Near Infrared Spectroscopy (NIRS) Model-Based Prediction for Protein Content in Cowpea

Cowpea (Vigna unguiculata L. Walp) is a multi-purpose legume with high quality protein for human consumption and livestock. The objective of this work was to develop near-infrared spectroscopy (NIRS) prediction models to estimate protein content in cowpea. A total of 116 cowpea breeding lines with a wide range of protein contents (19.28 % to 32.04%) were selected to build the model using whole seed and ground seed samples. Partial least-squares discriminant analysis (PLS-DA) regression technique with different pre-treatments (derivatives, standard normal variate, and multiplicative scatter correction) were carried out to develop the protein prediction model. Results showed: 1) spectral plots of both the whole seed and ground seed showed higher spectral scatter at higher wavelengths (>1450 nm), 2) data pre-processing affects prediction accuracy for bot whole seed and ground seed samples, 3) prediction using ground seed samples (0.64 R2 0.85) is better than the whole seed (0.33 R2 0.78), and 4) the data pre-processing second derivative with standard normal variate has the best prediction (R2_whole seed = 0.78, R2_ground seed = 0.85). The results will be of interest in cowpea breeding programs aimed at improving total seed protein content.

In situ infrared, Raman and X-ray spectroscopy for the mechanistic understanding of hydrogen evolution reaction

Hydrogen production by water reduction reactions has received considerable attention because hydrogen is considered a clean-energy carrier,key for a sustainable energy future.Computational methods have been widely used to study the reaction mechanism of the hydrogen evolution reaction(HER),but the calculation results need to be supported by experimental results and direct evidence to confirm the mechanistic insights.In this review,we discuss the fundamental principles of the in situ spectroscopic strategy and a theoretical model for a mechanistic understanding of the HER.In addition,we investigate recent studies by in situ Fourier transform infrared(FTIR),Raman spectroscopy,and X-ray absorption spectroscopy(XAS) and cover new findings that occur at the catalyst-electrolyte interface during HER.These spectroscopic strategies provide practical ways to elucidate catalyst phase,reaction intermediate,catalyst-electrolyte interface,intermediate binding energy,metal valency state,and coordination environment during HER.

Application of near-infrared spectroscopy for fast germplasm analysis and classification in multi-environment using intact-seed peanut(Arachis hypogaea L.)

Peanut is a worldwide oilseed crop and the need to assess germplasm in a non-destructive manner is important for seed nutritional breeding.In this study,Near Infrared Spectroscopy(NIRS)was applied to rapidly assess germplasm variability from whole seed of 699 samples,field-collected and assembled in four genetic and environmentbased sets:one set of 300 varieties of a core-collection and three sets of 133 genotypes of an interspecific population,evaluated in three environments in a large spatial scale of two countries,Mbalmayo and Bafia in Cameroon and Nioro in Senegal,under rainfed conditions.NIR elemental spectra were gathered on six subsets of seeds of each sample,after three rotation scans,with a spectral resolution of 16 cm-1over the spectral range of867 nm to 2530 nm.Spectra were then processed by principal component analysis(PCA)coupled with Partial least squares-discriminant analysis(PLS-DA).As results,a huge variability was found between varieties and genotypes for all NIR wavelength within and between environments.The magnitude of genetic variation was particularly observed at 11 relevant wavelengths such as 1723 nm,usually related to oil content and fatty acid composition.PCA yielded the most chemical attributes in three significant PCs(i.e.,eigenvalues>10),which together captured 93%of the total variation,revealing genetic and environment structure of varieties and genotypes into four clusters,corresponding to the four samples sets.The pattern of genetic variability of the interspecific population covers,remarkably half of spectrum of the core-collection,turning out to be the largest.Interestingly,a PLS-DA model was developed and a strong accuracy of 99.6%was achieved for the four sets,aiming to classify each seed sample according to environment origin.The confusion matrix achieved for the two sets of Bafia and Nioro showed 100%of instances classified correctly with 100%at both sensitivity and specificity,confirming that their seed quality was different from each other and all other samples.Overall,NIRS chemometrics is useful to assess and distinguish seeds from different environments and highlights the value of the interspecific population and core-collection,as a source of nutritional diversity,to support the breeding efforts.

