Static and dynamic performance of a venturi airflow sensor

Venturi-meters have received limited application to the measurement of human airflow. A 3-D printed Venturi airflow sensor (fV) was integrated to a differential pressure sensor for the purpose to document signal quality, dynamic performance during instantaneous flow, and airflow analysis with subsequent calibration for volume measurement. In addition, we assessed these results for potential use in the measurement of human pulmonary ventilation. The fV was developed via drafting software program and 3-D printer (38 mm ID inlet, 12 mm ID constriction, and 23° conical transition sections). The fV was matched to a differential pressure transducer (0–7 kPa input per 0–10 VDC output). The fV was connected in series with a Pneumotachometer (fP) and Turbine airflow sensor (fT), with 1 m length tracheal tubing (ID: 35 mm), separating the devices. Airflow conditions were controlled by an industrial vacuum for constant flow with a manual-operated ball valve connected in series, and a criterion 3 L calibrated syringe for instantaneous flow. Repeated baseline data were collected for signal stability, and to identify the signal-to-noise ratio for signal quality. Airflow estimates and signal quality analysis was performed on repeated constant airflow conditions that spanned the transducer 0–10 VDC output. Repeated syringe manoeuvres that spanned the transducer 0–10 VDC output were used to determine the dynamic performance of the fV. A paired t-test revealed no statistical difference between fV and fP for signal quality across all airflow conditions (p = 0.2028). A paired t-test of dynamic performance data revealed no statistical significance between response times for the fV Vs fP. A calibration method was tested for validation in volume measurements, which resulted in a mean estimate = 3.032 L; CV = 1.14%; n = 83. The fV is easily manufactured with a durable and simple design, making its potential use for ventilation measurement an affordable and reliable technology. These findings provide a reasonable basis to pursue fV technology in applied validation studies, such as inspired and expired ventilation measurement.     

基于双模态深度自编码的孤立性肺结节诊断方法

近年来,深度学习技术在肺癌诊断方面得到了广泛的应用,但现有的研究主要集中于肺部CT图像。为了有效提高肺结节的诊断性能,提出一种基于双模态深度降噪自编码的肺结节诊断方法。首先,分别从肺部CT和PET图像中得到肺结节区域的特征信息;然后,以候选结节的PET/CT图像作为整个深度自编码网络的输入,并对高层信息进行学习;最后,采用融合策略对多种特征进行融合并将其作为整个框架的输出。实验结果表明,提出的方法可以达到92.81%的准确率、91.75%的敏感度和1.58%的特异性,且优于其他方法的诊断性能,更适用于肺结节良/恶性的辅助诊断。     

多输入卷积神经网络肺结节检测方法研究

针对传统计算机辅助诊断系统中肺部结节检出过程复杂,检出结果依赖于分类前期每个步骤的性能,以及存在假阳性率高的问题,提出了一种基于卷积神经网络的端到端的肺结节检测方法。该方法首先使用大量带标签的肺结节数据对构建的多输入卷积神经网络进行训练,实现从原始数据到语义标签的有监督学习。然后采用快速边缘检测方法和二维高斯概率密度函数构建候选区域模板,从待检测CT序列中获取候选区域并将其作为多输入卷积神经网络的输入数据。最后采用判定阈值实现疑似肺结节区域标注,同时在相邻的CT影像中进行重点检测。在LIDC-IDRI数据集上的大量实验结果表明,所提方法在肺部CT影像中对微、小结节的检出率较高;同时,重点检测模板能够小幅降低微、小结节检测的假阳率。     

The untapped potential of proteomic analysis in pediatric pulmonary hypertension

Analysis of the human proteome has become increasingly sophisticated, and offers invaluable potential insight into the pathophysiology of human disease. The increasing standardization of methods, speed, and sophistication of mass spectrometric analysis, availability of reliable antibodies, and dissemination of information among the scientific community has allowed for exponential growth of our knowledge base. The continued effort to provide a molecular explanation for future medical applications based on biomarker discovery is epitomized by the outstanding efforts of the human proteome project, whose goal is to generate a map of the human proteome. However, proteomic analysis is underrepresented in pediatric illness; given the unique challenges of research in the pediatric population, proteomic analysis represents enormous untapped potential, especially in the further elucidation of the pathophysiology of rare diseases such as pulmonary hypertension (PH). In this article, we will describe the unique challenge of pediatric research, the importance of alternative avenues such as proteomics for in-depth analysis of pediatric pathobiology at the cellular level, the specific need for proteomic investigation of pediatric PH, the current status of PH proteomics, and future directions.     

