Anti-NMS monoclonal antibody Made to order

Purpose Our aim was to highlight the presence and the frequency of posterior staphyloma (PS) in non-highly myopic retinitis pigmentosa (RP) patients and to study the relationship between PS and choroidal thickness (CT). Methods This was a retrospective case-control study of 77 eyes (39 patients) with RP, axial length inferior to 26 mm and clinically preserved macular area. All patients underwent fundus photography, A- and B-scan ocular ultrasonography, fundus autofluorescence (FAF) and swept source optical coherence tomography (SS-OCT). PS was defined by an outward bowing of the sclera on SS-OCT and B-scans. The relationship between the PS and SS-OCT layers thicknesses was determined. Results Over 77 RP eyes of 39 patients studied, a PS was identified in 17 eyes (22%) of nine patients. Fifteen eyes had a narrow macular staphyloma (NMS), and two eyes had a wide macular staphyloma (WMS). Mean age in this group was 34.2 years (range 19-53 years), mean axial length was 23.60 +/- 0.61 mm and mean CT was 185.7 +/- 71 um versus 259.7 um in eyes without PS. The staphyloma edges corresponded to area of outer retina loss on SS-OCT and were larger than the hyperautofluorescent ring on FAF. We found a significant association between PS and CT in our RP patients (p = 0.003). The mean CT was significantly thinner in PS eyes compared to eyes without staphyloma. There was no significant association between PS and with visual acuity, years of progression, retinal thickness nor FAF findings. Conclusions PS was present in 22% of non-highly myopic eyes with RP. Narrow macular staphyloma was the most common type observed in our series. A marked thinning of the choroid was noted in PS eyes when compared to RP eyes without PS, as well as the outer retina degeneration.

In recent years, a large number of quantization schemes have been proposed for compressing convolutional neural networks (CNN). However, most of them have the following problems: 1) when there are outliers in the weight, post-training quantization cannot obtain the ideal effect, and the accuracy loss is unavoidable; 2) quantizing the non-maximum suppression (NMS) stage of oriented object detection networks is non-trivial so that such networks are difficult to deploy on edge computing devices that only support integer operations. In this letter, we propose the outlier-aware quantization (OAQ) to boost the robustness of the post-training quantization method. Besides, we design a multilayer perceptron network to approximate the intersection-over-union (IoU) of rotated boxes, making the NMS stage can be deployed on integer-arithmetic-only devices. The experiment results demonstrate that our solution outperforms the widely used post-training quantizationmethod. Meanwhile, to the best of our knowledge, this is the first study that focuses on the optimization and quantization of the NMS stage of oriented object detection networks.