Use in clinical practice of an automated screening method of diabetic retinopathy that can be derived using a diagnostic artificial intelligence system

Author(s):  
Cristina Peris-Martínez ◽  
Abhay Shaha ◽  
Warren Clarida ◽  
Ryan Amelon ◽  
María C Hernáez-Ortega ◽  
...  
2020 ◽  
pp. 193229682090621
Author(s):  
Abhay Shah ◽  
Warren Clarida ◽  
Ryan Amelon ◽  
Maria C. Hernaez-Ortega ◽  
Amparo Navea ◽  
...  

Purpose: The purpose of this study is to compare the diagnostic performance of an autonomous artificial intelligence (AI) system for the diagnosis of referable diabetic retinopathy (RDR) to manual grading by Spanish ophthalmologists. Methods: Subjects with type 1 and 2 diabetes participated in a diabetic retinopathy (DR) screening program in 2011 to 2012 in Valencia (Spain), and two images per eye were collected according to their standard protocol. Mydriatic drops were used in all patients. Retinal images—one disc and one fovea centered—were obtained under the Medical Research Ethics Committee approval and de-identified. Exams were graded by the autonomous AI system (IDx-DR, Coralville, Iowa, United States), and manually by masked ophthalmologists using adjudication. The outputs of the AI system and manual adjudicated grading were compared using sensitivity and specificity for diagnosis of both RDR and vision-threatening diabetic retinopathy (VTDR). Results: A total of 2680 subjects were included in the study. According to manual grading, prevalence of RDR was 111/2680 (4.14%) and of VTDR was 69/2680 (2.57%). Against manual grading, the AI system had a 100% (95% confidence interval [CI]: 97%-100%) sensitivity and 81.82% (95% CI: 80%-83%) specificity for RDR, and a 100% (95% CI: 95%-100%) sensitivity and 94.64% (95% CI: 94%-95%) specificity for VTDR. Conclusion: Compared to manual grading by ophthalmologists, the autonomous diagnostic AI system had high sensitivity (100%) and specificity (82%) for diagnosing RDR and macular edema in people with diabetes in a screening program. Because of its immediate, point of care diagnosis, autonomous diagnostic AI has the potential to increase the accessibility of RDR screening in primary care settings.


2021 ◽  
Vol 160 (6) ◽  
pp. S-64-S-65
Author(s):  
Ethan A. Chi ◽  
Gordon Chi ◽  
Cheuk To Tsui ◽  
Yan Jiang ◽  
Karolin Jarr ◽  
...  

Author(s):  
Mohamed Hossameldin khalifa ◽  
Ahmed Samir ◽  
Ayman Ibrahim Baess ◽  
Sara Samy Hendawi

Abstract Background Vascular angiopathy is suggested to be the major cause of silent hypoxia among COVID-19 patients without severe parenchymal involvement. However, pulmonologists and clinicians in intensive care units become confused when they encounter acute respiratory deterioration with neither severe parenchymal lung involvement nor acute pulmonary embolism. Other radiological vascular signs might solve this confusion. This study investigated other indirect vascular angiopathy signs on CT pulmonary angiography (CTPA) and involved a novel statistical analysis that was performed to determine the significance of associations between these signs and the CT opacity score of the pathological lung volume, which is calculated by an artificial intelligence system. Results The study was conducted retrospectively, during September and October 2020, on 73 patients with critical COVID-19 who were admitted to the ICU with progressive dyspnea and low O2 saturation on room air (PaO2 < 93%). They included 53 males and 20 females (73%:27%), and their age ranged from 18 to 88 years (mean ± SD=53.3 ± 13.5). CT-pulmonary angiography was performed for all patients, and an artificial intelligence system was utilized to quantitatively assess the diseased lung volume. The radiological data were analyzed by three expert consultant radiologists to reach consensus. A low CT opacity score (≤10) was found in 18 patients (24.7%), while a high CT opacity score (>10) was found in 55 patients (75.3%). Pulmonary embolism was found in 24 patients (32.9%); three of them had low CT opacity scores. Four other indirect vasculopathy CTPA signs were identified: (1) pulmonary vascular enlargement (57 patients—78.1%), (2) pulmonary hypertension (14 patients—19.2%), (3) vascular tree-in-bud pattern (10 patients—13.7%), and (4) pulmonary infarction (three patients—4.1%). There were no significant associations between these signs and the CT opacity score (0.3205–0.7551, all >0.05). Furthermore, both pulmonary vascular enlargement and the vascular tree-in-bud sign were found in patients without pulmonary embolism and low CT-severity scores (13/15–86.7% and 2/15–13.3%, respectively). Conclusion Pulmonary vascular enlargement or, less commonly, vascular tree-in-bud pattern are both indirect vascular angiopathy signs on CTPA that can explain the respiratory deterioration which complicates COVID-19 in the absence of severe parenchymal involvement or acute pulmonary embolism.


Sign in / Sign up

Export Citation Format

Share Document