Vortex vein engorgement and different shapes of venous drainage systems in polypoid choroidal vasculopathy vs. age­related macular degeneration on indocyanine green angiography.

There are differences in vortex vein engorgement and appearance in polypoid choroidal vasculopathy (PCV), age-related macular degeneration (AMD), and healthy eyes. The present study aimed to use indocyanine green angiography (ICGA) to find a simple, clinically meaningful method for evaluating the filling degree of vortex veins in various eye diseases. Participant clinical characteristics were recorded. The number of vortex veins (NVV), central vortex vein diameter (CVVD), mean root area of the vortex vein (MRAVV), mean diameter of the thickest peripheral branch (MDPTB), subfoveal choroidal thickness and percentage of vortex vein anastomosis (PVVA) were obtained by marking the vortex veins on ICGA. The proportion of subretinal haemorrhage and the numbers and types of vortex veins in each quadrant were counted separately. The CVVD and MDPTB were significantly increased in the PCV compared with those in the AMD group (P<0.05). The CVVD, MRAV, and PVVA were significantly increased in the PCV compared with those in the healthy group (P<0.05). The type IV vortex vein (complete with ampulla) proportion was the lowest while the type I (vortex vein absent) proportion was the highest in the PCV group (P<0.001). NVV in the inferior-temporal region was increased in the PCV compared with that in the AMD group (P=0.034). Subretinal haemorrhage occurred in the inferior temporal choroid in 47.62% of examined eyes in PCV group, and in the superior temporal choroid in 23.81% of the PCV group, with significant differences between the quadrants (P<0.001). Vortex vein engorgement and shape differed significantly between PCV, AMD and healthy eyes. The vortex vein branches in PCV eyes were significantly dilated in the posterior pole; moreover, the peripheral choroid and the lower proportion of type IV vortex veins may be pathognomonic for PCV.

A New Method for Automatic Detection of Defects in Selective Laser Melting Based on Machine Vision.

Selective laser melting (SLM) is a forming technology in the field of metal additive manufacturing. In order to improve the quality of formed parts, it is necessary to monitor the selective laser melting forming process. At present, most of the research on the monitoring of the selective laser melting forming process focuses on the monitoring of the melting pool, but the quality of forming parts cannot be controlled in real-time. As an indispensable link in the SLM forming process, the quality of powder spreading directly affects the quality of the formed parts. Therefore, this paper proposes a detection method for SLM powder spreading defects, mainly using industrial cameras to collect SLM powder spreading surfaces, designing corresponding image processing algorithms to extract three common powder spreading defects, and establishing appropriate classifiers to distinguish different types of powder spreading defects. It is determined that the multilayer perceptron (MLP) is the most accurate classifier. This detection method has high recognition rate and fast detection speed, which cannot only meet the SLM forming efficiency, but also improve the quality of the formed parts through feedback control.