-
公开(公告)号:US20200075134A1
公开(公告)日:2020-03-05
申请号:US16463867
申请日:2017-11-21
发明人: Kota Shiba , Ryo Tamura , Gaku Imamura , Genki Yoshikawa
摘要: The present invention provides a method and a device for estimating a value to be estimated associated with a specimen, by performing machine learning of a relationship between a value of an estimation object and an output corresponding thereto, based on an output from a chemical sensor with regard to a plurality of specimens for which specific values to be estimated are known, and using the result of the mechanical learning to estimate a specific value to be estimated on the basis of an output from the chemical sensor with regard to a given unknown specimen.
-
公开(公告)号:US11442051B2
公开(公告)日:2022-09-13
申请号:US16488866
申请日:2018-02-16
发明人: Gaku Imamura , Genki Yoshikawa , Takashi Washio
摘要: Provided is a novel analysis method which, when a chemical sensor is used to perform a measurement, makes it possible to identify a sample without controlling or monitoring a change in the time the sample is introduced. According to the present invention, a sample can be identified without knowing a change in the time the sample is introduced, by using a chemical sensor having a plurality of channels each having different characteristics to perform a measurement, and performing an analysis on the basis of responses obtained from each channel.
-
公开(公告)号:US12007405B2
公开(公告)日:2024-06-11
申请号:US17055687
申请日:2019-06-04
发明人: Kosuke Minami , Genki Yoshikawa , Gaku Imamura , Kota Shiba
摘要: Provided is a novel material analysis technique using a chemical sensor. By reversing the conventional approach, a material to be measured is provided as a receptor of the chemical sensor, and a response signal of the chemical sensor obtained by supplying a known gas or the like to the chemical sensor is obtained. From the response signal, it is possible to identify and distinguish the receptor material, and to obtain its composition and the like. By analyzing the response signal by means of a statistical or machine learning technique such as principal component analysis, linear discriminant analysis, or a support vector machine, the above-mentioned identification and the like can be performed with high accuracy.
-
4.
公开(公告)号:US11774346B2
公开(公告)日:2023-10-03
申请号:US17057747
申请日:2019-06-04
发明人: Kosuke Minami , Gaku Imamura , Kota Shiba , Genki Yoshikawa
摘要: In an analysis of a fluid component using a nanomechanical sensor covered with a receptor, the same receptor is caused to express different response characteristics. In a measuring system of analyzing a response when a sample gas and a purge gas are supplied to a nanomechanical sensor while switching the sample gas and the purge gas, a gas (external gas) different from both gases is mixed into a gas channel and supplied to the sensor for measurement. Since a response characteristic of a receptor is modulated by mixing of the external gas, the object described above is achieved.
-
公开(公告)号:US12100488B2
公开(公告)日:2024-09-24
申请号:US16463867
申请日:2017-11-21
发明人: Kota Shiba , Ryo Tamura , Gaku Imamura , Genki Yoshikawa
IPC分类号: G16C20/70 , G01N1/00 , G01N1/10 , G01N19/00 , G01N29/02 , G06N20/00 , G16C20/20 , G16C20/50 , G16C20/80 , G16C60/00
CPC分类号: G16C20/70 , G01N1/00 , G01N19/00 , G01N29/022 , G06N20/00 , G16C20/20 , G16C20/50 , G16C20/80 , G01N2001/1087 , G01N2001/1093 , G16C60/00
摘要: The present invention provides a method and a device for estimating a value to be estimated associated with a specimen, by performing machine learning of a relationship between a value of an estimation object and an output corresponding thereto, based on an output from a chemical sensor with regard to a plurality of specimens for which specific values to be estimated are known, and using the result of the mechanical learning to estimate a specific value to be estimated on the basis of an output from the chemical sensor with regard to a given unknown specimen.