Noninvasive Detection of Cerebral Blood Flow and Blood Oxygen Based on Inductive Sensing Technology and Near Infrared Spectroscopy

The synchronous monitoring of cerebral blood flow and blood oxygen levels plays a pivotal role in the prevention,diagnosis,and treatment of cerebrovascular diseases.This study introduces a novel noninvasive device utilizing inductive sensing and near-infrared spectroscopy technology to facilitate simultaneous monitoring of cerebral blood flow and blood oxygen levels.The device consists of modules for cerebral blood flow monitoring,cerebral blood oxygen monitoring,control,communication,and a host machine.Through experiments conducted on healthy subjects,it was confirmed that the device can effectively achieve synchronous monitoring and recording of cerebral blood flow and blood oxygen signals.The results demonstrate the device’s capability to accurately measure these signals simultaneously.This technology enables dynamic monitoring of cerebral blood flow and blood oxygen signals with potential clinical applications in preventing,diagnosing,treating cerebrovascular diseases while reducing their associated harm.

Investigation of Prefrontal Cortex Activity in University Students with Presenteeism: A Near-Infrared Spectroscopy (NIRS) Study

Presenteeism refers to impaired performance attributed to attending work with health problems. There has been no study examining the state of presenteeism with objective measures. We compared cerebral hemodynamic changes, measured by near-infrared spectroscopy (NIRS), during neuropsychological tests conducted by university students with presenteeism and healthy controls. Twenty-two university students participated in the study;11 of them with impaired performance caused by mental health problem were allocated to the presenteeism group and 11 without health problems to the control group. Presenteeism was assessed by the Presenteeism Scale for Students. To evoke hemodynamics changes, the participants completed a Word Fluency Test (WFT) and a Trail Making Test (TMT). The NIRS probes were located over the bilateral prefrontal area. Students with presenteeism had significantly higher incidences of depression than controls. However, there was no significant difference in behavioral performance examinations between the two groups. With regard to hemodynamics changes, the repeated measures analysis of covariance of the NIRS signals revealed significant interactions between group and task activation. Although we observed a significant increase in oxygenated hemoglobin concentration during the WFT among controls (simple main effect;left channel, F(1, 19) = 27.34, P F(1, 19) = 22.05, P < 0.001), no changes were found in students with presenteeism during either the WFT (simple main effect;left channel, F(1, 19) = 0.12, P F(1, 19) = 0.08, P t = ﹣0.94, P with Bonferroni correction = 0.745;right channel, t = ﹣2.19, P with Bonferroni correction < 0.113). This is the first study to reveal differences in activity in the cerebral cortex associated with presenteeism. The fact that students with presenteeism have prefrontal dysfunction might reinforce the concept of presenteeism.

Simulation and Modelling of near Infrared Spectroscopy (NIRS) as Brain Monitor

Near infrared spectroscopy (NIRS) is a method for non-invasive monitoring of cerebral oxygenation and haemodynamics. Different devices provide information on changes of oxygenated (HbO2) and deoxygenated haemoglobin (HHb), oxidized cytochrome aa3 (CytOx) or regional oxygen saturation (rSO2). NIRS has been used during adult and paediatric cardiac surgery.

