Neural architecture for robotic vision sensor that discounts uneven illumination in a manufacturing environment

Computer vision algorithms for inspection or pick-and-place operations often depend on spatially uniform illumination of a workplace. This necessitates expensive lighting fixtures. To discount effects of uneven illumination we designed and tested a neural network that can adaptively control light sensitivity at the photosensor level. Our neural network architecture consists of multiple layers with hexagonally arranged nodes. All nodes have partially overlapping receptive fields of different spatial frequencies. Feedforward connections are excitatory while feedback pathways subserve lateral inhibition. The outputs of these nodes are combined so as to maximize the signal-to-noise ratio while constantly resetting thresholds to maintain high sensitivity. Our connectionist architecture can account for many characteristics attributed to the lightness constancy phenomenon observed in biological systems. The results suggest that our module maintains high sensitivity over the whole domain of intensities without interfering with transmission of visual information embedded in spatial discontinuities of intensity.     

Differential Action of Connexin Hemichannel and Pannexin Channel Therapeutics for Potential Treatment of Retinal Diseases

Dysregulation of retinal function in the early stages of light-induced retinal degeneration involves pannexins and connexins. These two types of proteins may contribute to channels that release ATP, leading to activation of the inflammasome pathway, spread of inflammation and retinal dysfunction. However, the effect of pannexin channel block alone or block of both pannexin channels and connexin hemichannels in parallel on retinal activity in vivo is unknown. In this study, the pannexin channel blocker probenecid and the connexin hemichannel blocker tonabersat were used in the light-damaged rat retina. Retinal function was evaluated using electroretinography (ERG), retinal structure was analyzed using optical coherence tomography (OCT) imaging and the tissue response to light-induced injury was assessed immunohistochemically with antibodies against glial fibrillary acidic protein (GFAP), Ionized calcium binding adaptor molecule 1 (Iba-1) and Connexin43 (Cx43). Probenecid did not further enhance the therapeutic effect of connexin hemichannel block in this model, but on its own improved activity of certain inner retina neurons. The therapeutic benefit of blocking connexin hemichannels was further evaluated by comparing these data against results from our previously published studies that also used the light-damaged rat retina model. The analysis showed that treatment with tonabersat alone was better than probenecid alone at restoring retinal function in the light-damaged retina model. The results assist in the interpretation of the differential action of connexin hemichannel and pannexin channel therapeutics for potential treatment of retinal diseases.     

长春新碱治疗糖尿病肾病大鼠的实验研究

目的验证长春新碱对糖尿病肾病大鼠的治疗作用。方法用链脲佐菌素(STZ)诱导大鼠糖尿病模型。将模型大鼠分为治疗组与非治疗组,治疗组给予长春新碱(VCR)0.2 mg/kg,尾静脉注射,每周2次,观察给药后大鼠血糖、尿蛋白、肾功能、肾重,体重等值的改变以及肾脏病理的改善情况。结果 治疗组大鼠尿蛋白明显减少,肾重/体重低于非治疗组。光镜下观察治疗组肾脏病理学变化情况有明显改善。结论长春新碱可降低糖尿病肾病大鼠的尿蛋白并改善糖尿病肾病早期病理损害。     

一种改进的视网膜图像中微小动脉瘤的检测算法

基于糖尿病性视网膜病变中最早出现的微小动脉瘤病症进行了研究,提出一种有效的微小动脉瘤检测算法。首先在传统模板匹配算法的基础上提出了一种动态多参数模板匹配算法,并且使用相对误差和与相关系数来共同制约匹配度,从而实现了更为精确的匹配提取;其次提出了基于分布特性的计分策略和自适应加权的汇总策略,避免了单纯采用各个特征量作为独立约束指标进行筛选时忽视各个特征量的约束力大小的弊端。实验结果表明,该检测算法能够有效地提高微小动脉瘤的检测真阳性率。     

基于视觉感知机制的织物疵点轮廓检测

针对传统人工织物疵点检测存在的误检及低效等问题,提出了一种基于视觉感知机制的自适应织物疵点轮廓检测方法.首先,模拟视觉系统中视网膜感受野对视觉信息的处理机制对织物疵点图像进行滤波及疵点增强;其次,依据初级视皮层(V1)区对视觉信息响应的方向选择性机制构建织物疵点图像边缘检测模型,实现对织物疵点图像的边缘检测.最后,采用自适应阈值选择的方法对检测到的边缘进行二次处理,获得织物图像疵点的轮廓.为验证本文方法的有效性和准确性,对4类织物疵点图像进行测试,并定性和定量两方面进行比较分析,结果表明文中提出的方法能够较好地检测出织物疵点轮廓信息,不仅可以得到质量较高的织物疵点轮廓图像,而且在整个检测过程中能够自适应的选择参数,避免受人的主观因素影响,具有实际的应用价值.     

