The effectiveness of artificial intelligence-based automated grading and training system in education of manual detection of diabetic retinopathy.

Background: The purpose of this study is to develop an artificial intelligence (AI)-based automated diabetic retinopathy (DR) grading and training system from a real-world diabetic dataset of China, and in particular, to investigate its effectiveness as a learning tool of DR manual grading for medical students. Methods: We developed an automated DR grading and training system equipped with an AI-driven diagnosis algorithm to highlight highly prognostic related regions in the input image. Less experienced prospective physicians received pre- and post-training tests by the AI diagnosis platform. Then, changes in the diagnostic accuracy of the participants were evaluated. Results: We randomly selected 8,063 cases diagnosed with DR and 7,925 with non-DR fundus images from type 2 diabetes patients. The automated DR grading system we developed achieved accuracy, sensitivity/specificity, and AUC values of 0.965, 0.965/0.966, and 0.980 for moderate or worse DR (95 percent CI: 0.976-0.984). When the graders received assistance from the output of the AI system, the metrics were enhanced in varying degrees. The automated DR grading system helped to improve the accuracy of human graders, i.e., junior residents and medical students, from 0.947 and 0.915 to 0.978 and 0.954, respectively. Conclusion: The AI-based systemdemonstrated high diagnostic accuracy for the detection of DR on fundus images from real-world diabetics, and could be utilized as a training aid system for trainees lacking formal instruction on DR management.

Hyperintense Vessel Sign on T2-FLAIR on Patients with Carotid Endarterectomy / 中国医学影像学杂志

Purpose To evaluate the correlation between T2-FLAIR hyperintense vessel sign (HVS) and the stenotic degree of internal carotid artery (ICA) and assess the HVS changes after the carotid endarterectomy (CEA). Materials and Methods Fifty-one patients with CEA were retrospectively enrolled. The stenosis of the bilateral ICA were as:≥90%, and<90%. The distribution of HVS locations was classified as three regions:sylvian fissure, sulci of temporo-occipital lobe and other areas. The presence and the location of HVS were counted. The extrension of HVS on T2-FLAIR were graded as:I:the presence of HVS was<1/3 of the MCA territory, II:the presence of HVS was≥1/3 of the MCA territory.χ2-test was performed for correlation between HVS and ICA stenosis. The difference of HVS and stenosis of ICA and their effects on CEA was accessed. Results HVS was significantly higher in the ICA stenosis more than 90%group than in the less than 90% group (χ2=23.584, P<0.001). The frequencies of HVS were 12, 34 and 15 in sylvian fissure, sulci of temporo-occipital lobe and other area, respectively. The proportion of grade II HVS was higher in the ≥ 90% group than in the<90% group (χ2=8.395, P<0.05). After CEA, HVS on 29 affected hemispheres were showed to be disappeared (n=24) or remained (n=5) in the treated side. Conclusion The presence and the grade of HVS were correlated with the stenotic degree of ICA. In the patients with ICA stenosis, HVS was most frequently found in the sulci of temporal lobe and occipital lobe, and seldom found in sylvian fissure. HVS disappeared after CEA indicating that HVS can be considered as a marker for CEA treatment.