时变脑诱发电位的单次提取和麻醉控制

介绍了采用多种数字信号处理实现时变（有延时和幅度突然增大）脑诱发电位的单次提取的新的实验结果。这些方法是：采用小波变换法提高淹没在脑自发电位中的脑诱发电位信噪比作为自适应滤波器的原始输入,按相关系数最大值处对齐后作平均作为自适应滤波器的参考输入,自适应滤波后再进行后处理。实验表明,经过这样多重处理后,可以较好地单次提取时变脑诱发电位     

针对脑内源光学信号的自适应中值滤波处理

具有高空间分辨率的脑内源性信号光学成像术是研究大脑皮层功能构筑的有力工具.针对非常微弱的内源性光学信号的提取,本文首先讨论了其噪声来源和基于生理结构原因的信号特征:然后根据由标准化处理得到的功能图像的二阶统计量分析结果,提出了自适应滤波窗口设计:使用非线性中值空域滤波,对含有不同特征的图像区域,分别采取不同的滤波窗口进...     

应用小波变换检测细胞显微荧光图像中的囊泡

文章提出了一种用小波变换来检测生物荧光图像中囊泡的方法。作者用à trous小波对图像进行小波变换,然后求出每层系数的中值绝对偏差σ,并用t=kσ/0.67作为阈值对每层系数进行门限滤波,然后通过提取小波变换系数来重构图像。通过设计实验与常用的“rolling ball”算法对比,发现小波变换算法在低信噪比的情况下,具有更好的灵敏度;对于形状大小不同的信号,具有更好的稳定性;而且对于信号的细节信息具有更好的保真性。     

基于偏微分方程的保持纹理细节的医学图像去噪方法

基于偏微分方程的医学图像去噪方法已公认为具有显著效果的去噪技术。常用的偏微分方程去噪方法虽然可以去除变化平缓的图像中的噪声,同时保持边缘结构信息,但对带有较多纹理细节的医学图像的去噪效果却不太理想。在对目前有关纹理医学图像理解和综合的基础上,介绍了三种纹理图像去噪技术。第一种自适应调整参数的全变分方法,在不同尺度空间下去噪,可保持纹理和细节;第二种将医学图像空间由BV空间上升到G空间以保持纹理;第三种使用多尺度分解噪声以保持细节特征。这些方法均能在实际应用中达到一定的效果,但是如何更好的去除噪声,文章对其进行分析,并提出新的改进思路。     

不同滤波方法对超声图像分割的影响

采用各向异性滤波方法以及Gabor滤波方法对乳腺肿瘤超声图像进行处理,再使用snake方法以及levelset方法在设置相同参数的条件下,对过滤的图像分别进行分割。试验结果表明,不同的滤波方法在分害4目标图像的收敛度,边缘圆滑度及整体轮廓提取效果都有较大影响,为分割图像选取适当的滤波器提供了参考。     

基于计算机视觉的棉铃虫成虫雌雄自动判别研究

棉铃虫Helicoverpa armigera(Hübner)是棉花作物的主要寄主害虫,其性别的自动准确判别对区域性比、种群数量等方面的预测预报具有重要意义。本文通过CCD设备获取雌雄成虫的原始彩色图像,运用数学形态学和自适应图像增强法进行滤波分析,提高害虫RGB彩色图像及分通道图像的质量。针对RGB彩色图像提取害虫的颜色矩特征,针对B通道灰度图像提取基于灰度共生和差分统计矩阵的纹理、形态不变矩等特征,对提取的36个特征数据进行归一化处理。将惩罚因子和RBF核函数参数作为SVM分类器识别率的重要判断标准,利用K折交叉验证选取最优参数组合并建立模型,当C=4,g=0.0825,识别率达到最佳98.33%。证实了将计算机视觉应用于昆虫性别自动判别的可行性。     

诱发电位的多通道时频相干提取算法

多通道时频域相干成分提取算法是针对低信噪比的宽频带信号提取问题提出的。它采用多通道同步观测,在各通道的观测数据中信号成分具有较高的相干性,而噪声的相干性较低,因此根据其相干性的高低差别即可将信号与噪声分离,提取有效信号。为实现信号与噪声的分离,首先应用小波包分解将信号在时频域展开,然后通过计算相干系数确定信号的时频分布,最终通过小波包重构将信号从噪声中分离出来。这一算法不需要信号的任何先验知识,收敛快,可以有效地提取宽频带信号,极大地提高信号的信噪比,对非重复性信号具有良好的捕捉能力．应用此算法成功地实现了视觉诱发电位的单次提取。     