Correlation of prefrontal activity measured by near-infrared spectroscopy (NIRS) with mood, BDNF genotype and serum BDNF level in healthy individuals

Depression has been known to reduce the prefrontal activity associated with the execution of certain cognitive tasks, although whether a temporarily depressed or anxious mood in healthy individuals affects the prefrontal blood oxygen level during cognitive tasks is unknown. Combining the measurement of prefrontal activity with near-infrared spectroscopy (NIRS) and the two cognitive tasks, namely the letter version of the verbal fluency test (VFT-l) and the Stroop test, we measured the effect of a depressed or anxious mood and gender on the changes in the prefrontal oxygenated hemoglobin (Oxy-Hb) levels during those cognitive tests in healthy individuals. Depressed mood or anxious mood was assessed by the Hospital Anxiety and Depression Scale (HADS). Thereby we aimed to explore the possibility of NIRS measurement for detecting the early subclinical manifestation of major depression. Moreover, we examined the possible relationships between prefrontal activation and the functional Val66Met polymorphisms of the brain derived neurotropic factor (BDNF) gene and serum BDNF level. As a result, the increased prefrontal Oxy-Hb levels during cognitive tasks were significantly correlated with the severity of depressed mood in males. The course of the prefrontal Oxy-Hb increase was different depending on the cognitive tasks, i.e., the VFT-l or the Stroop test, in both genders. Correlations of BDNF genotype and serum BDNF level with the prefrontal Oxy-Hb levels during those cognitive tasks were negative. Our results suggest that the early subclinical manifestation of depressed mood in males might be detected by the NIRS measurement, which is not correlated with the individual properties of BDNF.

WIRELESS DISTRIBUTED ACQUISITION SYSTEM FOR NEAR INFRARED SPECTROSCOPY-WDA-NIRS

The wireless distributed acquisition system for near infrared spectrosecopy(WDA-NIRS)is a portable,ultra-compact,contimuous wave(CW)NIRS system.Its main advantage is that it allows continuous synchronized multi-site hemodynarnic monitoring.The WDA-NIRS syster calculates online changes in hemoglobin concentration based on modifed Beer-Lambert law and the tissue oxy genation index based on the spatial resolved spectroscopy method.It consists of up to seven signal acquisition units,sufficiently small to be easily attached to any part of the body.These units are remotely synchronized by a PC base station for independent acquisition of NIRS signals.Each acquisition module can be freely adapted to individual requirements such as local skin properties and the microcirculation of interest,e.g,different muscles,brain,skin,etc.For this purpose,the light emitted by each LED can be individually,interactively or automatically adjusted to local needs.Furthermore,the user can freely create an emitter time multiplexing protocol and choose the detector sensitivity most suitable to a particular situation.The potential diagnostic value of this advanced device is demonstrated by three typical applications.

Correlation between LIFG and Autonomic Activation during Stressful Tasks:A Functional Near-infrared Spectroscopy (fNIRS) Study

It remains unclear whether language tasks in one＇s first （L1） or second （L2） language can cause stress responses and whether frontal, autonomic and behavioral responses to stressful tasks are correlated. In this study, we studied 22 Chinese subjects whose L2 was English and measured the cerebral blood oxygenation in their frontal lobe by using functional near-infrared spectroscopy （fNIRS） as par- ticipants engaged in a mental arithmetic task （MAT） and verbal fluency tasks （VFTs） in L1 （Chinese） and L2 （English）. To examine the activated cortical areas, we estimated the channel location based on Montreal Neurological Institute （MNI） standard brain space by using a-probabilistic estimation method. We evaluated heart rate （HR） changes to analyze autonomic nervous system （ANS） functioning. We found that the MAT and VFTs induced greater increases in HR than did the control （Ctrl） task. Further- more, subjects developed greater increases in HR in the MAT and VFTt~ than they did in the VFTL1. Compared with the Ctrl task, the MAT and both VFTLland VFTL2 produced robust and widespread bi- lateral activation of the frontal cortex. Interestingly, partial correlation analysis indicated that the activity in the left inferior frontal gyrus （LIFG） [Brodmarm＇s area （BA） 47] was consistently correlated with the increases in HR across the three tasks （MAT, VFTL2, and VFTL1）, after controlling for the performance data. The present results suggested that a VFT in L2 may be more stressful than in L1. The LIFG may affect the activation of the sympathetic system induced by stressful tasks, includin~ MATs and VFTs.

