Pulse wave detection method, pulse wave detection apparatus, and recording medium

• US 9,986,922 B2
• Filed: 03/04/2015
• Issued: 06/05/2018
• Est. Priority Date: 09/07/2012
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Uchida, Daisuke, Sakata, Masato, Sekiguchi, Hidenori, Inomata, Akihiro
• Primary Examiner(s): Flory, Christopher A
• Application Number:
  US14/638,570
• Publication Number:
  US 20150173630A1
• Time in Patent Office: 1,189 Days
• Field of Search: 600479
• CPC Class Codes: A61B 2576/00 Medical imaging apparatus involving image processing or analysis A61B1/00009, A61B6/52 and A61B8/52 take precedence; image processing per se G06T; details of algorithms for analysing medical images G06T7/0012
• CPC Class Codes: A61B 5/0059 using light, e.g. diagnosis by transillumination, diascopy, fluorescence A61B5/0093, A61B5/1455, A61B5/02416 and A61B5/0261 take precedence
• CPC Class Codes: A61B 5/024 Detecting, measuring or recording pulse rate or heart rate A61B5/0205, A61B5/021 take precedence
• CPC Class Codes: A61B 5/02416 using photoplethysmograph signals, e.g. generated by infrared radiation A61B5/14552 takes precedence
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• CPC Class Codes: A61B 5/7257 using Fourier transforms
• CPC Class Codes: A61B 5/7275 Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
• CPC Class Codes: G16H 30/40 for processing medical images, e.g. editing

Abstract

A pulse wave detection method includes obtaining an image obtained by photographing a subject with an imaging device, extracting intensities representative of signal components of a specific frequency band for respective wavelength components among signals of a plurality of wavelength components included in the image, calculating, using the intensities extracted for the respective wavelength components, a weight coefficient by which a signal is multiplied when the signals are calculated between the wavelength components to minimize an arithmetic value of the signal components in the specific frequency band after multiplication, multiplying at least one of the signals of the respective wavelength components by the weight coefficient, performing arithmetic operation on the signals between the wavelength components after multiplication by the weight coefficient, and detecting pulse waves of the subject using a signal after the arithmetic operation.

Claims

1. A pulse wave detection method comprising:

obtaining images obtained by photographing a subject with an imaging device;

extracting intensities representative of signal components of a specific frequency band for respective wavelength components among signals of a plurality of wavelength components included in the images, the specific frequency band having a frequency band to which pulse waves do not correspond, and having a section having less than or equal to a predetermined length overlapping a frequency band that the pulse waves are enabled to take;

calculating, using the intensities extracted for the respective wavelength components, a weight coefficient by which a signal is multiplied when the signals are calculated between the wavelength components to minimize an arithmetic value of the signal components in the specific frequency band after multiplication;

multiplying at least one of the signals of the respective wavelength components by the weight coefficient;

performing arithmetic operation on the signals between the wavelength components after multiplication by the weight coefficient; and

detecting pulse waves of the subject using a signal after the arithmetic operation to enable suppression of an increase in processing load or a decrease in accuracy when noise is reduced, wherein

the extracting includes first extracting the signal components of the specific frequency band for the respective wavelength components from the signals of the respective wavelength components using a low-pass filter or a band-pass filter and second extracting signal components of the frequency band that the pulse waves are enabled to take for the respective wavelength components from the signals of the respective wavelength components using a band-pass filter,

the calculating includes calculating the weight coefficient using the intensities of the signal components of the specific frequency band, the signal components being extracted for the respective wavelength components,

the multiplying includes multiplying the signal components of the frequency band that the pulse waves are enabled to take by the weight coefficient,

the performing includes performing arithmetic operation on the signal components of the frequency band that the pulse waves are enabled to take between the wavelength components after multiplication by the weight coefficient, and

the detecting includes detecting the pulse waves of the subject using a signal waveform after the arithmetic operation, wherein

the first extracting includes extracting absolute intensity values in the specific frequency band that are extracted for the respective wavelength components, and

the calculating includes calculating the weight coefficient by calculating a ratio of the absolute intensity values between the wavelength components, wherein

the first extracting includes extracting the absolute intensity value of the signal component of the specific frequency band, by executing a multiplication process of exponentiating the signal component of an R component in the specific frequency band, and extracting the absolute intensity value of the signal component of the specific frequency band, by executing a multiplication process of exponentiating the signal component of a G component in the specific frequency band, wherein

the pulse wave detection method further includes executing smoothing on time-series data of the absolute intensity value in the specific frequency band to respond to time change.

2. The pulse wave detection method according to claim 1, further comprising:

converting the signals of the respective wavelength components into frequency components, wherein

the multiplying includes multiplying one of spectrums obtained by converting the signals of the wavelength components into the frequency components by the weight coefficient,

the performing includes performing arithmetic operation of the spectrums between the wavelength components after multiplication by the weight coefficient, and

the detecting includes detecting the pulse waves of the subject using a spectrum after the arithmetic operation.

3. The pulse wave detection method according to claim 2, wherein

the extracting includes extracting an average power intensity of the specific frequency band from the spectrums obtained by converting the signals of the respective wavelength components into the frequency components, and

the calculating includes calculating the weight coefficient by calculating a ratio of the average power intensity between the wavelength components.

4. The pulse wave detection method according to claim 1, wherein the signals include two or more wavelengths, and include different light absorption sensitivities to hemoglobin.