异质背景下黄河三角洲潮沟的遥感提取方法

以黄河三角洲为研究区域,针对该区域潮滩背景异质性强、潮沟宽度不一和潮沟各向异性强等特点,选取高分二号多光谱影像作为数据源,首先利用归一化水体指数(NDWI)和最大类间方差法(OTSU)提取宽阔潮沟,其次,使用改进的模糊C均值算法(MFCM)和多尺度高斯匹配滤波(MGMF),在削弱潮滩背景异质性的基础上增强细小潮沟,接着利用自适应阈值分割提取细小潮沟,最后合并细小潮沟和宽阔潮沟,形成完整的潮沟网络.不仅充分利用了高分二号影像的空间分辨率和光谱信息,也顾及了线状要素的几何特征,保证了潮沟提取结果的空间连续性.在4个局部测试区域,提取结果的Kappa系数大于0.8,总体精度高于97%,优于最大似然法和支持向量机.结果表明: 本研究提出的方法能够较完整地提取不同类型的潮沟,表现出较好的提取精度和稳定性,能够为潮沟的实时动态监测及其发育演化规律研究提供科学参考.     

三江平原农田土壤跳虫多尺度空间自相关性

土壤动物多尺度空间自相关性特征是土壤动物空间异质性和空间共存格局研究的重要基础。以土壤跳虫为研究对象,以中国科学院三江平原沼泽湿地生态实验站为研究区,于2011年8月和10月,选取具30 a以上耕作历史的旱地为实验样地,以5 m为步长设置一个50 m×50 m的实验样地,采用地统计中的全局Moran's I指数和局部Moran's I指数相结合的方法,研究土壤跳虫多尺度空间自相关性特征。全局空间自相关结果表明,土壤跳虫群落总密度和多数跳虫种类存在显著的多尺度空间自相关特征,多数为显著的正的空间自相关性,形成明显的空间集群。局部空间自相关结果表明,土壤跳虫群落总密度和所有跳虫种类均存在显著的局部空间自相关性,在局部地区形成"高—高"和/或"低—低"的空间集群,并和"高—低"和/或"低—高"的空间孤立区相伴发生,即形成"斑块"和"孔隙"相间分布的水平镶嵌格局;这种镶嵌格局具一定的时间动态特征,从夏季到秋季斑块大小和空间分布范围略有变化。研究表明,土壤跳虫具有明显的多尺度空间自相关性;空间集群是土壤跳虫的常见空间分布格局,在样地内形成"斑块"和"孔隙"相间分布的水平镶嵌格局,这种格局具年内时间变异性。     

利用参数模型提取诱发脑电（EP）及其在40Hz EEG和EP间关联研究中的应用

提出一种基于参数模型的信息处理方法,用于诱发脑电的提取,这种方法可以得到单次ＥＰ波形的无偏,最小方差估计,用刺激强度实验验证了该方法在实际情况下的有效,进一步将其应用于刺激前ＥＥＧ与ＥＰ的关联研究,得到了刺激前４０Ｈｚ的ＥＥＧ成分与晚潜伏期听觉诱发脑电幅度有极显著负相关的结果,表明刺激前及状态确定对ＥＰ有影响。     

Realistic model of a solid substrate fermentation packed-bed pilot bioreactor

For any given large-scale solid substrate fermentation (SSF) bioreactor, to assess how well a control system will work in practice requires the most realistic model possible. This model needs to account fully for complicated dynamic reactor behaviour and, in addition, has to include a specific noise model that is capable of reproducing the disturbances observed in SSF bioreactor measurements. In this work, noisy data collected historically from SSF pilot scale fermentations was used to develop such a model. Applying standard signal processing techniques, each measured variable was separated into deterministic and noise signals. Deterministic signals were used to calibrate a previously developed phenomenological model of the bioreactor. Noise signals were used to construct a realistic noise model for each measured variable in turn. Finally, the two models were combined to attain simulations that compared well with real measurements. This integrated model will provide realistic simulations that will prove useful in the design of effective control systems for intermittently mixed SSF bioreactors.     