幼儿体质健康水平与工作记忆加工的差异特征:来自fNIRS的证据

目的:探讨幼儿体质健康发展水平与工作记忆加工的大脑神经关联,为幼儿体质健康促进与认知发展提供依据。方法:选取98名4~6岁学龄前儿童,按照《国民体质测定标准手册(幼儿部分)》的内容评估幼儿体质健康发展水平,采用功能性近红外光谱脑成像(fNIRS)技术,监测幼儿执行工作记忆任务时的大脑前额叶皮层血氧动力学信号,采用4(体质健康,分为优秀、良好、合格、不合格)×2(记忆难度,分为低记忆难度、高记忆难度)的混合实验设计探讨幼儿体质健康水平与大脑工作记忆时的行为学和前额叶氧合血红蛋白浓度差异。结果:1)在低记忆难度条件下,体质健康优秀组的氧合血红蛋白浓度在右侧背外侧前额叶皮层(R-DLPFC)显著大于良好组,在R-DLPFC、右侧三角部布罗卡氏区(R-PTBA)、右侧额极区(R-FPA)及左侧背外侧前额叶皮层(L-DLPFC)显著大于合格组,在L-DLPFC显著大于不合格组;2)在高记忆难度条件下,优秀组的氧合血红蛋白浓度在R-DLPFC显著大于良好组,在R-DLPFC、R-PTBA、R-FPA及L-DLPFC显著大于合格组和不合格组;良好组的氧合血红蛋白浓度在L-DLPFC和R-PTBA显著大于合格组,在R-DLPFC、R-PTBA、R-FPA及L-DLPFC显著大于不合格组,合格组的氧合血红蛋白浓度在R-DLPFC、R-FPA及L-DLPFC的氧合血红蛋白浓度显著大于不合格组;3)优秀组和良好组在高记忆难度条件下的R-DLPFC、R-PTBA、R-FPA及L-DLPFC显著大于低记忆难度条件下;4)合格组在高记忆难度条件下的R-FPA和L-DLPFC的氧合血红蛋白浓度显著大于低记忆难度条件下;5)不合格组在高记忆难度条件下的R-PTBA的氧合血红蛋白浓度显著大于低记忆难度条件下。结论:幼儿的体质健康发展水平与工作记忆存在密切关联,表现为体质健康发展水平越高,大脑在执行工作记忆任务时的前额叶相关皮层激活程度越高。

定向运动练习对大学生空间定向能力的影响:来自行为学和fNIRS的证据

探讨定向运动练习对大学生空间定向能力的影响及其脑机制,为提升人的空间认知能力提供新途径和科学依据。以48名大一学生为研究对象,随机分为实验组和对照组,对实验组进行8周定向运动练习干预,采用功能性近红外光谱成像系统(fNIRS)记录被试者大脑氧合血红蛋白(Oxy-Hb)浓度变化,分析被试者干预前后空间定向认知任务的行为表现及神经机制。结果表明:(1)定向运动练习显著提升练习者空间定向认知任务的正确率且显著优于对照组。(2)在空间定向任务中,实验组左侧腹外侧前额叶(L-VLPFC)和左侧背外侧前额叶(L-DLPFC)的Oxy-Hb激活显著降低且低于对照组。(3)实验组腹外侧前额叶(VLPFC)的Oxy-Hb浓度与正确率显著相关。研究认为,定向运动练习可以有效改善练习者的空间定向认知能力,且空间定向能力的改善效益与大脑前额叶皮层激活程度有关。定向运动练习改善练习者空间定向能力的脑机制可能是通过改变练习者的空间认知加工模式,从而促进其脑神经活动的优化。

EEG-fNIRS技术在神经精神疾病研究中的应用进展

通过脑电图(EEG)、功能近红外光谱成像(fNIRS)、功能磁共振成像对神经精神疾病已经开展广泛的研究与应用。近年来随着技术的不断发展,EEG与fNIRS的同步采集设备被开发且逐步应用于神经精神疾病的研究中。本文首先对EEG-fNIRS设备的同步检测与数据分析技术进行简要概述,总结脑卒中、癫痫以及其他神经精神疾病研究中EEG-fNIRS的分析方法与最新发现,并探讨研究的发展方向。

