我们要感谢沃达丰美洲基金会，比尔和梅琳达·盖茨基金会，圣保罗研究基金会（FAPESP）为这项研究提供资金。我们还要感谢从昆虫学系美国加州大学河滨分校的许多教师，他们对这个项目的意见。

1，昆虫群体和饲养<ol><li>蚊子殖民地和饲养<ol><li>背面淡色tarsalis，致倦库蚊，淡色stigmatosoma和埃及伊蚊成虫从实验室的殖民地，它源于野生捕捞的个体。 </li><li>在标准实验室条件下，搪瓷锅背面蚊子幼虫（27℃，16：8小时光照：黑暗[LD]周期1小时黄昏/黎明时期），并喂它们随意在地面啮齿动物饲料混合物，啤酒酵母（3:1​​，体积比）。 </li><li>收集蚊蛹入200ml杯，并将其放置在试验室。另外，吸出成蚊成实验室1周之内出现。确保各实验腔室含有相同的物种/性别的20到40个人。 </li><li>饲料成蚊随意在10％的蔗糖和水的混合物。更换食品周刊。 </li><li>棉花浸湿毛巾每周两次和PL王牌他们的实验室顶到笼子内保持湿度。此外，放置杯子的自来水200毫升在腔室在任何时候，以帮助保持整体的湿度水平。 </li><li>保持试验室上在16：8小时光照：黑暗[LD]周期，20.5-22℃，30-50％RH的条件下的实验的持续时间。 </li></ol></li><li>家蝇和果蝇群体和饲养<ol><li>从实验室的殖民地后方家蝇 ，从野外捕获而得的个体。钓野生果蝇拟黄的个人和后他们在实验室。 </li><li>在塑料桶的标准实验室条件下后方家蝇幼虫（12时12小时光照：黑暗[LD]循环，26℃，40％相对湿度）的水，米糠粕，苜蓿，酵母和奶粉的混合物。在饲养室后部拟黄果蝇幼虫喂它们自由采食腐烂的混合物水果。 </li><li>吸成人家蝇到1周之内出现的实验室。成年果蝇拟黄可直接饲养在实验室。此前的数据收集，确保每个实验室包含不超过10-15个人家蝇或20-30个人果蝇拟黄。 </li><li>喂成年家蝇自由采食对糖和低脂肪奶粉的混合物，自由饮水。饲料成年果蝇拟黄随意腐烂水果的混合物。更换食品周刊。 </li><li>保持试验室在16：8小时光照：黑暗[LD]循环，20.5-22℃，30〜50％RH的实验时间。 </li></ol></li></ol> 2，记录飞行声音实验钱伯斯<ol><li>试验室设置 注：“实验çhamber“指设计在我们的实验室笼，其中，所述数据被记录。该传感器是相当便宜的。当内置大容量，一组最多可以制造不到10美元。 <ol><li>建构一个实验室，无论是大尺寸的67厘米（长）X 22厘米宽x24.75厘米H或更小尺寸：30厘米（长）×20厘米（W）×20厘米H的实验室由一个光电晶体管阵列和中激光线指向光电晶体管阵列。 注意：另外，该腔室包括被修改为包括所述传感器装置以及安装于一块PVC管道，以允许进入昆虫的套筒Kritter拾得的。 </li><li>光电晶体管阵列连接到一个电子板。电子板的输出馈入数字录音机，并记​​录为MP3格式的音频数据。看到图1.II传感器和图1.I腔的物理版本的逻辑设计。 </li><lI&gt;修改的试验室用一片网布贴在里面的盖子，以防止昆虫逃逸。 注：当昆虫飞行穿越激光束，它的翅膀部分遮挡光线，导致小光的波动。光波动是由光电晶体管阵列作为电流变化的捕捉，以及该信号被滤波和放大由设计的电子电路板的定制。 </li></ol></li><li>设置系统由飞行昆虫制作录制声音<ol><li>试验室连接到电源。打开电源。 </li><li>在试验室中，发现激光灯和photoarray。调整激光灯的photoarray。为了实现适当的对准，调整使用的磁铁对实验室的外面相对应的附连到photoarray在腔室内部的磁体，直到激光光的中心在所有的个体photodiod的photoarrayES。 </li><li>执行两个完整性检查，以确保系统的设置是否正确。 注意：第一个步骤是，以确保该系统的电源，所有电线都正确地连接以及激光指向的照片阵列。第二个步骤是进行进一步的检查对激光对准，以确保它能够捕捉昆虫的wingbeats的声音。 <ol><li>将耳机（而非记录器）插入到音频插孔。沉浸手进出试验室，邻近photoarray，打破了激光的平面。确保激光灯亮（这将是光的红色光束），并且你打破光平面几次用你的手。听的变化噪声电平作为你的手进入和退出光束。如果检测可听差，传感器能够捕获由大物体运动产生的声音。如果成功的话，就移动到下一个完整性检查，否则，请检查耳机是否连接正确，无论是激光指着photoarray。调整photoarray相应地，直到手的声音进出实验室可以听到。 </li><li>附上一块薄薄的电线到自动牙刷。打开的牙刷，并沉浸在布线和移出试验室靠近phototarray的。确保激光射到一块布线的，因为它移动。如果检测到的频率的可听变化，当片布线的断裂激光的平面，系统就准备好捕捉由细小的物体， 即 ，昆虫声音的运动所产生的声音。如果没有检测到有声的差异，回到步骤2.2.2重新调整激光灯和photoarray。 </li></ol></li><li>在系统的设置是否正确，盖上盖子，并添加了昆虫。 </li></ol></li><li>数据采集​​：通过飞行昆虫制作录制声音<ol><li>打开录音机，并作出voicË注解，包括以下信息：该物种在实验室的名称，昆虫，日期和时间，当前环境室温和相对湿度的年龄。暂停录制。 </li><li>记录器连接到该系统中，通过音频电缆，并且恢复记录。离开记录器来记录3天，然后停止录制。 </li><li>从记录的数据下载到PC上的一个新文件夹。通过删除数据清空记录。 </li><li>重复上述的记录处理，直到剩余的昆虫死亡关闭并有不大于5的昆虫在笼活着。 </li></ol></li></ol> 3，传感器数据处理和检测声音由飞行昆虫产生的<ol><li>用软件来检测声音由飞行昆虫产生的。 注意：软件（检测算法）比实时更快。整个过程不到3个小时来处理记录会话，DAT的即三天A，A标机采用英特尔（R）酷睿™CPU以2.00 GHz和8 GB的RAM上。 <ol><li>对于包含从记录会话数据的每个文件夹，运行检测软件来检测昆虫的声音。运行软件，打开MATLAB，然后键入“circandian_wbf（DATADIR）”，在命令窗口，DATADIR是记录数据的目录。然后按“Enter”键开始。 注：从引用＃16下载检测软件circadian_wbf。 </li><li>等到算法终止，再检查的检测结果。该算法输出所有检测到的昆虫声音在名为“DATADIR _extf”的新文件夹，其中DATADIR是一样的，在前面的步骤。每个声音文件是一个1秒长的音频剪辑最初提取从原始的记录，用数字滤波器来去除噪声。每个检测到的声音的发生时间将被保存在一个名为“DATADIR _time.mat及文件＃8221 ;.观察图2中的检测到的昆虫声音的例子。 </li></ol></li><li>检测算法<ol><li>用0.1秒长的滑动窗口滑动通过录音。滑动窗口开始从记录的开始。对于每一个窗口，按照下面的步骤。 <ol><li>计算当前窗口的基本频率。 </li><li>如果基波频率为100赫兹到1200赫兹的范围内，则执行下列操作： <ol><li>提取1秒长的音频剪辑的中心从记录当前的窗口;应用数字滤波以去除在该剪辑中的噪声，并保存滤波的音频到文件夹“DATADIR _extf”。 </li><li>保存当前窗口的出现时间到文件“DATADIR _时间”。 </li><li>移动滑动窗口对紧跟在提取的音频剪辑的点。 </li></ol></li><li>如果根本FRequency不是100赫兹到1200赫兹的范围内，只需将滑动窗口0.01秒前进。 </li></ol></li><li>重复该过程，直到滑动窗口到达记录的结尾。 </li></ol></li></ol> 4，昆虫分类<ol><li>贝叶斯分类仅使用飞音 注：贝叶斯分类器是一个概率分类器的目的是其最有可能的阶级划分。 <ol><li>声音特征计算<ol><li>对于每个昆虫的声音，计算使用离散傅立叶变换（DFT）的声音的频谱。截断的频谱只包括对应于频率范围的那些数据点：100赫兹到2000赫兹。截短的频谱，然后在分类为“代表”的昆虫声音使用。 注：该DFT是一种算法，其将时域信号转换到频域。这是一个BUILT-函数在大多数编程库，可以在程序中调用，只需一行代码。 </li></ol></li><li>训练贝叶斯分类器<ol><li>使用k近邻密度估计方法14学习用声音特征的后验概率分布。与邻近法，训练阶段是建立一个训练集。 <ol><li>随机抽取的一些昆虫的声音从收集各种昆虫的数据。 </li><li>按照第4.1.1节中的步骤和计算截断频谱每个采样声音。被截断的频谱连同样本的类标签（昆虫名）组成的训练集。 </li></ol></li></ol></li><li>使用贝叶斯分类到一个未知的昆虫分类<ol><li>计算的未知昆虫声音的截断频率的频谱。 </li><li>计算未知的截光谱之间的欧几里德距离n对象，并在训练数据集中的所有截断的频谱。 </li><li>找到相关的前k（K = 8在本文中）中的训练数据集的未知物体的最近邻居。计算属于一类作为其标记为类中的前k近邻的部分未知的昆虫声音的后验概率。 </li><li>分类的未知物体以具有最高的后验概率的类。 </li></ol></li></ol></li><li>添加一个功能，该分类：昆虫昼夜节律<ol><li>学习的昆虫声音的发生时间，即，昼夜节律对每个物种的昆虫类制冷分布。 <ol><li>取得从检测结果中的每个声音（CF 3.2节）的发生时间。 </li><li>对于每一个物种，打造的昆虫声音发生时间的直方图。 </li><li>归一化的直方图，使得直方图的面积为1。正常美化版直方图是在给定物种的昼夜节律。它讲述了一个特定的时间周期内观察昆虫，该物种的，在飞行的概率。 </li></ol></li><li>通过结合“昆虫声音”与昼夜节律分类未知的“昆虫声音” <ol><li>给定时刻，其中未知昆虫的声音发生的特定点，获得观测类的基于类的昼夜节律昆虫的概率。 注：昼夜节律是一个概率分布。这是一个数组，指定检测健全，昆虫的特定物种所产生的概率，在一天中的特定时间。因此，一旦时间被定，可以简单地检查阵列得到的概率。 </li><li>按照第4.1.2节中的步骤来计算使用的声音的功能，对未知的声音属于类的后验概率。相乘后验概率的结果来回米上一步中得到新的后验概率。 </li><li>分类“未知的声音”，以拥有最高新的后验概率的类。 </li></ol></li></ol></li><li>增加一个功能的分类：昆虫地理分布<ol><li>了解该物种的地理分布的利益，无论是从公布的历史数据记录，相关文献，或者干脆从现场技术人员/生物学家收集第一手资料。出于演示的目的，使用模拟的图形分布， 如图7。 </li><li>分类“未知的昆虫声音”用“飞声”和两个附加功能。 <ol><li>定所述地理位置，其中昆虫声音被截获，计算使用物种的图形分布在观察从类昆虫在该特定位置的概率。 <ol><li>按照第4.2.2步骤次计算采用的声音特点和生理节奏的“未知的声音”属于类的后验概率。相乘的这个步骤，以从先前的步骤的结果的结果，以获得新的后验概率。 </li></ol></li><li>分类“未知的声音”，以拥有最高新的后验概率的类。 </li></ol></li></ol></li><li>总体框架中添加功能<ol><li>请考虑使用刚刚的声音特点作为主要分类贝叶斯分类器。按照下面的步骤来添加新的功能分类。 <ol><li>在训练阶段，学习新功能的类空调的密度函数。 </li><li>在分类阶段，考虑到“未知的声音”的新功能，通过计算在上一步中了解到的密度函数观测要素类的概率。乘新概率公关对应于“未知声音”属于类基于刚刚奇数的特点，分别计算evious后验概率，以获得新的后验概率。分类未知对象具有最高的新的后验概率的类。 </li></ol></li></ol></li></ol>

