Neural classifier construction using regularization, pruning and test error estimation

In this paper we propose a method for construction of feed-forward neural classifiers based on regularization and adaptive architectures. Using a penalized maximum likelihood scheme, we derive a modified form of the entropic error measure and an algebraic estimate of the test error. In conjunction with optimal brain damage pruning, a test error estimate is used to select the network architecture. The scheme is evaluated on four classification problems.     

Global bifurcation structure of chaotic neural networks and its application to traveling salesman problems

This paper studies global bifurcation structure of the chaotic neural networks applied to solve the traveling salesman problem (TSP). The bifurcation analysis clarifies the dynamical basis of the chaotic neuro-dynamics which itinerates a variety of network states associated with possible solutions of TSP and efficiently ‘searches’ for the optimum or near-optimum solutions. By following the detailed merging process of chaotic attractors via crises, we find that the crisis-induced intermittent switches among the ruins of the previous localized chaotic attractors underly the ‘chaotic search’ for TSP solutions. On the basis of the present study, efficiency of the ‘chaotic search’ to optimization problems is discussed and a guideline is provided for tuning the bifurcation parameter value which gives rise to efficient ‘chaotic search’.     

The anterior inferior cerebellar arterial network supplying the rat cochlea and its role in autoregulation of cochlear blood flow

Summary The sensitivity of the cochlea is dependent upon maintenance of a delicate homeostatic environment. One mechanism which participates in providing this environment is the autoregulation of cochlear blood flow. This autoregulation is ensured through the interaction of sympathetic, peptidergic and hemodynamic mechanisms. The current study demonstrates an adaptation that also participates in cochlear blood flow autoregulation. Specifically, an anterior inferior cerebellar arterial network is described and the relative contributions of each of its vessels to total cochlear blood flow is measured using laser Doppler flowmetry. The results show that each collateral vessel contributes to the blood supply of the cochlea and that reperfusion is accompanied by hyperemia. These findings suggest an adaptation that provides stable blood flow through redundancy and compensatory potential. Additionally, these observations have implications for experimental models of ischemia.     

真皮下血管网皮瓣修复感染创面的研究

目的：探讨真皮下血管网皮瓣的抗菌能力和临床修复感染创面的应用。方法：以家猪为实验动物，在臀部两侧分别形成传统的岛状皮瓣和真皮下血管网岛状皮瓣，用细菌计数、吖啶橙荧光染色测定白细胞吞噬指数和白细胞内杀菌率等方法，进行对比。结果：真皮下血管网岛状皮瓣的抗菌力明显降低。 １９８９年起临床应用 ５例，创面彻底清创后用吻合血管的真皮下血管网游离皮瓣修复，全部成活。结论：该皮瓣抗菌力虽低于传统皮瓣，但仍可用于修复感染创面。     

用神经网络法预测药物在体透过人皮肤的渗透性

目的：预测药物在体透过人皮肤的渗透性。方法：以正辛醇／水分配系数(logP)、分子体积(V)、氢键酸度(∑β2^H)和氢键碱度(∑β2^H)等理化参数作为输入层神经元，以药物在一定时间内在体透过人皮肤的透过比的对数值(R，透过量／未透过量)作为输出层神经元，建立起合适的BP(Back—propagation)神经网络。结果：17个药物在一定时间内在体透过人皮肤的透过比的神经网络计算值和实测值均相当符合。结论：用BP神经网络法可以较好地预测药物在体透过人皮肤的渗透性。     

应用计算机网络系统提高门急诊管理水平

作者介绍的门急诊计算机管理网络系统，应用软件使用ＦＯＸＰＲＯ２．５ＦＯＲＤＯＳ语言，在两台网络服务器之间应用镜像技术，门诊号的录入使用光笔和条形码技术，实行划价、收费一体化，采用星型拓扑网络结构，保证了网络的先进性、安全性、稳定性和可扩充性。网络系统实现挂号、划价收费、调剂、药品使用管理、工作量统计及经济核算的全程闭环管理格局，有效地堵塞了管理和经济上的漏洞，提高了工作效率和管理水平。该系统已正常运行了１２个月，并通过省级科技成果鉴定。     

Heterogeneity of Red Blood Cell Velocity in Skeletal Muscle Decreases with Increased Flow

