Journal of medical Internet research

1438-8871 16 10 2014 Journal of medical Internet research J. Med. Internet Res. Rapid Grading of Fundus Photographs for Diabetic Retinopathy Using Crowdsourcing. e233 Screening for diabetic retinopathy is both effective and cost-effective, but rates of screening compliance remain suboptimal. As screening improves, new methods to deal with screening data may help reduce the human resource needs. Crowdsourcing has been used in many contexts to harness distributed human intelligence for the completion of small tasks including image categorization. Our goal was to develop and validate a novel method for fundus photograph grading. An interface for fundus photo classification was developed for the Amazon Mechanical Turk crowdsourcing platform. We posted 19 expert-graded images for grading by Turkers, with 10 repetitions per photo for an initial proof-of-concept (Phase I). Turkers were paid US $0.10 per image. In Phase II, one prototypical image from each of the four grading categories received 500 unique Turker interpretations. Fifty draws of 1-50 Turkers were then used to estimate the variance in accuracy derived from randomly drawn samples of increasing crowd size to determine the minimum number of Turkers needed to produce valid results. In Phase III, the interface was modified to attempt to improve Turker grading. Across 230 grading instances in the normal versus abnormal arm of Phase I, 187 images (81.3%) were correctly classified by Turkers. Average time to grade each image was 25 seconds, including time to review training images. With the addition of grading categories, time to grade each image increased and percentage of images graded correctly decreased. In Phase II, area under the curve (AUC) of the receiver-operator characteristic (ROC) indicated that sensitivity and specificity were maximized after 7 graders for ratings of normal versus abnormal (AUC=0.98) but was significantly reduced (AUC=0.63) when Turkers were asked to specify the level of severity. With improvements to the interface in Phase III, correctly classified images by the mean Turker grade in four-category grading increased to a maximum of 52.6% (10/19 images) from 26.3% (5/19 images). Throughout all trials, 100% sensitivity for normal versus abnormal was maintained. With minimal training, the Amazon Mechanical Turk workforce can rapidly and correctly categorize fundus photos of diabetic patients as normal or abnormal, though further refinement of the methodology is needed to improve Turker ratings of the degree of retinopathy. Images were interpreted for a total cost of US $1.10 per eye. Crowdsourcing may offer a novel and inexpensive means to reduce the skilled grader burden and increase screening for diabetic retinopathy. Brady Christopher J CJ http://orcid.org/0000-0001-7847-3914 Wills Eye Hospital, Retina Service: Mid Atlantic Retina, Philadelphia, PA, United States. brady@jhmi.edu. Villanti Andrea C AC http://orcid.org/0000-0003-3104-966X Pearson Jennifer L JL http://orcid.org/0000-0002-1400-5932 Kirchner Thomas R TR http://orcid.org/0000-0001-5764-4980 Gupta Omesh P OP http://orcid.org/0000-0003-4845-0409 Shah Chirag P CP http://orcid.org/0000-0001-6369-4917 ENG JOURNAL ARTICLE 2014 10 30

2041-4889 5 2014 Cell death & disease Cell Death Dis Pathogenic role of lncRNA-MALAT1 in endothelial cell dysfunction in diabetes mellitus. e1506 10.1038/cddis.2014.466 Long noncoding RNAs (lncRNAs) have important roles in diverse biological processes. Our previous study has revealed that lncRNA-MALAT1 deregulation is implicated in the pathogenesis of diabetes-related microvascular disease, diabetic retinopathy (DR). However, the role of MALAT1 in retinal vasculature remodeling still remains elusive. Here we show that MALAT1 expression is significantly upregulated in the retinas of STZ-induced diabetic rats and db/db mice. MALAT1 knockdown could obviously ameliorate DR in vivo, as shown by pericyte loss, capillary degeneration, microvascular leakage, and retinal inflammation. Moreover, MALAT1 knockdown could regulate retinal endothelial cell proliferation, migration, and tube formation in vitro. The crosstalk between MALAT1 and p38 MAPK signaling pathway is involved in the regulation of endothelial cell function. MALAT1 upregulation represents a critical pathogenic mechanism for diabetes-induced microvascular dysfunction. Inhibition of MALAT1 may serve as a potential target for anti-angiogenic therapy for diabetes-related microvascular complications. Liu J-Y JY 1] Eye Hospital, Nanjing Medical University, Nanjing, China [2] The Fourth School of Clinical Medicine, Nanjing Medical University, Nanjing, China. Yao J J 1] Eye Hospital, Nanjing Medical University, Nanjing, China [2] The Fourth School of Clinical Medicine, Nanjing Medical University, Nanjing, China. Li X-M XM Eye Hospital, Nanjing Medical University, Nanjing, China. Song Y-C YC 1] Eye Hospital, Nanjing Medical University, Nanjing, China [2] The Fourth School of Clinical Medicine, Nanjing Medical University, Nanjing, China. Wang X-Q XQ 1] Eye Hospital, Nanjing Medical University, Nanjing, China [2] The Fourth School of Clinical Medicine, Nanjing Medical University, Nanjing, China. Li Y-J YJ 1] Eye Hospital, Nanjing Medical University, Nanjing, China [2] The Fourth School of Clinical Medicine, Nanjing Medical University, Nanjing, China. Yan B B 1] Eye Hospital, Nanjing Medical University, Nanjing, China [2] The Fourth School of Clinical Medicine, Nanjing Medical University, Nanjing, China. Jiang Q Q 1] Eye Hospital, Nanjing Medical University, Nanjing, China [2] The Fourth School of Clinical Medicine, Nanjing Medical University, Nanjing, China [3] Institute of Integrated Medicine, Nanjing Medical University, Nanjing, China. ENG JOURNAL ARTICLE 2014 10 30