基于卷积自编码的fNIRS信号运动校正算法研究

功能性近红外光谱技术(functional near-infrared spectroscopy,fNIRS)作为一种高时间分辨率、成本低廉、便携性高的脑成像系统,近年来深受脑神经科学等研究领域的关注。但fNIRS信号中的运动伪迹会干扰后期数据分析的结果,且现有的一些算法去噪效果较为单一。因此,本文提出了一种基于多层卷积自编码的fNIRS信号运动伪迹校正算法——MCAN算法,并使用该算法对fNIRS信号中的3种运动伪迹进行校正;然后用仿真数据和实验数据对所提算法的性能进行验证,将其与现有的几种常用算法进行对比,结果表明:MCAN算法在剩余运动伪迹数量、均方误差、信噪比、皮尔逊相关系数的平方、峰峰误差几种指标上表现良好,说明所提算法可作为一种全新的fNIRS信号预处理算法。

基于fNIRS对女性不同膀胱状态下盆底肌收缩任务的前额叶激活情况

目的为探讨前额叶在人体控尿功能中发挥的作用提供神经影像学依据。方法利用近红外脑功能成像技术(fNIRS)收集20名健康成年女性志愿者膀胱充盈和空虚状态下盆底肌收缩时大脑前额皮层的血流动力学数据,并对数据进行相应的处理,通过分析各个通道所提取的氧合血红蛋白浓度的相对变化量所对应的Beta值,对比前额叶不同脑区间激活情况差异。结果膀胱充盈状态下前额皮层共有30个通道激活,膀胱空虚状态下前额皮层共有8个通道激活(P均<0.05),其中共同激活通道有7个;相较于空虚状态,膀胱充盈时的前额叶激活更明显,且以右侧脑区激活为主,差异主要体现在右侧背外侧前额叶和额极(P均<0.05)。结论盆底肌收缩运动可以使大脑前额皮层被激活。当膀胱处于充盈状态时大脑前额皮层可能会通过神经反射活动感知膀胱的压力变化,从而参与调控盆底肌的自主舒缩运动,实现其在人体控尿功能中的作用,其中右侧背外侧前额叶脑区在此过程中发挥的作用可能更为明显。

基于fNIRS的主动、被动与想象屈肘运动脑激活的特点及比较

目的 利用功能性近红外光谱成像(functional near-infrared spectroscopy, fNIRS)技术探究健康成年人主动屈肘、被动屈肘和想象屈肘运动脑激活的特点。方法 选取30名健康大学生为被试者,男女各15例,年龄(21.22±1.42)岁,选择右手主动、被动和想象屈肘任务作为刺激模式,采用近红外光谱脑功能成像系统实时采集被试者3种不同运动模式下脑功能区氧合血红蛋白(Oxyhemoglobin, HbO2)水平作为脑激活的特征值,并以静息态(resting state, RS)的HbO2水平为基线值,比较3种运动模式下脑激活的特点。结果 非参数检验结果表明,不同运动模式下,右侧初级运动皮层(H=13.309,P=0.004),右侧前额叶前部皮层(H=20.30,P<0.001),右侧背外侧前额叶皮层(H=12.194,P=0.007),左侧初级体感皮层(H=12.642,P=0.005),左侧初级运动皮层(H=20.03,P<0.001),左侧前额叶前部皮层(H=18.187,P<0.001),左侧背外侧前额叶皮层(H=13.379,P=0.004)的HbO2水平比较差异有统计学意义;而其右侧初级体感皮层(H=4.284,P=0.232),右侧前运动区和辅助运动皮层(H=4.201,P=0.241),左侧前运动区和辅助运动皮层(H=8.793,P=0.052)HbO2水平比较差异无统计学意义。结论 不同运动模式下存在特定的脑功能区激活,其激活程度既有相似性又有差异性,主动运动和被动运动任务下的脑激活区最类似。

