The efficacy of automated "disease/no disease'' grading for diabetic retinopathy in a systematic screening programme

Biomedical Physics and Grampian Retinal Screening Programme, University of Aberdeen, Foresterhill, Aberdeen.
British Journal of Ophthalmology (Impact Factor: 2.81). 12/2007; 91(11):1512-7. DOI: 10.1136/bjo.2007.119453
Source: PubMed

ABSTRACT To assess the efficacy of automated "disease/no disease" grading for diabetic retinopathy within a systematic screening programme.
Anonymised images were obtained from consecutive patients attending a regional primary care based diabetic retinopathy screening programme. A training set of 1067 images was used to develop automated grading algorithms. The final software was tested using a separate set of 14 406 images from 6722 patients. The sensitivity and specificity of manual and automated systems operating as "disease/no disease" graders (detecting poor quality images and any diabetic retinopathy) were determined relative to a clinical reference standard.
The reference standard classified 8.2% of the patients as having ungradeable images (technical failures) and 62.5% as having no retinopathy. Detection of technical failures or any retinopathy was achieved by manual grading with 86.5% sensitivity (95% confidence interval 85.1 to 87.8) and 95.3% specificity (94.6 to 95.9) and by automated grading with 90.5% sensitivity (89.3 to 91.6) and 67.4% specificity (66.0 to 68.8). Manual and automated grading detected 99.1% and 97.9%, respectively, of patients with referable or observable retinopathy/maculopathy. Manual and automated grading detected 95.7% and 99.8%, respectively, of technical failures.
Automated "disease/no disease" grading of diabetic retinopathy could safely reduce the burden of grading in diabetic retinopathy screening programmes.

Download full-text


Available from: Gordon J Prescott, Jul 22, 2015
  • Source
    • "Regarding decision making, automatic DR screening systems either partially follow clinical protocols (e.g. MAs indicate presence of DR) (Jelinek et al., 2006) (Antal and Hajdu, 2012a) (Philip et al., 2007) (Fleming et al., 2010a) or use a machine learning classifier (Abramoff et al., 2008) (Fleming et al., 2010b) (Agurto et al., 2011). A common way to improve reliability in machine learning based applications is to use ensemble-based approaches (Kuncheva, 2004). "
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, an ensemble-based method for the screening of diabetic retinopathy (DR) is proposed. This approach is based on features extracted from the output of several retinal image processing algorithms, such as image-level (quality assessment, pre-screening, AM/FM), lesion-specific (microaneurysms, exudates) and anatomical (macula, optic disc) components. The actual decision about the presence of the disease is then made by an ensemble of machine learning classifiers. We have tested our approach on the publicly available Messidor database, where 90% sensitivity, 91% specificity and 90% accuracy and 0.989 AUC are achieved in a disease/no-disease setting. These results are highly competitive in this field and suggest that retinal image processing is a valid approach for automatic DR screening.
    Knowledge-Based Systems 04/2014; 60. DOI:10.1016/j.knosys.2013.12.023 · 3.06 Impact Factor
  • Source
    • "So for a fair competition, automated screening systems should be compared by human experts routinely involved. In this sense, manual grading is imperfect, since graders required to be highly specific missed more than 5% of the cases of referable diabetic retinopathy in a study [7]. With an automated decision support of the grading wider access could be assured to the service and improvements could be realized in personal and community DR care. "
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper we give a brief review on the present status of automated detection systems describe for the screening of diabetic retinopathy. We further detail an enhanced detection procedure that consists of two steps. First, a pre-screening algorithm is considered to classify the input digital fundus images based on the severity of abnormalities. If an image is found to be seriously abnormal, it will not be analysed further with robust lesion detector algorithms. As a further improvement, we introduce a novel feature extraction approach based on clinical observations. The second step of the proposed method detects regions of interest with possible lesions on the images that previously passed the pre-screening step. These regions will serve as input to the specific lesion detectors for detailed analysis. This procedure can increase the computational performance of a screening system. Experimental results show that both two steps of the proposed approach are capable to efficiently exclude a large amount of data from further processing, thus, to decrease the computational burden of the automatic screening system.
    Journal of Computational Science 09/2012; 3(5). DOI:10.1016/j.jocs.2012.01.001 · 1.57 Impact Factor
  • Source
    • "To further enhance rural and remote area screening, automated image analysis programs have been developed and are now in use as a first line screening for microaneurysms in Scotland [6]. Several algorithms have been proposed for detecting parts of the retina, the presence/absence of retinopathy as well as specific lesions from mild nonproliferative to proliferative retinopathy and maculopathy (see [7] and references therein). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Sufficient image quality is a necessary prerequisite for reliable automatic detection systems in several healthcare environments. Specifically for Diabetic Retinopathy (DR) detec-tion, poor quality fundus makes more difficult the analysis of discontinuities that characterize lesions, as well as to generate evidence that can incorrectly diagnose the presence of anomalies. Several methods have been applied for classification of image quality and recently, have shown satisfactory results. However, most of the authors have focused only on the visibility of blood vessels through detection of blurring. Furthermore, these studies frequently only used fundus images from specific cameras which are not validated on datasets obtained from different retinographers. In this paper, we propose an approach to verify essential requirements of retinal image quality for DR screening: field definition and blur detection. The methods were developed and validated on two large, representative datasets collected by different cameras. The first dataset comprises 5,776 images and the second, 920 images. For field definition, the method yields a performance close to optimal with an area under the Receiver Operating Characteristic curve (ROC) of 96.0%. For blur detection, the method achieves an area under the ROC curve of 95.5%.
    Sibgrapi 2013; 01/2012
Show more