<ol>
	<li>Kahn, M. C., Celestin, W., Offenhauser, W. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Recording+of+sounds+produced+by+certain+disease-carrying+mosquitoes.">Recording of sounds produced by certain disease-carrying mosquitoes.</a> Science. 101 (2622), 335-336 (1945).</li><li>Reed, S. C., Williams, C. M., Chadwick, L. E. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Frequency+of+wing-beat+as+a+character+for+separating+species+races+and+geographic+varieties+of+Drosophila.">Frequency of wing-beat as a character for separating species races and geographic varieties of Drosophila.</a> Genetics. 27 (3), 349 (1942).</li><li>Belton, P., Costello, R. A. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Flight+sounds+of+the+females+of+some+mosquitoes+of+Western+Canada.">Flight sounds of the females of some mosquitoes of Western Canada.</a> Entomologia experimentalis et applicata. 26 (1), 105-114 (1979).</li><li>Mankin, R. W., Machan, R., Field Jones, R. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=testing+of+a+prototype+acoustic+device+for+detection+of+Mediterranean+fruit+flies+flying+into+a+trap.">testing of a prototype acoustic device for detection of Mediterranean fruit flies flying into a trap.</a> Proc. 7th Int Symp Fruit Flies of Economic Importance. , 10-15 (2006).</li><li>Raman, D. R., Gerhardt, R. R., Wilkerson, J. B. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Detecting+insect+flight+sounds+in+the+field+Implications+for+acoustical+counting+of+mosquitoes.">Detecting insect flight sounds in the field Implications for acoustical counting of mosquitoes.</a> Transactions of the ASABE. 50 (4), 1481 .</li><li>Moore, A., Miller, J. R., Tabashnik, B. E., Gage, S. H. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Automated+identification+of+flying+insects+by+analysis+of+wingbeat+frequencies.">Automated identification of flying insects by analysis of wingbeat frequencies.</a> Journal of economic entomology. 79 (6), 1703-1706 (1986).</li><li>Moore, A., Miller, R. H. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Automated+identification+of+optically+sensed+aphid+(Homoptera:+Aphidae)+wingbeat+waveforms.">Automated identification of optically sensed aphid (Homoptera: Aphidae) wingbeat waveforms.</a> Annals of the Entomological Society of America. 95 (1), 1-8 (2002).</li><li>Li, Z., Zhou, Z., Shen, Z., Yao, Q. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Automated+identification+of+mosquito+(diptera:+Culicidae)+wingbeat+waveform+by+artificial+neural+network.">Automated identification of mosquito (diptera: Culicidae) wingbeat waveform by artificial neural network.</a> Artificial Intelligence Applications and Innovations. , 483-489 (2005).</li><li>Moore, A. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Artificial+neural+network+trained+to+identify+mosquitoes+in+flight.">Artificial neural network trained to identify mosquitoes in flight.</a> Journal of insect behavior. 4 (3), 391-396 (1991).</li><li>Halevy, A., Norvig, P., Pereira, F. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+unreasonable+effectiveness+of+data.">The unreasonable effectiveness of data.</a> Intelligent Systems, IEEE. 24 (2), 8-12 (2009).</li><li>Banko, M., Brill, E. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Mitigating+the+paucity-of-data+problem+exploring+the+effect+of+training+corpus+size+on+classifier+performance+for+natural+language+processing.">Mitigating the paucity-of-data problem exploring the effect of training corpus size on classifier performance for natural language processing.</a> , 1-5 (2001).</li><li>Shotton, J., et al. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Real-time+human+pose+recognition+in+parts+from+single+depth+images.">Real-time human pose recognition in parts from single depth images.</a> Communications of the ACM. 56 (1), 116-124 (2013).</li><li>Ilyas, P. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Classifying+insects+on+the+fly.">Classifying insects on the fly.</a> Ecological Informatics. , (2013).</li><li>Mack, Y. P., Rosenblatt, M. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Multivariate+k-nearest+neighbor+density+estimates.">Multivariate k-nearest neighbor density estimates.</a> Journal of Multivariate Analysis. 9 (1), 1-15 (1979).</li><li>Benedict, M., Robinson, A. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=The+first+releases+of+transgenic+mosquitoes+an+argument+for+the+sterile+insect+technique.">The first releases of transgenic mosquitoes an argument for the sterile insect technique.</a> TRENDS in Parasitology. 19 (8), 349-355 (2003).</li><li>. Supporting Materials Available from: <a target="_blank" href="https://sites.google.com/site/insectclassification">https://sites.google.com/site/insectclassification</a></li><li>Chen, Y., Why, A., Batista, G., Mafra-Neto, A., Keogh, E. <a target="_blank" href="http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed&cmd=Search&doptcmdl=Citation&defaultField=Title+Word&term=Flying+Insect+Classification+with+Inexpensive+Sensors.">Flying Insect Classification with Inexpensive Sensors.</a> Journal of Insect. 27 (5), 657-677 (2014).</li></ol>