Near-infrared spectroscopy method for rapid proximate quantitative analysis of nutrient composition in Pacific oyster Crassostrea gigas

Glycogen,amino acids,fatty acids,and other nutrient components affect the flavor and nutritional quality of oysters.Methods based on near-infrared reflectance spectroscopy(NIRS)were developed to rapidly and proximately determine the nutrient content of the Pacific oyster Crassostreagigas.Samples of C.gigas from 19 costal sites were freeze-dried,ground,and scanned for spectral data collection using a Fourier transform NIR spectrometer(Thermo Fisher Scientific).NIRS models of glycogen and other nutrients were established using partial least squares,multiplication scattering correction first-order derivation,and Norris smoothing.The R_(C) values of the glycogen,fatty acids,amino acids,and taurine NIRS models were 0.9678,0.9312,0.9132,and 0.8928,respectively,and the residual prediction deviation(RPD)values of these components were 3.15,2.16,3.11,and 1.59,respectively,indicating a high correlation between the predicted and observed values,and that the models can be used in practice.The models were used to evaluate the nutrient compositions of 1278 oyster samples.Glycogen content was found to be positively correlated with fatty acids and negatively correlated with amino acids.The glycogen,amino acid,and taurine levels of C.gigas cultured in the subtidal and intertidal zones were also significantly different.This study suggests that C.gigas NIRS models can be a cost-effective alternative to traditional methods for the rapid and proximate analysis of various slaughter traits and may also contribute to future genetic and breeding-related studies in Pacific oysters.

基于fNIRS和神经反馈的记忆认知训练研究

该研究使用功能性近红外光谱技术和神经反馈机制开发了记忆任务训练系统,并对被试的脑电信号进行采集和分析。结果显示,参与神经反馈任务的被试在前额叶部右半脑区具有更高的相关性脑网络节点度和平均集群系数,中介中心度的分散程度相对降低。此外,被试的前额叶部左半脑区在神经反馈任务中的核心地位上升。通过通道网络模型和支持向量机模型对脑数据进行分类,结果表明两个模型在任务态和静息态的分类准确率高于反馈任务和对照任务,且通道网络模型的分类准确率更高。该研究结果表明神经反馈任务的被试具有明显的脑数据特征,并且这些特征能够被有效识别。

基于NIRS技术的款冬花药材质控指标定量分析模型的建立

目的建立基于近红外光谱(NIRS)技术的款冬花药材中款冬酮、水分、醇溶性浸出物和总灰分含量的定量分析模型,为款冬花药材及其制剂的快速质量评价提供新思路。方法参照2020年版《中国药典》,分别采用高效液相色谱法、烘干法、热浸法及灰分测定法测定19个产地130批款冬花药材中主要质控指标款冬酮、水分、醇溶性浸出物、总灰分的含量,采集款冬花药材的NIRS数据信息,然后采用NIRS结合偏最小二乘法建立样品中上述质控指标的各个定量分析模型,经验证集样品验证后得到NIRS含量预测模型。结果130批款冬花药材样品中款冬酮、水分、醇溶性浸出物和总灰分的含量范围分别为0.0514%~0.1035%、7.75%~10.93%、20.17%~31.12%、7.68%~12.10%。所建立的款冬花药材中款冬酮、水分、醇溶性浸出物和总灰分定量分析模型的内部交叉验证决定系数(R2)分别为0.9858、0.9684、0.9734、0.9880;校正集均方差(RMSEC)分别为0.00154、0.187、0.478、0.127;预测均方差(RMSEP)分别为0.00181、0.212、0.543、0.149;RMSEP/RMSEC分别为1.1753、1.1337、1.1360、1.1732,均在合理范围内(1<RMSEP/RMSEC≤1.2)。验证集样品中上述4个质控指标真实值与模型预测值的平均绝对误差分别为-0.00036、0.06143、0.14400和0.01043,平均预测回收率分别为99.65%、100.72%、100.66%和100.15%。结论所建NIRS定量分析模型稳定性好、测定结果可靠,可用于款冬花药材中相关质控指标含量的快速批量预测。

