Detection of retinal abnormalities using smartphone-captured fundus images: a survey

Abstract

Several retinal pathologies lead to severe damages that may achieve vision lost. Moreover, some damages require expensive treatment, other ones are irreversible due to the lack of therapies. Therefore, early diagnoses are highly recommended to control ocular diseases. However, early stages of several ocular pathologies lead to the symptoms that cannot be distinguish by the patients. Moreover, ageing population is an important prevalence factor of ocular diseases which is the cases of most industrial counties. Further, this feature involves a lake of mobility which presents a limiting factor to perform periodical eye screening. Those constraints lead to a late of ocular diagnosis and hence important ocular pathology patients are registered. The forecast statistics indicates that affected population will be increased in coming years. Several devices allowing the capture of the retina have recently been proposed. They are composed by optical lenses which can be snapped on Smartphone, providing fundus images with acceptable quality. Thence, the challenge is to perform automatic ocular pathology detection on Smartphone captured fundus images that achieves higher performance detection while respecting timing constraint with respect to the clinical employment. This paper presents a survey of the Smartphonecaptured fundus image quality and the existing methods that use them for retinal structures and abnormalities detection. For this purpose, we first summarize the works that evaluate the Smartphone-captures fundus image quality and their FOV (field-of-view). Then, we report the capability to detect abnormalities and ocular pathologies from those fundus images. Thereafter, we propose a flowchart of processing pipeline of detecting methods from Smartphone captured fundus images and we investigate about the implementation environment required to perform the detection of retinal abnormalities.

Fast and efficient retinal blood vessel segmentation method based on deep learning network

“…Additionally, the provided results allow our dedicated future work to implement retinal vessel tree segmentation into mobile devices, as a higher interest in mobile health within the clinical context. Indeed, several mobile digital imaging devices have recently appeared, hence enabling the capture of fundus images with good quality on mobile devices [64]. Furthermore, the performed computational performances promote targeting future work for using higher resolution fundus images, in particular the ultra-wide field, and achieving real time implementations with higher segmentation accuracy.…”

Mobile‐aided screening system for proliferative diabetic retinopathy

“…Thereafter, due to the handheld aspect of the smartphone, the fundus images are always blurred and noised. 20 Thus, we apply the point spread function to the OD sub-image to enhance the image quality. 30,31 Then, the Contrast Limited Adaptive Histogram Equalization (CLAHE) approach 36,37 is applied, where the pre-processed OD is shown in Figure 3C1,C2.…”

“…The smartphone captured fundus images are legible between 86% and 100% on average and have an accepted quality between 93% and 100%. 20 In addition, several clinical studies have approved the ability to detect DR from smartphone-captured fundus images. [21][22][23] However, those images are characterized by a moderate quality, caused by the handheld aspect of the capture process.…”

The Usability and Trustworthiness of Medical Eye Images

“…for cataract detection [20,21], to identify high cholesterol levels [22,23], to diagnose concussions [24] and for glaucoma screening [23,25]. Akil and Elloumi [26] present a meta paper, investigating the image quality and diagnosis performance achieved in eight prior works using smartphones equipped with additional lenses for retinal examination. Most recently deep learning has been presented as an approach to identify eye diseases [13,14,15,16].…”