作者宣称，他们没有竞争的经济利益。

这里描述的传感器/分类框架允许飞虫的廉价和可扩展的分类。该精度达到由系统不够好，让商业产品的开发，并可能在昆虫学研究的有用工具。 使用廉价的，非侵入性的传感器能​​够准确，自动分类飞虫的能力将会对昆虫学研究显著影响。例如，通过部署在系统中的字段进行计数和分类，昆虫媒介，该系统可提供的目标物种的实时计数，产生可用于计划介入/抑制方案来防...

利用其飞行附带声音自动分类的昆虫的想法可以追溯到计算机和初期市售录音设备1。然而，很少取得了进展在这个问题上，在中间的几十年。进展的这种追求，缺乏可归结为几个相关的因素。 第一，缺乏有效的传感器取得数据收集困难。大多数收集数据的努力已经用声麦克风2-5。这样的装置是极其敏感的风噪声和环境噪声的环境中，从而导致非常稀疏的，低质量的数据。 第二，配合这些数据质量问题的事实是，许多研究者试图学习非常复杂的分类模型，尤其是神经网络6-8。试图了解复杂的分类模型，用仅有的几十例，是秘方过拟合。 第三，获得的数据的困难意味着许多研究者试图建立分类模型具有非常有限的数据，少至300实例9或更小。然而，已知的是用于构建准确的分类模式，更多的数据是更好10-13。 这项工作涉及的所有三个问题。光学（而不是声 ）传感器可用于录制的“声音”从米昆虫的飞行距离，拥有完整的不变性风噪声和周围的声音。这些传感器都允许的百万标记的训练实例，比以前的所有努力结合更多的数据的记录，并因此有助于避免过度拟合一直困扰以前的研究工作。如下所示是有原则的方法允许更多的信息到分类模型的结合。这种额外的信息可以作为司空见惯，并为易于获得的时间 - 日期的，但仍然产生显著收益模型的准确性。最后，它被证明数据，我们收集了巨额允许我们以“ 数据的不合理有效性 ”10优势，以产生简单，准确和鲁棒的分类器。 综上所述，飞行昆虫的分类已经超越了研究实验室创建的可疑索赔，现在准备实际部署。在这项工作中提出的传感器和软件将提供全球研究人员强大的工具来加速他们的研究。