Utility of a novel error-stepping method to improve gradient-based parameter identification by increasing the smoothness of the local objective surface: A case-study of pulmonary mechanics

Accurate model parameter identification relies on accurate forward model simulations to guide convergence. However, some forward simulation methodologies lack the precision required to properly define the local objective surface and can cause failed parameter identification.     

老年糖尿病合并肺结核52例临床分析

目的了解糖尿病合并肺结核的临床特点,提高对老年糖尿病肺结核的认识,争取早发现、早治疗。方法对2年间52例老年糖尿病肺结核病例进行临床分析。结果治疗3个月后,血糖控制在正常范围者32例,结核病变吸收和好转者39例,痰菌阴转者25例,空洞闭合或缩小者7例,空洞无变化者3例,死亡2例。结论积极控制血糖,使血糖控制在正常水平,合理的抗结核治疗对控制老年糖尿病肺结核亦起积极作用。     

Convergence conditions for iterative methods seeking multi-component solitary waves with prescribed quadratic conserved quantities

We obtain linearized (i.e., non-global) convergence conditions for iterative methods that seek solitary waves with prescribed values of quadratic conserved quantities of multi-component Hamiltonian nonlinear wave equations. These conditions extend the ones found for single-component solitary waves in a recent publication by Yang and the present author. We also show that, and why, these convergence conditions coincide with dynamical stability conditions for ground-state solitary waves.Notably, our analysis applies regardless of whether the number of quadratic conserved quantities, s, equals or is less than the number of equations, S. To illustrate the situation when s < S, we use one of our iterative methods to find ground-state solitary waves in spin-1 Bose-Einstein condensates in a magnetic field (s = 2, S = 3).     

人发中微量元素与某些呼吸系统疾病关系的研究

用仪器中子活化法测定了正常人和慢性支气管炎,肺气肿,肺心病患者头发中Ｃａ,Ｍｇ,Ｆｅ,Ｓｅ,Ｃｕ,Ｚｎ等１９种元素的含量。测得患者发中Ｃａ,Ｍｇ含量低于正常人,Ｆｅ,Ａｓ,Ｃｏ等含量高于正常人;慢性支气管炎急发期患者发中Ｃａ,Ｍｇ含量低于缓解期患者,Ｆｅ含量高于缓解期患者,均有显著性差异（Ｐ＜０．０５－０．０１）。人发Ｃａ与Ｍｇ呈高度正相关（Ｐ＜０．００１）。还测定了用于治疗慢性支气管炎的中药“咳     

大鼠吸入~(239)PuO_2后肺巨噬细胞核吸收剂量的估算

本文利用大鼠吸入~(239)PuO_2后肺钚滞留函数,肺巨噬细胞的更新函数及有关的生物学资料,采用 Monte-Carlo 方法,估算了大鼠吸入~(239)PuO_2后肺和肺巨噬细胞核的平均吸收剂量。对于初始肺钚沉积量分别为8.7、6.7、2.7和3.0kBq,照射时间分别为2、7、13和28d 的4组实验动物,估算出肺平均吸收剂量(?)_P为0.69、1.7、1.2和2.6Gy,肺巨噬细胞陔的平均吸收剂量(?)_(PM)为0.95、2.21、1.71和3.23Gy。(?)_(PM)与(?)_P的比值的平均值为1.3。肺巨噬细胞核的受照剂量主要求自沉积于肺巨噬细胞内的钚,周围钚粒子的剂量贡献仅占1%左右。