Predictive controller evaluation including non-stationary high frequency noise and outliers for batch solid substrate fermentation bioreactors

Optimum operation and automatic control of large-scale solid substrate fermentation (SSF) bioreactors is difficult. Though advanced control algorithms can handle most challenges encountered properly, for real-time SSF processes such controllers are expensive and time consuming to design and tune. With these considerations, advanced control algorithm tests using realistic simulations appear more appropriate. We used a phenomenological process model of an SSF pilot bioreactor, coupled with a realistic noise model, to test linear model predictive controllers. We focused on the effect noise has on the performance of the control algorithms, and how to enhance performance using a combination of low-pass (Butterworth) and outlier shaving (Hampel) filters. In simulations undertaken directly with the phenomenological model it was relatively straightforward to achieve good control performance. Nevertheless, control degraded sharply when the output of the phenomenological model was contaminated with noise using our realistic noise model, even with proper signal filtering.     

Model selection and parameter estimation for ion channel recordings with an application to the K + outward-rectifier in barley leaf

We present a statistical method, and its accompanying algorithms, for the selection of a mathematical model of the gating mechanism of an ion channel and for the estimation of the parameters of this model. The method assumes a hidden Markov model that incorporates filtering, colored noise and state-dependent white excess noise for the recorded data. The model selection and parameter estimation are performed via a Bayesian approach using Markov chain Monte Carlo. The method is illustrated by its application to single-channel recordings of the K⁺ outward-rectifier in barley leaf.Acknowledgement The authors thank Sake Vogelzang, Bert van Duijn and Bert de Boer for their helpful advice and useful comments and suggestions.     

Pointwise Hölder exponents of a model for skin laser Doppler flowmetry signals based on six nonlinear coupled oscillators with linear and parametric couplings: Comparison with experimental data from young healthy subjects

We study pointwise Hölder exponents of experimental and numerically simulated skin laser Doppler flowmetry (LDF) data that give a peripheral view of the cardiovascular system. The experimental signals are recorded in the forearm of young healthy subjects. The numerically simulated LDF data are computed from a model containing six nonlinear coupled oscillators reflecting six almost periodic rhythmic activities present in experimental LDF signals. Simulated LDF signals are for the first time generated with both linear and parametric couplings in order to represent cardiovascular system behaviors. Moreover, we propose the use of a parametric generalised quadratic variation (GQV) based estimation method for the estimation of the pointwise Hölder exponents. The latter identify possible multifractal characteristics of data. The GQV method is first tested on a white noise measure and then applied on the LDF data. The results of our signal processing analysis show that experimental LDF signals recorded in the forearm are weakly multifractal for young healthy subjects at rest. Furthermore, our findings show that the simulated data have a complexity similar to the one of signal recorded in young healthy subjects. However, their pointwise Hölder exponents have differences that we explain. This paper provides useful information to go deeper into the modeling of LDF data, which could bring enlightenment for a better understanding of the peripheral cardiovascular system.     

Solid-state fermentation for production of griseofulvin on rice bran using Penicillium griseofulvum

Griseofulvin is a secondary metabolite produced from fungal species that have morphology suitable for solid-state fermentation (SSF). Reports on production of griseofulvin by SSF are scarce. The present work investigates SSF for griseofulvin production, optimization of its process parameters vis-à-vis the conventional submerged fermentation and its downstream processing from the same. Rice bran adjusted to an initial moisture content (IMC) of 50% (v/w) inoculated with 1 mL of a suspension of 10(6) spores/mL under agitation at 250 rpm containing the modified Czapek-Dox medium and additional 0.1% choline chloride as a precursor gave a yield of griseofulvin in 9 days that was comparable to submerged fermentation after 28 days. The yield of griseofulvin (microg/g dry biomass) was comparable in SSF and submerged fermentation. The biomass was estimated by estimation of chitin. Discussions on the effect of each parameter in SSF have also been included.     