<table border="0" cellpadding="0" cellspacing="0" width="1293">
	<tbody>
		<tr height="21">
			<td height="21" style="height:21px;width:328px;">Name of Material/ Equipment</td>
			<td style="width:237px;">Company</td>
			<td style="width:257px;">Catalog Number</td>
			<td style="width:471px;">Comments/Description</td>
		</tr>
		<tr height="21">
			<td height="21" style="height:21px;width:328px;">Audio Recorder: ICD-PX312</td>
			<td style="width:237px;">Sony</td>
			<td style="width:257px;">4-267-065-11(2)</td>
			<td>With a 8 GB microSD extra memory</td>
		</tr>
		<tr height="21">
			<td height="21" style="height:21px;width:328px;">Insectary</td>
			<td style="width:237px;">Lee&#39;s Aquarium &#38; Pet Products.</td>
			<td style="width:257px;">20088 HerpHaven&#174;, Large Rectangle</td>
			<td style="width:471px;">14 1/2&#34; Long x 8 3/4&#34; Wide x 9 3/4&#34; high. Modified to house insects.&#160;</td>
		</tr>
		<tr height="42">
			<td height="42" style="height:42px;width:328px;">Laser Line Generator, 650nm (red)</td>
			<td style="width:237px;">Apinex (www.apinex.com)</td>
			<td style="width:257px;">LN60-650</td>
			<td style="width:471px;">5mW. This is a low powered laser, similar to a teachers lasers pointer</td>
		</tr>
		<tr height="63">
			<td height="63" style="height:63px;width:328px;">Photodiode Array</td>
			<td style="width:237px;">VISHAY SEMICONDUCTOR&#160; TEFD4300&#160; PIN PHOTODIODE, 650NM, 20DEG, T1</td>
			<td style="width:257px;">TEFD4300</td>
			<td style="width:471px;">We made a custom array of 15 of these Photodiodes wired in parallel</td>
		</tr>
		<tr height="63">
			<td height="63" style="height:63px;">Analogue to Digital Convertor Integrated Circuit&#160;</td>
			<td style="width:237px;">Custom made in our lab</td>
			<td style="width:257px;"></td>
			<td style="width:471px;">We made this item ourselves, but an easily available commercial product, Gino PCF8591 AD/DA Converter, provides the same functionality.&#160;</td>
		</tr>
	</tbody>
</table>

flying insect detection and classification with inexpensive sensors

一种廉价的，非侵入性的系统，可以准确分类飞虫会对昆虫学研究的重要意义，并允许用于医疗和农业昆虫学的载体和病虫害防治许多有用的应用程序的开发。鉴于此，在过去60年看过致力于这项工作的许多研究工作。迄今为止，但是，没有这种研究产生了深远的影响。在这项工作中，我们表明，伪声光传感器能产生优良的数据;该附加的特征，这两个内在的和外在的昆虫的飞行特性，可以被利用来改善昆虫分类;一个贝叶斯分类方法允许有效地学习分类模型是非常强劲的过度拟合，和一般的分类框架，允许轻松地将功能任意数字。我们证明了这一切相形见绌以前的作品COMBI大规模的实验结果奈德，如由昆虫的数目，并考虑物种的数量来衡量。

两个实验都在这里呈现。对于这两个实验中，所使用的数据，随机从包含超过10万个对象的数据集进行采样。 第一个实验表明所提出的分类准确分类，昆虫的不同物种/性别的能力。作为分类精度依赖于昆虫进行分类，分类精度单个绝对值不会让读者对系统的性能有良好的直觉。代替，而不是报告于一组固定的昆虫的分类器的精确度，该分类器被应用到数据集的递增增加数种?...

Watch this Scientific Journal Video about Flying Insect Detection and Classification with Inexpensive Sensors at JoVE.com

Flying Insect Detection and Classification with Inexpensive Sensors

我们提出了使用廉价的，非侵入性的伪声光感应器可以自动，准确地检测，计数和分类飞基于他们的飞行昆虫的声音系统。

飞行昆虫检测与分类用便宜的传感器

An inexpensive, noninvasive system that could accurately classify flying insects would have important implications for entomological research, and allow ...

flying-insect-detection-and-classification-with-inexpensive-sensors

Research

JoVE Journal

Bioengineering

25.0K 次观看.  University of California, Riverside. 我们提出了使用廉价的，非侵入性的伪声光感应器可以自动，准确地检测，计数和分类飞基于他们的飞行昆虫的声音系统。 

视频: Flying Insect Detection and Classification with Inexpensive Sensors

Bacterial Detection &amp; Identification Using Electrochemical Sensors

Electrochemical sensors are widely used for rapid and accurate measurement of blood glucose and can be adapted for detection of a wide variety of analytes. Electrochemical sensors operate by transducing a biological recognition event into a useful electrical signal. Signal transduction occurs by coupling the activity of a redox enzyme to an amperometric electrode. Sensor specificity is either an inherent characteristic of the enzyme, glucose oxidase in the case of a glucose sensor, or a product of linkage between the enzyme and an antibody or probe.
Here, we describe an electrochemical sensor assay method to directly detect and identify bacteria. In every case, the probes described here are DNA oligonucleotides. This method is based on sandwich hybridization of capture and detector probes with target ribosomal RNA (rRNA). The capture probe is anchored to the sensor surface, while the detector probe is linked to horseradish peroxidase (HRP). When a substrate such as 3,3',5,5'-tetramethylbenzidine (TMB) is added to an electrode with capture-target-detector complexes bound to its surface, the substrate is oxidized by HRP and reduced by the working electrode. This redox cycle results in shuttling of electrons by the substrate from the electrode to HRP, producing current flow in the electrode.

We describe an electrochemical sensor assay method for rapid bacterial detection and identification. The assay involves a sensor array functionalized with DNA oligonucleotide capture probes for ribosomal RNA (rRNA) species-specific sequences. Sandwich hybridization of target rRNA with the capture probe and a horseradish peroxidase-linked DNA oligonucleotide detector probe produces a measurable amperometric current.

使用电化学传感器的细菌检测和鉴定

Electrochemical  sensors are widely used for rapid and accurate measurement of blood glucose and  can be adapted for detection of a wide variety of ...

