Short utterance recognition using a network with minimum training

A feedforward network is used to recognize short, digitized, isolated utterances. A high, multispeaker recognition rate is achieved with a small vocabulary with a single training utterance. This approach makes use of the pattern recognition property of the network architecture to classify different temporal patterns in the multidimensional feature space. The network recognizes the utterances without the need of segmentation, phoneme identification, or time alignment. We train the network with four words spoken by one single speaker. The network is then able to recognize 20 tokens spoken by 5 other speakers. We repeat the above training and testing procedure using a different speaker's utterances for training each time. The overall accuracy is 97.5%. We compare this approach to the traditional dynamic programming (DP) approach, and find that DP with slope constraints of 0 and 1 achieve 98.5% and 85% accuracies respectively. Finally we validate out statistics by training and testing the network of a four-word subset of the Texas Instruments (Tl) isolated word database. The accuracy with this vocabulary exceeds 96%. By doubling the size of the training set, the accuracy is raised to 98%. Using a suitable threshold, we are able to raise the accuracy of one network from 87% to 98.5%. Thresholding applied to all networks would then raise the overall accuracy to well over 99%.

This technique is especially promising because of the low overhead and computational requirements, which make it suitable for a low cost, portable, command recognition type of application.     

The universal algorithm of maturation for secretory and excretory protein precursors.

During maturation, most proteins undergo different posttranslational modifications. In most simple cases, signal peptidases remove the signal or leader peptide from the precursors of the secretory proteins during their translocation across the ER membrane. For biologically active proteins, such as enzymes, regulatory and defense proteins, toxins, etc., additional maturation-regulating mechanisms were shown to proceed with limited proteolysis of inactive precursors by specific enzymes. A number of specific enzymes from different cell types selectively cleave proproteins at specific processing sites. In this work, we analyzed the sequences of protein precursors synthesized in the excretory glands of different animals and identified new, non-traditional processing sites. They differ from the motifs previously identified in secreted proteins' precursors and enabled us to reconstruct the sequence of events leading to the conversion of protein precursors into the final products (mature proteins). We also found that in animals, the maturation mechanism of secretory and excretory proteins and the set of enzymes involved are species specific. The processing sites identified in protein precursors in this study are useful for a more detailed genome analysis and more accurate mature protein sequence prediction.     

睡眠状态下人体生理信号模糊分析系统的研究

采用LabVIEW技术完成对心电、呼吸、无创血氧和无创血压信号的采集与处理,而后通过MATLAB模糊测量系统对采集的人体生理信号进行模糊预测分析。通过实验检验了人体生理信号检测、处理和模糊预测分析方法的可靠性。LabVIEW技术对人体生理信号的模糊预测分析是有效的,可用于检测阻塞性睡眠呼吸暂停患者的生理状态信号。     

熵指数对异氟烷麻醉中切皮体动反应的预测作用

目的 研究熵指数(包括反应熵和状态熵)对异氟烷复合麻醉中切皮体动反应的预测作用并与双频谱指数(BIS)比较.方法 ASA Ⅱ级择期腹部手术患者20例,按切皮有无体动反应分为体动组(9例)和非体动组(11例).入室后常规监测,然后静注异丙酚、琥珀胆碱、芬太尼快诱导气管插管,持续吸入异氟烷,维持呼气末异氟烷浓度为1.3MAC,维持10min后开始手术,切皮前不用神经肌肉阻滞剂,如果发生体动,则按10%提高下一个患者异氟烷呼气末浓度,如果不动则按10%降低下一个患者异氟烷呼气末浓度.分别于入室、切皮前60s、30s、切皮后45s、120s对每例患者同时行熵指数和BIS监测.结果 两组患者的熵指数值有显著性差异(P＜0.05),而BIS无显著性差异(P＞0.05).反应熵和状态熵可以预测切皮体动,其预测概率(Pk)值分别为0.893 7和0.885 2,BIS不能预测切皮体动,Pk值是0.591 9,熵指数和BIS的Pk值之间有显著性差异(P＜0.05),Pk与反应熵和状态熵的Logistic回归方程分别为ln [Pk/(1-Pk)]=32.517-0.481x和ln [Pk/(1-Pk)]=28.728-0.438x.结论 熵指数监测可以预测异氟烷麻醉中的切皮体动反应,值得临床推广应用.     

