Automated detection and classification of early AMD biomarkers using deep learning.

Age-related macular degeneration (AMD) affects millions of people and is a leading cause of blindness throughout the world. Ideally, affected individuals would be identified at an early stage before late sequelae such as outer retinal atrophy or exudative neovascular membranes develop, which could produce irreversible visual loss. Early identification could allow patients to be staged and appropriate monitoring intervals to be established. Accurate staging of earlier AMD stages could also facilitate the development of new preventative therapeutics. However, accurate and precise staging of AMD, particularly using newer optical coherence tomography (OCT)-based biomarkers may be time-intensive and requires expert training which may not feasible in many circumstances, particularly in screening settings. In this work we develop deep learning method for automated detection and classification of early AMD OCT biomarker. Deep convolution neural networks (CNN) were explicitly trained for performing automated detection and classification of hyperreflective foci, hyporeflective foci within the drusen, and subretinal drusenoid deposits from OCT B-scans. Numerous experiments were conducted to evaluate the performance of several state-of-the-art CNNs and different transfer learning protocols on an image dataset containing approximately 20000 OCT B-scans from 153 patients. An overall accuracy of 87% for identifying the presence of early AMD biomarkers was achieved.

Reliability of Graders and Comparison with an Automated Algorithm for Vertical Cup-Disc Ratio Grading in Fundus Photographs.

INTRODUCTION: We aimed to investigate the intergrader and intragrader reliability of human graders and an automated algorithm for vertical cup-disc ratio (CDR) grading in colour fundus photographs. MATERIALS AND METHODS: Two-hundred fundus photographs were selected from a database of 3000 photographs of patients screened at a tertiary ophthalmology referral centre. The graders included glaucoma specialists (n = 3), general ophthalmologists (n = 2), optometrists (n = 2), family physicians (n = 2) and a novel automated algorithm (AA). In total, 2 rounds of CDR grading were held for each grader on 2 different dates, with the photographs presented in random order. The CDR values were graded as 0.1-1.0 or ungradable. The grading results of the 2 senior glaucoma specialists were used as the reference benchmarks for comparison. RESULTS: The intraclass correlation coefficient values ranged from 0.37-0.74 and 0.47-0.97 for intergrader and intragrader reliability, respectively. There was no significant correlation between the human graders' level of reliability and their years of experience in grading CDR (P = 0.91). The area under the curve (AUC) value of the AA was 0.847 (comparable to AUC value of 0.876 for the glaucoma specialist). Bland Altman plots demonstrated that the AA's performance was at least comparable to a glaucoma specialist. CONCLUSION: The results suggest that AA is comparable to and may have more consistent performance than human graders in CDR grading of fundus photographs. This may have potential application as a screening tool to help detect asymptomatic glaucoma-suspect patients in the community.

Retinal hemorrhage detection by rule-based and machine learning approach.

Robust detection of hemorrhages (HMs) in color fundus image is important in an automatic diabetic retinopathy grading system. Detection of the hemorrhages that are close to or connected with retinal blood vessels was found to be challenge. However, most methods didn't put research on it, even some of them mentioned this issue. In this paper, we proposed a novel hemorrhage detection method based on rule-based and machine learning methods. We focused on the improvement of detection of the hemorrhages that are close to or connected with retinal blood vessels, besides detecting the independent hemorrhage regions. A preliminary test for detecting HM presence was conducted on the images from two databases. We achieved sensitivity and specificity of 93.3% and 88% as well as 91.9% and 85.6% on the two datasets.

A pilot study of a mobile-phone-based home monitoring system to assist in remote interventions in cases of acute exacerbation of COPD.

We conducted a six-month feasibility study of a mobile-phone-based home monitoring system, called M-COPD. Patients with a history of moderate Acute Exacerbation of COPD (AECOPD) were given a mobile phone to record major symptoms (dyspnoea, sputum colour and volume), minor symptoms (cough and wheezing) and vital signs. A care team remotely monitored the recorded data and provided clinical interventions. Eight patients (mean age 65 years) completed the trial. Ten acute exacerbations occurred during the trial and were successfully treated at home. Prior to the AECOPD episode, the combined score of the major symptoms increased significantly (P < 0.05). Following the intervention, it decreased significantly (P < 0.05) within two weeks and returned to the baseline. The score of the minor symptoms also increased significantly (P < 0.05), but the decrease following the intervention was not significant. There were significantly fewer hospital admissions during the trial, fewer ED presentations and fewer GP visits than in a six-month matched period in the preceding year. The results demonstrate the potential of home monitoring for analysing respiratory symptoms for early intervention of AECOPD.

Drusen quantification for early identification of age related macular degeneration (AMD) using color fundus imaging.

Age-related macular degeneration (AMD) is a major cause of visual impairment in the elderly and identifying people with the early stages of AMD is important when considering the design and implementation of preventative strategies for late AMD. Quantification of drusen size and total area covered by drusen is an important risk factor for progression. In this paper, we propose a method to detect drusen and quantify drusen size along with the area covered with drusen in macular region from standard color retinal images. We used combined local intensity distribution, adaptive intensity thresholding and edge information to detect potential drusen areas. The proposed method detected the presence of any drusen with 100% accuracy (50/50 images). For drusen detection accuracy (DDA), the segmentations produced by the automated method on individual images achieved mean sensitivity and specificity values of 74.94% and 81.17%, respectively.