(18)(19)(20), high beta (20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30) and low gamma .…”

“…These ConvNets exploit the hierarchical structure present in many natural signals. Recently, deep ConvNets trained end-to-end were, for example, able to more accurately diagnose skin cancer types from images than human dermatologists [9] and could segment retinal vessels better than human annotators [10].Deep ConvNets are nowadays also being applied to EEG analyses, such as decoding task-related information from EEG [11][12][13][14][15][16]. We have recently developed and validated the Braindecode toolbox1 for this purpose, and showed that the performance of deep ConvNets trained end-to-end is comparable to that of algorithms using hand-engineered features to decode task-related information.…”

“…These ConvNets exploit the hierarchical structure present in many natural signals. Recently, deep ConvNets trained end-to-end were, for example, able to more accurately diagnose skin cancer types from images than human dermatologists [9] and could segment retinal vessels better than human annotators [10].…”