An odor detection system based on automatically trained mice by relative go no-go olfactory operant conditioning
An odor detection system based on automatically trained mice by relative go no-go olfactory operant conditioning
复制标题
基于相对go-no-go嗅觉操作条件自动训练的小鼠的气味检测系统
DOI：
10.1038/srep10019
复制
发表时间：
2015-05-06
影响因子：
4.6
通讯作者：
Ma, YuanYe
中科院分区：
文献类型：
--
作者：
He, Jing;Wei, JingKuan;Ma, YuanYe
人类社会在各种情况下都需要气味检测应用。 Rodent 在开发生物气味检测系统方面具有独特的优势。本报告概述了一种新颖的装置，旨在训练最多 5 只小鼠使用新的嗅觉、相对走-不走、操作条件反射范式自动检测气味。新范例提供了通过不断变化的反应来测量个体动物检测行为的实时可靠性的机会。所有 15 只缺水小鼠都能够通过触摸传感器学会对以更高的触摸频率传递的不可预测的目标气味做出反应。持续降低目标气味（正丁醇）浓度对小鼠进行训练，在 0.01% 溶液浓度下训练时，平均正确率显着下降；通过训练，报警算法对合格小鼠组的气味检测行为表现出良好的识别能力。然后，针对4个区块的模拟场景对报警算法进行了反复测试。小鼠在测试期间的表现与训练期间相当，在 59 次随机递送中总共针对目标气味发出了 58 次警告，并且 0 次误报。结果表明，这种气味检测方法有望在各种类型的气味检测应用中得到进一步发展。
Odor detection applications are needed by human societies in various circumstances. Rodent offers unique advantages in developing biologic odor detection systems. This report outlines a novel apparatus designed to train maximum 5 mice automatically to detect odors using a new olfactory, relative go no-go, operant conditioning paradigm. The new paradigm offers the chance to measure real-time reliability of individual animal’s detection behavior with changing responses. All of 15 water-deprivation mice were able to learn to respond to unpredictable delivering of the target odor with higher touch frequencies via a touch sensor. The mice were continually trained with decreasing concentrations of the target odor (n-butanol), the average correct percent significantly dropped when training at 0.01% solution concentration; the alarm algorithm showed excellent recognition of odor detection behavior of qualified mice group through training. Then, the alarm algorithm was repeatedly tested against simulated scenario for 4 blocks. The mice acted comparable to the training period during the tests and provided total of 58 warnings for the target odor out of 59 random deliveries and 0 false alarm. The results suggest this odor detection method is promising for further development in respect to various types of odor detection applications.