5. The pulse wave detection method according to claim 4, wherein at least one signal among the signals of the wavelength components is a signal having a band based on a light wavelength of 525 nm, and the other signals are signals having bands of light wavelengths other than the band.

6. The pulse wave detection method according to claim 1, wherein signal components less than 1 Hz are used as the signal components of the specific frequency band.

7. The pulse wave detection method according to claim 1, wherein

the detecting includes, when canceling a noise component in a time-series space to detect pulse waves, obtaining a heart beat waveform serving as a form of pulse waves without Fourier transform.

8. The pulse wave detection method according to claim 1, wherein

the time-series data of a signal after arithmetic operation has peaks located in positions substantially identical with positions of peaks of an electrocardiogram waveform of a reference, and they have relation substantially similar to each other.

9. A pulse wave detection apparatus comprising:

a processor configured to;

obtain images obtained by photographing a subject with an imaging device;

extract intensities representative of signal components of a specific frequency band for respective wavelength components among signals of a plurality of wavelength components included in the images, the specific frequency band having a frequency band to which pulse waves do not correspond, and having a section having less than or equal to a predetermined length overlapping a frequency band that the pulse waves are enabled to take;

calculate, using the intensities extracted for the respective wavelength components, a weight coefficient by which a signal is multiplied when the signals are calculated between the wavelength components to minimize an arithmetic value of the signal components in the specific frequency band after multiplication;

multiply at least one of the signals of the respective wavelength components by the weight coefficient;

perform arithmetic operation on the signals between the wavelength components after multiplication by the weight coefficient; and

detect pulse waves of the subject using a signal after the arithmetic operation to enable suppression of an increase in processing load or a decrease in accuracy when noise is reduced, wherein

the extracting includes first extracting the signal components of the specific frequency band for the respective wavelength components from the signals of the respective wavelength components using a low-pass filter or a band-pass filter and second extracting signal components of the frequency band that the pulse waves are enabled to take for the respective wavelength components from the signals of the respective wavelength components using a band-pass filter,

the calculating includes calculating the weight coefficient using the intensities of the signal components of the specific frequency band, the signal components being extracted for the respective wavelength components,

the multiplying includes multiplying the signal components of the frequency band that the pulse waves are enabled to take by the weight coefficient,

the performing includes performing arithmetic operation on the signal components of the frequency band that the pulse waves are enabled to take between the wavelength components after multiplication by the weight coefficient, and

the detecting includes detecting the pulse waves of the subject using a signal waveform after the arithmetic operation, wherein

the first extracting includes extracting absolute intensity values in the specific frequency band that are extracted for the respective wavelength components, and

the calculating includes calculating the weight coefficient by calculating a ratio of the absolute intensity values between the wavelength components, wherein

the first extracting includes extracting the absolute intensity value of the signal component of the specific frequency band, by executing a multiplication process of exponentiating the signal component of an R component in the specific frequency band, and extracting the absolute intensity value of the signal component of the specific frequency band, by executing a multiplication process of exponentiating the signal component of a G component in the specific frequency band, wherein

the processor is further configured to execute smoothing on time-series data of the absolute intensity value in the specific frequency band to respond to time change.

10. A non-transitory computer readable recording medium having stored therein a pulse wave detection program that causes a computer to execute a process comprising:

obtaining images obtained by photographing a subject with an imaging device;

extracting intensities representative of signal components of a specific frequency band for respective wavelength components among signals of a plurality of wavelength components included in the images, the specific frequency band having a frequency band to which pulse waves do not correspond, and having a section having less than or equal to a predetermined length overlapping a frequency band that the pulse waves are enabled to take;

calculating, using the intensities extracted for the respective wavelength components, a weight coefficient by which a signal is multiplied when the signals are calculated between the wavelength components to minimize an arithmetic value of the signal components in the specific frequency band after multiplication;

multiplying at least one of the signals of the respective wavelength components by the weight coefficient;

performing arithmetic operation on the signals between the wavelength components after multiplication by the weight coefficient; and

detecting pulse waves of the subject using a signal after the arithmetic operation to enable suppression of an increase in processing load or a decrease in accuracy when noise is reduced, wherein

the extracting includes first extracting the signal components of the specific frequency band for the respective wavelength components from the signals of the respective wavelength components using a low-pass filter or a band-pass filter and second extracting signal components of the frequency band that the pulse waves are enabled to take for the respective wavelength components from the signals of the respective wavelength components using a band-pass filter,

the calculating includes calculating the weight coefficient using the intensities of the signal components of the specific frequency band, the signal components being extracted for the respective wavelength components,

the multiplying includes multiplying the signal components of the frequency band that the pulse waves are enabled to take by the weight coefficient,

the performing includes performing arithmetic operation on the signal components of the frequency band that the pulse waves are enabled to take between the wavelength components after multiplication by the weight coefficient, and

the detecting includes detecting the pulse waves of the subject using a signal waveform after the arithmetic operation, wherein

the first extracting includes extracting absolute intensity values in the specific frequency band that are extracted for the respective wavelength components, and

the calculating includes calculating the weight coefficient by calculating a ratio of the absolute intensity values between the wavelength components, wherein

the first extracting includes extracting the absolute intensity value of the signal component of the specific frequency band, by executing a multiplication process of exponentiating the signal component of an R component in the specific frequency band, and extracting the absolute intensity value of the signal component of the specific frequency band, by executing a multiplication process of exponentiating the signal component of a G component in the specific frequency band, wherein

the process further includes executing smoothing on time-series data of the absolute intensity value in the specific frequency band to respond to time change.