Rh phenotype prediction by DNA typing and its application to practice

The complexity of the RHD and RHCE genes, which is the greatest of all blood group systems, confounds analysis at the molecular level. RH DNA typing was introduced in 1993 and has been applied to prenatal testing. PCR-SSP analysis covering multiple polymorphisms was recently introduced for the screening and initial characterization of partial D. Our objective is to summarize the accrued knowledge relevant to the approaches to Rh phenotype prediction by DNA typing, their possible applications beyond research laboratories and their limitations. The procedures, results and problems encountered are highly detailed. It is recommended that DNA typing comprises an analysis of more than one polymorphism. We discuss future directions and propose a piecemeal approach to improve reliability and cost-efficiency of blood group genotyping that may eventually replace the prevalent serology-based techniques even for many routine tasks. Transfusion medicine is in the unique position of being able to utilize the most extensive phenotype databases available to check and develop genotyping strategies.     

应用因特网对SARS病毒M蛋白B细胞抗原表位的预测

目的：预测SARS冠状病毒M蛋白的B细胞表位。方法：采用EMBOSS软件结合Hopp＆woods亲水性参数、Janin可及性参数、极性参数、柔韧性参数及二级结构方案对SARS冠状病毒M蛋白的B细胞表位进行了预测，并用吴玉章等已建立的B细胞表位预测方法进行评述。结果：SARS冠状病毒M蛋白B细胞识别的表位可能位于第3～13和153～166位氮基酸等区域内或附近，这2个被预测的表位均含有β—转角和无规卷曲结构。结论：该研究对应用合成肽抗原进行SARS早期诊断研究及相关抗体的制备具有重要指导意义。     

缓和加氢裂化动力学模型

参考缓和加氢裂化（ＭＨＣ）反应机理，开发了一种既能预测产品产率，又能预测产品性质的ＭＨＣ动力学网络模型。以文献上发表的ＭＨＣ实验数据为例，进行了参数估计，确定了动力学参数，给出了模型的标准差。本模型所预测的ＭＨＣ结果，与文献数据吻合良好，具有良好的预测性能。     

SARS病毒S蛋白部分片段B-细胞表位的多参数预测

目的预测SARS病毒S蛋白(spike glycoprotein)部分片段(aa No.400～600 & aa No.1 100～1 195)的B-细胞表位.方法应用多种参数和方法进行综合分析,包括跨膜分析、保守区分析、同源性比较、Hopp ＆ Woods亲水性参数、抗原性参数、可及性参数、β-转角、万氏法等综合预测方法.结果显示B-细胞识别的表位可能在436～456和1 136～1 146残基或其附近,这两个被预测的表位均含有β-转角和不规则卷曲结构.结论本研究为应用合成肽抗原制备抗SARS病毒S蛋白抗体、诊断非典型肺炎(severe acute respiratory syndrome, SARS)提供了依据.     

Serum digoxin and beta-methyldigoxin in elderly patients on hospital admission: correlation with home compliance and clinical variables

Summary Serum digoxin and beta-methyldigoxin (BMD) were measured in 165 elderly patients (age >60 years) admitted to hospital, of whom 109 had been treated at home with digoxin and 56 with BMD.The mean BMD level was significantly lower than that of digoxin (1.1 vs. 1.4 ng/ml). Creatinine clearance and daily dose were the variables most strongly associated with digoxin level, and the prescribed dose and serum albumin were the best predictors of the BMD concentration. Compliance was assessed by a compliance index (CI), namely the ratio of the measured glycoside concentration, corrected for creatinine clearance, over the expected steady-state dose, calculated from a hospitalized reference group. Compliant individuals in both treatment groups, i.e. those with a CI > the median value, were characterized by a lower daily dose and dosage frequency.Toxicity, whether clinical or electrocardiographic, was present in 9% of the patients and was associated only with a significantly higher mean serum level of the drug.