基于LSTM长短期记忆网络的樟子松木材气干密度NIRS模型预测

【目的】木材密度不仅与木材的各种材性密切相关,而且是衡量木材质量与价值的重要指标。采用近红外光谱(Near infrared spectroscopy,NIRS)分析技术能够快速、高效地预测木材密度,避免了传统试验中繁琐的检测步骤。长短期记忆网络(Long short-term memory network,LSTM)作为循环神经网络(Recurrent neural network,RNN)的变体,不仅可以学习序列数据之间的高阶特征信息,而且克服了RNN中的长距离依赖、梯度爆炸与梯度消失等问题。将LSTM与NIRS结合,提出一种能够准确预测樟子松木材气干密度的无损检测技术,为提高NIRS模型预测木材气干密度精度提供理论依据。【方法】该研究以樟子松木材样本为研究对象,用近红外光谱仪获得106个樟子松样本的光谱数据,并在恒温(20±2℃)恒湿(65%±3%)的环境下测定样本的气干密度。通过对比多组预处理方法和特征选择方法,采用Savitzky-Golay卷积平滑(Savitzky-Golay smoothing,SGS)等方法进行预处理,采用竞争性自适应加权算法(Competitive adaptive reweighted sampling,CARS)进行波段选择,剔除NIRS数据中的高频噪声与冗余信息,提升光谱数据质量、建模速度与精度。为验证LSTM模型预测能力,将其与偏最小二乘回归(Partial least squares regression,PSLR)、卷积神经网络(Convolution neural network,CNN)等建模算法对比分析。上述3种算法被分别应用于建立樟子松木材气干密度近红外预测模型。【结果】基于上述3种建模方法建立的NIRS模型均可实现樟子松气干密度的有效预测。且LSTM模型的预测精度与回归拟合度均优于PLSR与CNN模型。其中SGS+CARS处理后的LSTM模型的预测精度最高、泛化性能最强、拟合效果最好(R^(2)=0.959,RMSEP=0.005,RPD=5.033)。【结论】通过对樟子松木材光谱数据与气干密度的采集,建立了一种新型的基于NIRS分析技术与LSTM的木材气干密度检测方法。LSTM预测模型相较于传统的回归模型,模型的预测精度更高,回归效果更好,鲁棒性更强。该检测方法既可保证木材的完整性,又可以提高气干密度的预测精度,实现了对樟子松木材气干密度的快速无损检测,为木材近红外光谱分析提供了可参考的模型与理论依据。

弹力带抗阻运动对高龄老年人工作记忆的影响——来自fNIRS的证据

目的观察长期弹力带运动改善高龄老年人工作记忆的效果,采用近红外光谱技术探究可能的脑机制。方法将60名高龄老年人随机分为弹力带组与对照组,弹力带组接受16周、每周3次、每次40 min的弹力带干预,对照组保持原有生活状态;采集实验前后受试者完成工作记忆任务期间的前额叶血流动力学指标。结果反应时的时间×组别交互效应具有统计学意义(P<0.001);正确率的时间×记忆负荷交互效应具有统计学意义(P=0.003);正确反应率的时间×组别交互效应具有统计学意义(P<0.001);氧合血红蛋白结果显示,9条通道组别×时间×记忆负荷交互效应均具有统计学意义(P<0.002)。结论规律性中低强度弹力带抗阻运动能改善高龄老年人的工作记忆表现;在低记忆负荷时前额叶激活不明显,中、高记忆负荷时双侧腹外侧前额叶、左背外侧前额叶、左额极区显著激活,这种前额叶激活优化模式可能是弹力带抗阻运动改善工作记忆的脑机制。