科研

生物工程

Glutamine Flux Imaging Using Genetically Encoded Sensors

Genetically encoded sensors allow real-time monitoring of biological molecules at a subcellular resolution. A tremendous variety of such sensors for biological molecules became available in the past 15 years, some of which became indispensable tools that are used routinely in many laboratories.
One of the exciting applications of genetically encoded sensors is the use of these sensors in investigating cellular transport processes. Properties of transporters such as kinetics and substrate specificities can be investigated at a cellular level, providing possibilities for cell-type specific analyses of transport activities. In this article, we will demonstrate how transporter dynamics can be observed using genetically encoded glutamine sensor as an example. Experimental design, technical details of the experimental settings, and considerations for post-experimental analyses will be discussed.

This article will demonstrate how to monitor glutamine dynamics in live cells using FRET. Genetically encoded sensors allow real-time monitoring of biological molecules at a subcellular resolution. Experimental design, technical details of the experimental settings, and considerations for post-experimental analyses will be discussed for genetically encoded glutamine sensors.

谷氨酰胺通量成像使用基因编码的传感器

Genetically encoded sensors allow real-time monitoring of biological molecules at a subcellular resolution. A tremendous variety of such sensors for ...

A Microfluidic-based Electrochemical Biochip for Label-free DNA Hybridization Analysis

Miniaturization of analytical benchtop procedures into the micro-scale provides significant advantages in regards to reaction time, cost, and integration of pre-processing steps. Utilizing these devices towards the analysis of DNA hybridization events is important because it offers a technology for real time assessment of biomarkers at the point-of-care for various diseases. However, when the device footprint decreases the dominance of various physical phenomena increases. These phenomena influence the fabrication precision and operation reliability of the device. Therefore, there is a great need to accurately fabricate and operate these devices in a reproducible manner in order to improve the overall performance. Here, we describe the protocols and the methods used for the fabrication and the operation of a microfluidic-based electrochemical biochip for accurate analysis of DNA hybridization events. The biochip is composed of two parts: a microfluidic chip with three parallel micro-channels made of polydimethylsiloxane (PDMS), and a 3 x 3 arrayed electrochemical micro-chip. The DNA hybridization events are detected using electrochemical impedance spectroscopy (EIS) analysis. The EIS analysis enables monitoring variations of the properties of the electrochemical system that are dominant at these length scales. With the ability to monitor changes of both charge transfer and diffusional resistance with the biosensor, we demonstrate the selectivity to complementary ssDNA targets, a calculated detection limit of 3.8 nM, and a 13% cross-reactivity with other non-complementary ssDNA following 20 min&#160;of incubation. This methodology can improve the performance of miniaturized devices by elucidating on the behavior of diffusion at the micro-scale regime and by enabling the study of DNA hybridization events.

We present a microfluidic-based electrochemical biochip for DNA hybridization detection. Following ssDNA probe functionalization, the specificity, sensitivity, and detection limit are studied with complementary and non-complementary ssDNA targets. Results illustrate the influence of the DNA hybridization events on the electrochemical system, with a detection limit of 3.8 nM.

微流体为基础的电化学生物芯片的无标记的DNA杂交分析

Miniaturization of analytical benchtop procedures into the micro-scale provides significant advantages in regards to reaction time, cost, and integration ...

Sensitive Detection of Proteopathic Seeding Activity with FRET Flow Cytometry

Increasing evidence supports transcellular propagation of toxic protein aggregates, or proteopathic seeds, as a mechanism for the initiation and progression of pathology in several neurodegenerative diseases, including Alzheimer's disease and the related tauopathies. The potentially critical role of tau seeds in disease progression strongly supports the need for a sensitive assay that readily detects seeding activity in biological samples. 
By combining the specificity of fluorescence resonance energy transfer (FRET), the sensitivity of flow cytometry, and the stability of a monoclonal cell line, an ultra-sensitive seeding assay has been engineered and is compatible with seed detection from recombinant or biological samples, including human and mouse brain homogenates. The assay employs monoclonal HEK 293T cells that stably express the aggregation-prone repeat domain (RD) of tau harboring the disease-associated P301S mutation fused to either CFP or YFP, which produce a FRET signal upon protein aggregation. The uptake of proteopathic tau seeds (but not other proteins) into the biosensor cells stimulates aggregation of RD-CFP and RD-YFP, and flow cytometry sensitively and quantitatively monitors this aggregation-induced FRET. The assay detects femtomolar concentrations (monomer equivalent) of recombinant tau seeds, has a dynamic range spanning three orders of magnitude, and is compatible with brain homogenates from tauopathy transgenic mice and human tauopathy subjects. With slight modifications, the assay can also detect seeding activity of other proteopathic seeds, such as &#945;-synuclein, and is also compatible with primary neuronal cultures. The ease, sensitivity, and broad applicability of FRET flow cytometry makes it useful to study a wide range of protein aggregation disorders.

Cell-to-cell transfer of protein aggregates, or proteopathic seeds, may underlie the progression of pathology in neurodegenerative diseases. Here, a novel FRET flow cytometry assay is described that enables specific and sensitive detection of seeding activity from recombinant or biological samples.

与FRET流式细胞仪灵敏地检测Proteopathic播种活动

Increasing evidence supports transcellular propagation of toxic protein aggregates, or proteopathic seeds, as a mechanism for the initiation and ...