脑磁图神经活动源数目的估计

在脑磁图信号的分析中,正确估计出脑磁图神经活动源的数目是进一步分析脑磁图信号的前提。目前广泛采用的信息论方法和主成分分析方法都是根据特征值来确定源的数目,这两种方法在源数目较多、噪声较强的情况下,会导致误判。该文提出了一种噪声调节自动阈值的脑磁图源数目判断方法,利用基于噪声调节的主成分分析并结合聂曼- 皮尔逊准则对脑磁图源数目进行估计。同时,该方法采用了基于小波的噪声方差估计,实现了脑磁图信号中噪声方差的精确估计。通过对基于信息论方法、主成分分析方法以及该文所提议方法的实验结果的比较,表明该文所提议方法能更准确地估计脑磁图源数目,特别是在源数目较多、信噪比较小的情况下,仍能准确地估计脑磁图源数目,具有较大的实际意义。     

Production of pectinase from deseeded sunflower head by Aspergillus niger in submerged and solid-state conditions

Studies were carried out on the production of pectinases using deseeded sunflower head by Aspergillus niger DMF 27 and DMF 45 in submerged fermentation (SmF) and solid-state fermentation (SSF). Higher titres of endo- and exo-pectinases were observed when medium was supplemented with carbon (4% glucose for SmF and 6% sucrose for SSF) and nitrogen (ammonium sulphate, 0.3% for both SmF and SSF) sources. Green gram husk proved to be relatively a better supplement to attain higher yield of endo-pectinase (11.7 U/g) and exo-pectinase (30.0 U/g) in solid-state conditions. Maximum production of endo-pectinase (19.8 U/g) and exo-pectinase (45.9 U/g) by DMF 45 were recorded in SSF when compared to endo-pectinase (18.9 U/ml) and exo-pectinase (30.3 U/ml) by DMF 27 in SmF under optimum process conditions.     

A better way to estimate population trends

Estimation of a population trend from a time series of abundance data is an important task in ecology, yet such estimation remains logistically and conceptually challenging in practice. First, the extent to which unequal intervals in the time series, due to missing observations or irregular sampling, compromise trend estimation is not well‐known. Furthermore, the predominant trend estimation method (loglinear regression of abundance data against time) ignores the possibility of process noise, while an alternative method (the ‘diffusion approximation’) ignores observation error in the abundance data. State‐space models that account for both process noise and observation error exist but have been little used. We study an adaptation of the exponential growth state‐space (EGSS) model for use with missing data in the time series, and we compare its trend estimation to the status quo methods. The EGSS model provides superior estimates of trend across wide ranges of time series length and sources of variation. The performance of the EGSS model even with half of the counts in the time series missing implies that trend estimates may be improved by diverting effort away from annual monitoring and towards increasing time series length or improving precision of the abundance estimates for years that data are collected.     

Least-squares methods for identifying biochemical regulatory networks from noisy measurements

Background  

We consider the problem of identifying the dynamic interactions in biochemical networks from noisy experimental data. Typically, approaches for solving this problem make use of an estimation algorithm such as the well-known linear Least-Squares (LS) estimation technique. We demonstrate that when time-series measurements are corrupted by white noise and/or drift noise, more accurate and reliable identification of network interactions can be achieved by employing an estimation algorithm known as Constrained Total Least Squares (CTLS). The Total Least Squares (TLS) technique is a generalised least squares method to solve an overdetermined set of equations whose coefficients are noisy. The CTLS is a natural extension of TLS to the case where the noise components of the coefficients are correlated, as is usually the case with time-series measurements of concentrations and expression profiles in gene networks.     

Attenuation of current and voltage noise signals recorded from epithelia

Transepithelially recorded current and voltage fluctuations are filtered by the impedance of the electrical equivalent parameters of the preparation, in series or in parallel, with the noise source. Fluctuations in voltage and current are assumed to be caused by fluctuations in conductance of the apical membrane. In order to obtain an estimation of the intrinsic noise amplitudes, calculations are presented to correct the transepithelial fluctuations. The influence of different model parameters on the recorded noise spectra is investigated. It is shown that the shape of the transepithelially recorded noise spectra may differ from the intrinsic ones, e.g. “peaking” in the power spectra may be explained by the assumption of a positive (referred to cell inside) e.m.f. at the basolateral membrane or a polarization impedance in series with the epithelium. Furthermore it is demonstrated that the ratio of voltage to current noise power may differ from the squared magnitude of the impedance.