Objective: Effective material exchange between blood and tissue depends on the heterogeneity of microvascular flow. The objective was to address inconsistencies between intravital studies regarding this dependency. We tested the hypothesis that heterogeneity of red blood cell velocity (V_RBC) in capillary beds varies with the strength of metabolic stimulus and with capillary bed geometry. Methods: We used videomicroscopy to measure V_RBC in a bed of 10–24 capillaries at the surface of extensor digitorum longus (EDL) muscle in anesthetized rats. The coefficient of variation (CV = standard deviation/mean; an index of spatial heterogeneity) was computed in the same bed before and after (i) 1, 2, 4, or 8 Hz supramaximal muscle contraction or (ii) adenosine superfusion (10^?7–10^?3 M). Beds with or without arteriolar—venular capillary shunts were used. Results: Although control V_RBC differed between beds (shunt: 232 μm/s; no shunt: 130 μm/s), the percentage increases in postcontraction V_RBC did not (range: 111–326%). In both beds, control CV varied greatly (overall range: 28–117%) and 2–8 Hz muscle contractions reduced CV significantly by 25%. Similar results were obtained for adenosine. In confirmatory experiments using the rat cremaster muscle, contractions (4 Hz) and adenosine (10^?4 M) also reduced CV. Based on all data, CV = 63–0.022 V_RBC (r = 0.82, P < 0.001). Conclusions: The heterogeneity of V_RBC decreased with metabolic stress, regardless of capillary bed geometry. We propose that both the large variability in control CV and the relatively shallow dependence of CV on velocity could be responsible for the present inconsistencies between intravital studies.     

A neural network confirms that physical exercise reverses EEG changes in depressed rats

The use of an artificial neural network (ANN) system to differentiate the EEG power density spectra in depressed from normal rats was tried. The beneficial effects of chronic physical exercise in reducing the effects of stress and therefore depression was also to be tested in animals by the same method. In this study, rats were divided into 4 groups, subjected to (i) chronic stress (D group); (ii) chronic exercise by treadmill running (EO group); (iii) exercise with stress (ES group) and (iv) handling (C group). The prefrontal cortical EEG, EMG and EOG were recorded simultaneously on paper and the digitized EEG signals were also stored in the hard-disk of a PC-AT through an ADC. After filtering the digitize signals, the EEG power spectra were calculated by an FFT routine. Three successive 4 s artefact-free epochs were averaged. The REM and NREM sleep periods as well as the awake period signals were analyzed separately. The FFT values from each of the 3 states, in the 4 groups of animals were tested by an ANN with 30 first layer neurons and a 2nd layer of a majority-vote-taker. The ANN could distinguish the depressed from the normal rats' EEG very well in REM (99%) sleep, NREM (95%) sleep and awake (81%) states. In most of the cases it identified the exercised rats' EEG as normal.     

基于贝叶斯多变量联合模型的体检人群脑卒中发病风险因素的纵向研究

背景 脑卒中是目前影响人类健康的主要公共卫生问题之一;健康体检纵向数据累积了大量的健康信息,由于缺失数据多、样本量小等诸多问题,导致其利用率低、重要信息未能得到充分挖掘,进而对常见慢性病的有效防控等工作带来一定困难。目的 基于贝叶斯多变量联合模型,探讨体检人群脑卒中发病风险因素,为慢性病风险因素分析提供新的方法。方法 本研究使用空军军医大学西京医院健康医学中心2008—2015年的体检资料。随访情况：以首次发生脑卒中为结局事件,发生结局事件立即停止随访;若未发生,到2015年体检信息收集完成后结束随访;体检间隔时间为1年。依据随访过程中是否发生脑卒中分为脑卒中组和非脑卒中组。纵向观察变量包括总胆固醇（TC）、三酰甘油（TG）、低密度脂蛋白胆固醇（LDL-C）、高密度脂蛋白胆固醇（HDL-C）、体质指数（BMI）和收缩压（SBP）。采用多因素Cox回归模型分析基线情况对脑卒中结局事件的影响;采用贝叶斯多变量联合模型,分析随访过程中TC、TG、LDL-C、HDL-C、BMI和SBP的纵向变化轨迹对脑卒中发病的影响。结果 本研究共纳入234例研究对象,1 581条纵向随访记录,平均随访时间为...     

兔耳纯静脉皮瓣血流量和血管网面密度观测

目的:观测纯静脉皮瓣术后血流量和血管网面密度变化。方法:利用家兔耳背设计纯静脉皮瓣和动脉皮瓣模型①观测术后血流量;②墨汁灌注后软蜡厚蜡切片,图像分析血管网面密度。结果:①纯静脉皮瓣术后48h血流量最低,第14d恢复至术前65%。②纯静脉皮瓣术后第1～7d的血管网面密度高于动脉皮瓣(P<0.05)。结论:纯静脉皮瓣术后血流量及血管网密度变化与动脉皮瓣不同。