Fast Enzymatic Processing of Proteins for MS Detection with a Flow-through Microreactor

The vast majority of mass spectrometry (MS)-based protein analysis methods involve an enzymatic digestion step prior to detection, typically with trypsin. This step is necessary for the generation of small molecular weight peptides, generally with MW &#60; 3,000-4,000 Da, that fall within the effective scan range of mass spectrometry instrumentation. Conventional protocols involve O/N enzymatic digestion at 37 &#186;C. Recent advances have led to the development of a variety of strategies, typically involving the use of a microreactor with immobilized enzymes or of a range of complementary physical processes that reduce the time necessary for proteolytic digestion to a few minutes (e.g., microwave or high-pressure). In this work, we describe a simple and cost-effective approach that can be implemented in any laboratory for achieving fast enzymatic digestion of a protein. The protein (or protein mixture) is adsorbed on C18-bonded reversed-phase high performance liquid chromatography (HPLC) silica particles preloaded in a capillary column, and trypsin in aqueous buffer is infused over the particles for a short period of time. To enable on-line MS detection, the tryptic peptides are eluted with a solvent system with increased organic content directly in the MS ion source. This approach avoids the use of high-priced immobilized enzyme particles and does not necessitate any aid for completing the process. Protein digestion and complete sample analysis can be accomplished in less than ~3 min and ~30 min, respectively.

A quick protocol for proteolytic digestion with an in-house built flow-through tryptic microreactor coupled to an electrospray ionization (ESI) mass spectrometer is presented. The fabrication of the microreactor, the experimental setup and the data acquisition process are described.

蛋白质的MS检测快速酶法处理与流通型微反应器

The vast majority of mass spectrometry (MS)-based protein analysis methods involve an enzymatic digestion step prior to detection, typically with trypsin. ...

Microfluidic Platform with Multiplexed Electronic Detection for Spatial Tracking of Particles

Microfluidic processing of biological samples typically involves differential manipulations of suspended particles under various force fields in order to spatially fractionate the sample based on a biological property of interest. For the resultant spatial distribution to be used as the assay readout, microfluidic devices are often subjected to microscopic analysis requiring complex instrumentation with higher cost and reduced portability. To address this limitation, we have developed an integrated electronic sensing technology for multiplexed detection of particles at different locations on a microfluidic chip. Our technology, called Microfluidic CODES, combines Resistive Pulse Sensing with Code Division Multiple Access to compress 2D spatial information into a 1D electrical signal. In this paper, we present a practical demonstration of the Microfluidic CODES technology to detect and size cultured cancer cells distributed over multiple microfluidic channels. As validated by the high-speed microscopy, our technology can accurately analyze dense cell populations all electronically without the need for an external instrument. As such, the Microfluidic CODES can potentially enable low-cost integrated lab-on-a-chip devices that are well suited for the point-of-care testing of biological samples.

We demonstrate a microfluidic platform with an integrated surface electrode network that combines resistive pulse sensing (RPS) with code division multiple access (CDMA), to multiplex detection and sizing of particles in multiple microfluidic channels.

与复用电子检测的粒子的空间跟踪微流体平台

Microfluidic processing of biological samples typically involves differential manipulations of suspended particles under various force fields in order to ...

Ovarian Cancer Detection Using Photoacoustic Flow Cytometry

Many studies suggest that the enumeration of circulating tumor cells (CTCs) may show promise as a prognostic tool for ovarian cancer. Current strategies for the detection of CTCs include flow cytometry, microfluidic devices, and real-time polymerase chain reaction (RT-PCR). Despite recent advances, methods for the detection of early ovarian cancer metastasis still lack the sensitivity and specificity required for clinical translation. Here, a novel method is presented for the detection of ovarian circulating tumor cells by photoacoustic flow cytometry (PAFC) utilizing a custom three dimensional (3D) printed system, including a flow chamber and syringe pump. This method utilizes folic acid-capped copper sulfide nanoparticles (FA-CuS NPs) to target SKOV-3 ovarian cancer cells by PAFC. This work demonstrates the affinity of these contrast agents for ovarian cancer cells. The results show NP characterization, PAFC detection, and NP uptake by fluorescence microscopy, thus demonstrating the potential of this novel system to detect ovarian CTCs at physiologically relevant concentrations.

A protocol is presented to detect circulating ovarian tumor cells utilizing a custom-made photoacoustic flow system and targeted folic acid-capped copper sulfide nanoparticles.

使用光声流细胞测定卵巢癌检测

Many studies suggest that the enumeration of circulating tumor cells (CTCs) may show promise as a prognostic tool for ovarian cancer. Current strategies ...

