TIME-SERIES-DATA FEATURE EXTRACTION DEVICE, TIME-SERIES-DATA FEATURE EXTRACTION METHOD AND TIME-SERIES-DATA FEATURE EXTRACTION PROGRAM

    公开(公告)号:US20190228291A1

    公开(公告)日:2019-07-25

    申请号:US16330346

    申请日:2017-08-28

    摘要: A time-series-data feature extraction device includes: a data processing unit that processes a received unevenly spaced time-series-data group into an evenly spaced time-series-data group including omissions and an omission information group indicating presence or absence of omissions, based on a received input time-series data length and a received minimum observation interval; a model learning unit that learns a weight vector of each layer of a model with a difference between an element not missing in a matrix of the evenly spaced time-series-data group including omissions and an element of an output result of an output layer of the model being taken as an error, and stores the weight vector as a model parameter in a storage unit, the difference being; and a feature extraction unit that receives time-series data of a feature extraction target, calculates a value of the intermediate layer of the model with use of the model parameter stored in the storage unit by inputting the received time-series data of the feature extraction target into the model, and outputs the calculated value of the intermediate layer as a feature that represents temporal changes in data.

    LEARNING-TYPE SIGNAL SEPARATION METHOD AND LEARNING-TYPE SIGNAL SEPARATION DEVICE

    公开(公告)号:US20200050937A1

    公开(公告)日:2020-02-13

    申请号:US16607357

    申请日:2018-04-23

    IPC分类号: G06N3/08 G06N3/04 G06N3/063

    摘要: A learning-type signal separation method performed using a model formulation unit, which performs learning processing based on a training-use signal including a specific component, and a training-use signal not including the specific component, the training-use signals including a common characteristic. The learning-type signal separation method includes: generating learned data by causing the model formulation unit to perform learning processing based on the training-use signal and information indicating whether or not the specific component is included in the training-use signal, to generate a data series signal in which the specific component has been separated and removed from a data series of the training-use signal; acquiring an arbitrary signal including the common characteristic; and generating, based on the acquired arbitrary signal and the generated learned data, a data series signal in which the specific component has been separated and removed from a data series of the arbitrary signal.

    BIOLOGICAL SIGNAL PROCESSING METHOD AND BIOLOGICAL SIGNAL PROCESSING APPARATUS

    公开(公告)号:US20190038164A1

    公开(公告)日:2019-02-07

    申请号:US16075116

    申请日:2017-01-25

    IPC分类号: A61B5/0456 A61B5/00

    摘要: There is provided a biological signal processing apparatus. The biological signal processing apparatus includes a biological signal extraction unit (2) configured to extract a biological signal from an electrocardiographic waveform measured by an electrocardiograph (1), an averaging processing unit (3) configured to calculate averaged data using time-series data of the biological signals extracted by the biological signal extraction unit (2), an abnormal value determination unit (4) configured to determine, for each data, whether the data of the biological signal extracted by the biological signal extraction unit (2) is appropriate, based on the averaged data calculated using the data of the biological signals that have occurred before the data, and an abnormal value processing unit (5) configured to perform one of deletion and interpolation of the data of the biological signal determined to be inappropriate by the abnormal value determination unit (4).

    HEARTBEAT DETECTION METHOD AND HEARTBEAT DETECTION DEVICE

    公开(公告)号:US20190320926A1

    公开(公告)日:2019-10-24

    申请号:US16342174

    申请日:2017-09-25

    IPC分类号: A61B5/0456 A61B5/00 G16H40/63

    摘要: There is provided a heartbeat detection device. The heartbeat detection device includes a time difference value calculation unit (3) configured to calculate a time difference value from a sampling data string of an electrocardiographic waveform of a living body, an index value calculation unit (4) configured to calculate an index value for heartbeat detection based on the time difference value, a threshold setting unit (5) configured to determine whether a peak of the index value exceeding a threshold can be detected, to calculate, when not less than a predetermined number of peaks is consecutively detected, a threshold candidate based on an average value of the predetermined number of the most recent peaks among these peaks, not to update the threshold when the threshold candidate exceeds a threshold limit value based on a difference limit value that is likely be a time difference value of the electrocardiographic waveform of the living body, and to set the threshold candidate as a new threshold when the threshold candidate has a value equal to or less than the threshold limit value, and a heartbeat time determination unit (6) configured to set the sampling time of the peak as the heartbeat time when the peak of the index value exceeding the threshold is detected.

    HEARTBEAT DETECTION METHOD AND HEARTBEAT DETECTION DEVICE

    公开(公告)号:US20190069791A1

    公开(公告)日:2019-03-07

    申请号:US16080645

    申请日:2017-02-13

    摘要: There is provided a heartbeat detection device. The heartbeat detection device includes a time difference value calculation unit (3) configured to calculate a time difference value of sampling data of an electrocardiographic waveform, FIFO buffers (4-1, 4-2) configured to receive the time difference value, a FIFO buffer (4-3) configured to receive an output from the FIFO buffer (4-2), a FIFO buffer (4-4) configured to receive an output from the FIFO buffer (4-3), a minimum value detection unit (5) configured to detect, for each sampling time, a minimum value M out of the time difference value stored in the FIFO buffer (4-2) and the time difference value stored in the FIFO buffer (4-4), and a heartbeat time determination unit (6) configured to set, when a difference value M−a between the minimum value M and an output value a of the FIFO buffer (4-1) is equal to or greater than a threshold, sampling time of the output value a as heartbeat time.

    RESPIRATION ESTIMATION METHOD AND APPARATUS
    8.
    发明申请

    公开(公告)号:US20180333064A1

    公开(公告)日:2018-11-22

    申请号:US15774964

    申请日:2016-11-04

    摘要: There is provided a respiration estimation apparatus. The respiration estimation apparatus includes an R-wave amplitude detection unit (5) configured to detect an amplitude of an R wave from a cardiac potential waveform of a subject, an R-R interval detection unit (6) configured to detect an R-R interval as an interval between an R wave and an immediately preceding R wave from the cardiac potential waveform, an acceleration displacement detection unit (7) configured to detect an angular displacement of an acceleration vector from a triaxial acceleration signal by a respiratory motion of the subject, a Fourier transform unit (10) configured to Fourier-transform each of time-series signals of the R-wave amplitude, the R-R interval, and the angular displacement to obtain a frequency spectrum of each of the signals of the R-wave amplitude, the R-R interval, and the angular displacement, and a signal selection unit (11) configured to extract a frequency as a candidate of a respiration frequency of the subject from each of the frequency spectrum of the R-wave amplitude, the frequency spectrum of the R-R interval, and the frequency spectrum of the angular displacement, and select best data from the frequencies as the respiration frequency of the subject.