基于自适应阈值和区域生长的SD-OCT糖网图像亮斑分割

硬性渗出是糖尿病性视网膜病变的一个比较显著的症状,频域光学相干断层视网膜图像中的高信号亮斑与渗出有着密切的联系。为了研究渗出与病变的关系,有必要找到一种提取亮斑的方法。但是,目前关于糖网图像亮斑提取的研究还非常少。首先运用层分割算法限制亮斑所在区域,然后采用自适应阈值法确定种子集合,最后使用基于人类视觉特性的区域生长方法提取出亮斑。实验结果表明,本方法可以较准确地分割出糖网图像中的亮斑。     

An effective deep learning model for grading abnormalities in retinal fundus images using variational auto-encoders

Diabetic retinopathy (DR) and Diabetic Macular Edema (DME) are severe diseases that affect the eyes due to damage in blood vessels. Computer-aided automated grading will help clinicians conduct disease diagnoses at ease. Experiments of automated image processing with deep learning techniques using CNN produce promising results, especially in the medical imaging domain. However, the disease grading tasks in retinal images using CNN struggle to retain high-quality information at the output. A novel deep learning model based on variational auto-encoder to grade DR and DME abnormalities in retinal images is proposed. The objective of the proposed model is to extract the most relevant retinal image features efficiently. It focuses on addressing less relevant candidate region generation and translational invariance present in images. The experiments are conducted in IDRID dataset and evaluated using accuracy, U-kappa, sensitivity, specificity and precision metrics. The results outperform compared with other state-of-art techniques.     

Development of an image-based model for capillary vasculature of retina

The paper presents a method of development of a detailed network model to represent retinal capillary vasculature. The capillary model is a circular mesh consisting of concentric rings with an increasing diameter. Each of the rings has uniformly distributed bifurcation nodes to represent capillary vessels. The model is customized using the data that has been measured from confocal microscopic images of a mouse retina. The capillary model developed can be connected to networks of larger vessels of the vasculature such as arterial and venous networks to form a complete model of the retinal network. A method to automate such interface connections between capillary and other vascular networks using connecting vessels (i.e., pre-capillary and post-capillary) is also presented in the paper. Such a detailed image-based capillary model together with the arterial and venular networks can be used for various circulation simulations to obtain accurate information on hemodynamic quantities such as the spatial distribution of pressure and flow in the vasculature for both physiological and pathological conditions. The method presented for the development of the capillary model can also be adopted for vasculatures of other organs.     

The discovery of biomarkers for type 2 diabetic nephropathy by serum proteome analysis

Diabetic nephropathy (DN) is a serious kidney complication of diabetes, and constitutes the leading cause of end-stage renal disease. The earliest clinical evidence of DN is microalbuminuria, a term which refers to the appearance of small but abnormal amounts of albumin in the urine. However, screening methods for DN, such as biomarker assays, are yet to be developed for type 2 DN. In the present study, in an attempt to identify the biomarkers for initial diagnoses of type 2 DN, the protein profiles of human sera collected from 30 microalbuminuric type 2 diabetic patients were compared with those collected from 30 normoalbuminuric type 2 diabetic patients, via 2-DE. As a result, a total of 18 spots were determined to have different protein levels in the microalbuminuric patients. Twelve spots had lower protein levels of approximately 50%, and the other six had higher levels of approximately 100-300% as compared to the spots of normoalbuminuric patients. These spots were identified with ESI-Q-TOF (ESI-quadrupole-TOF) MS. Among the identified proteins, vitamin D-binding protein (DBP) and pigment epithelium-derived factor (PEDF) were verified by Western blotting. The results of this study indicate that the DBP may be employed as diagnostic and monitoring biomarkers of type 2 DN, contingent on further study into the matter.