Design and Analysis for Fall Detection System Simplification

This paper presents a methodology based on multimodal sensors to configure a simple, comfortable and fast fall detection and human activity recognition system that can be easily implemented and adopted. The methodology is based on the configuration of specific types of sensors, machine-learning methods and procedures. The protocol is divided into four phases: (1) database creation (2) data analysis (3) system simplification and (4) evaluation. Using this methodology, we created a multimodal database for fall detection and human activity recognition, namely UP-Fall Detection. It comprises data samples from 17 subjects that perform 5 types of falls and 6 different simple activities, during 3 trials. All information was gathered using 5 wearable sensors (tri-axis accelerometer, gyroscope and light intensity), 1 electroencephalograph helmet, 6 infrared sensors as ambient sensors, and 2 cameras in lateral and front viewpoints. The proposed novel methodology adds some important stages to perform a deep analysis of the following design issues in order to simplify a fall detection system: a) select which sensors or combination of sensors are to be used in a simple fall detection system, b) determine the best placement of the sources of information, and c) select the most suitable machine learning classification method for fall and human activity detection and recognition. Even though some multimodal approaches reported in literature only focus on one or two of the above-mentioned issues, our methodology allows simultaneously solving these three design problems related to a human fall and activity detection and recognition system.

We present a methodology based on multimodal sensors to configure a simple, comfortable and fast fall detection and human activity recognition system. The goal is to build a system for accurate fall detection that can be easily implemented and adopted.

坠落检测系统简化的设计与分析

This paper presents a methodology based on multimodal sensors to configure a simple, comfortable and fast fall detection and human activity recognition ...

The Fabrication and Operation of a Continuous Flow, Micro-Electroporation System with Permeabilization Detection

Current therapeutic innovations, such as CAR-T cell therapy, are heavily reliant on viral-mediated gene delivery. Although efficient, this technique is accompanied by high manufacturing costs, which has brought about an interest in using alternative methods for gene delivery. Electroporation is an electro-physical, non-viral approach for the intracellular delivery of genes and other exogenous materials. Upon the application of an electric field, the cell membrane temporarily allows molecular delivery into the cell. Typically, electroporation is performed on the macroscale to process large numbers of cells. However, this approach requires extensive empirical protocol development, which is costly when working with primary and difficult-to-transfect cell types. Lengthy protocol development, coupled with the requirement of large voltages to achieve sufficient electric-field strengths to permeabilize the cells, has led to the development of micro-scale electroporation devices. These micro-electroporation devices are manufactured using common microfabrication techniques and allow for greater experimental control with the potential to maintain high throughput capabilities. This work builds off a microfluidic-electroporation technology capable of detecting the level of cell membrane permeabilization at a single-cell level under continuous flow. However, this technology was limited to 4 cells processed per second, and thus a new approach for increasing the system throughput is proposed and presented here. This new technique, denoted as cell-population-based feedback control, considers the cell permeabilization response to a variety of electroporation pulsing conditions and determines the best-suited electroporation pulse conditions for the cell type under test. A higher-throughput mode is then used, where this 'optimal' pulse is applied to the cell suspension in transit. The steps for fabricating the device, setting up and running the microfluidic experiments, and analyzing the results are presented in detail. Finally, this micro-electroporation technology is demonstrated by delivering a DNA plasmid encoding for green fluorescent protein (GFP) into HEK293 cells.

This protocol describes the microfabrication techniques required to build a lab-on-a-chip, microfluidic electroporation device. The experimental setup performs controlled, single-cell-level transfections in a continuous flow and can be extended to higher throughputs with population-based control. An analysis is provided showcasing the ability to electrically monitor the degree of cell membrane permeabilization in real-time.

具有透化检测功能的连续流动微电穿孔系统的制造和操作

Current therapeutic innovations, such as CAR-T cell therapy, are heavily reliant on viral-mediated gene delivery. Although efficient, this technique is ...

用于锥虫病检测和分类的 AI 程序

Superior Auto-Identification of Trypanosome Parasites by Using a Hybrid Deep-Learning Model

Trypanosomiasis is a significant public health problem in several regions across the world, including South Asia and Southeast Asia. The identification of hotspot areas under active surveillance is a fundamental procedure for controlling disease transmission. Microscopic examination is a commonly used diagnostic method. It is, nevertheless, primarily reliant on skilled and experienced personnel. To address this issue, an artificial intelligence (AI) program was introduced that makes use of a hybrid deep learning technique of object identification and object classification neural network backbones on the in-house low-code AI platform (CiRA CORE). The program can identify and classify the protozoan trypanosome species, namely Trypanosoma cruzi, T. brucei, and T. evansi, from oil-immersion microscopic images. The AI program utilizes pattern recognition to observe and analyze multiple protozoa within a single blood sample and highlights the nucleus and kinetoplast of each parasite as specific characteristic features using an attention map.
To assess the AI program&#39;s performance, two unique modules are created that provide a variety of statistical measures such as accuracy, recall, specificity, precision, F1 score, misclassification rate, receiver operating characteristics (ROC) curves, and precision versus recall (PR) curves. The assessment findings show that the AI algorithm is effective at identifying and categorizing parasites. By delivering a speedy, automated, and accurate screening tool, this technology has the potential to transform disease surveillance and control. It could also assist local officials in making more informed decisions on disease transmission-blocking strategies.

作者聚焦：从显微图像中检测 AI 驱动的锥虫物种 

Worldwide medical blood parasites were automatically screened using simple steps on a low-code AI platform. The prospective diagnosis of blood films was improved by using an object detection and classification method in a hybrid deep learning model. The collaboration of active monitoring and well-trained models helps to identify hotspots of trypanosome transmission.

使用混合深度学习模型对锥虫寄生虫进行卓越的自动识别

Trypanosomiasis is a significant public health problem in several regions across the world, including South Asia and Southeast Asia. The identification of ...