-
公开(公告)号:US11353437B2
公开(公告)日:2022-06-07
申请号:US16463585
申请日:2017-11-22
发明人: Gaku Imamura , Genki Yoshikawa , Takashi Washio
摘要: Provided is a novel analysis method that enables identification of a sample even when any sample is introduced during measurement carried out by using a chemical sensor. An input in which the amount of an unknown sample changes over time is provided to the chemical sensor, a response which is from the chemical sensor and which changes over time is measured, a sensor function (transmission function) of the chemical sensor with respect to the unknown sample is calculated on the basis of the input and the response, and the unknown sample is identified on the basis of the sensor function of the chemical sensor with respect to the unknown sample.
-
公开(公告)号:US12130270B2
公开(公告)日:2024-10-29
申请号:US17619024
申请日:2020-05-08
发明人: Gaku Imamura , Kosuke Minami , Kota Shiba , Genki Yoshikawa
IPC分类号: G01N33/00 , C08F222/06 , G01L1/18 , G01N5/02 , G01N29/02
CPC分类号: G01N33/0054 , C08F222/06 , G01L1/18 , G01N5/02 , G01N29/022 , G01N2291/0256
摘要: An object of the present invention is to detect ammonia with high sensitivity and high selectivity using a nanomechanical sensor with a structure that is as simple as possible. A method for detecting ammonia according to an embodiment of the present invention comprises supplying a sample gas possibly containing ammonia to a nanomechanical sensor that detects a stress or a displacement using poly(methyl vinyl ether-alt-maleic anhydride) as a material of a receptor layer, and detecting presence or absence of ammonia or a content of ammonia in the sample gas based on an output signal from the nanomechanical sensor, in which the sample gas is a humidified sample gas with controlled relative humidity.
-
公开(公告)号:US10564082B2
公开(公告)日:2020-02-18
申请号:US15543298
申请日:2015-08-31
发明人: Kota Shiba , Genki Yoshikawa , Yusuke Yamauchi , Norihiro Suzuki , Gaku Imamura , Kosuke Minami , Hamish Hei-Man Yeung
摘要: According to improvement of the receptor layer of various sensors of the type for detecting physical parameters (for example, a surface stress sensor, QCM, and SPR), all of high sensitivity, selectivity, and durability are achieved simultaneously. A porous material or a particulate material, e.g., nanoparticles, is used in place of a uniform membrane which has been conventionally used as a receptor layer. Accordingly, the sensitivity can be controlled by changing the membrane thickness of the receptor layer, the selectivity can be controlled by changing a surface modifying group to be fixed on the porous material or particulate material, and the durability can be controlled by changing the composition and surface properties of the porous material or particulate material.
-
公开(公告)号:US10458939B2
公开(公告)日:2019-10-29
申请号:US15552949
申请日:2016-02-25
发明人: Kota Shiba , Genki Yoshikawa , Gaku Imamura
IPC分类号: G01N27/12 , G01N5/02 , G01N29/02 , G01N29/036 , G01N3/00 , G01N21/00 , B82Y15/00 , G01N9/00 , G01N11/16
摘要: As a receptor layer, a film of a composite material of a base material such as a polymer and particles that adsorb an analyte is used. When the present invention is applied to a surface stress sensor or the like, the Young's modulus of the receptor layer, which significantly affects detection sensitivity, can be preset with a high degree of freedom, by independently selecting particles that adsorb a desired analyte and a base material that disperses said particles therein.
-
10.
公开(公告)号:US20190317066A1
公开(公告)日:2019-10-17
申请号:US16463585
申请日:2017-11-22
发明人: Gaku Imamura , Genki Yoshikawa , Takashi Washio
摘要: Provided is a novel analysis method that enables identification of a sample even when any sample is introduced during measurement carried out by using a chemical sensor. An input in which the amount of an unknown sample changes over time is provided to the chemical sensor, a response which is from the chemical sensor and which changes over time is measured, a sensor function (transmission function) of the chemical sensor with respect to the unknown sample is calculated on the basis of the input and the response, and the unknown sample is identified on the basis of the sensor function of the chemical sensor with respect to the unknown sample.
-
-
-
-
-
-
-
-
-