JPWO2021149048A5

JPWO2021149048A5 -

Info

Publication number: JPWO2021149048A5
Application number: JP2022543697A
Authority: JP
Publication date: 2024-01-23

Claims

A method for calculating pulse transition time ( PTT ) of a subject, the method comprising:
obtaining light data from the subject's face and fingers by a camera;
analyzing the light data to select data related to the subject 's face and fingers, respectively, by a computing device in communication with the camera;
detecting optical data from the facial skin and the finger skin;
determining a time series from the light data by collecting the light data until an elapsed period is reached, and then calculating the time series from the light data collected for the elapsed period;
Calculating the PTT from the time series ;
Equipped with
the light data comprises video data, and obtaining the light data from the finger skin comprises obtaining video data of the subject's face;
Obtaining the light data from the finger comprises obtaining video data of the subject's fingertip, and obtaining video data of the fingertip comprises detecting the fingertip by placing the fingertip over the camera. comprises acquiring video data of the skin of
Detecting the light data from the skin of the face comprises determining a plurality of face boundaries, selecting the most probable face boundary, and applying a histogram analysis to video data from the face;
Detecting the optical data from the finger skin includes determining a plurality of skin boundaries for the finger skin, selecting the skin boundary with the highest probability, and applying a histogram analysis to the video data from the finger skin. A method of preparing to do .

The camera comprises a plurality of mobile phone cameras, and the obtaining the light data further comprises obtaining video data from the plurality of mobile phone cameras, and the light data from the face is transmitted to the plurality of mobile phone cameras. 2. The method of claim 1 , wherein the light data from the finger is acquired by a first mobile phone camera of the cameras and the light data from the finger is acquired by a second mobile phone camera of the plurality of mobile phone cameras.

3. To obtain video data of the finger and the face, the subject places the finger over a rear mobile phone camera and the subject 's face is located in front of the front mobile phone camera . Method described.

4. The method of claim 3 , further comprising a fingertip on the rear mobile phone camera activating a flash associated with the rear mobile phone camera to provide light.

4. The video data of claim 3 , wherein each of the video data from the first mobile phone camera and the second mobile phone camera is analyzed to provide pulse signal information from the finger and the facial skin. Method.

6. The method of claim 5 , wherein the delay between pulse signals determined from the first mobile phone camera and the second mobile phone camera is determined according to synchronization via a single hardware clock.

the pulse signals are synchronized and then interpolated to yield the same sampling rate ;
The method comprises calculating a facial pulse waveform and a fingertip pulse waveform according to synchronization of the pulse signal,
7. The method of claim 6 , wherein the interpolation further comprises interpolating time series data from each of the face and the fingertip to convert a variable frame acquisition rate to a given fixed frame rate. .

2. The method of claim 1 , wherein determining the plurality of face boundaries comprises applying a multi-parameter convolutional neural network (CNN) to video data from the face to determine the face boundaries.

2. The method of claim 1 , wherein determining the plurality of skin boundaries comprises applying a multiparameter convolutional neural network (CNN) to video data from the finger to determine the skin boundaries.

Detecting the optical data from the finger skin comprises determining a plurality of fingertip boundaries, selecting the most probable fingertip boundary, and applying histogram analysis to video data from the fingertip. , the method of claim 1 .

11. The method of claim 10 , wherein determining the plurality of fingertip boundaries comprises applying a multi-parameter convolutional neural network (CNN) to video data from the fingertips to determine the fingertip boundaries.

Calculating the PTT further comprises combining measurements and metadata from light data from the facial skin and the finger skin, the metadata including the subject's weight, age, height, 2. The method of claim 1 , comprising one or more of biological sex, body fat percentage, and body muscle percentage.

The method further comprises determining the PTT from at least one additional physiological signal ;
2. The method of claim 1 , wherein the physiological signal is selected from the group consisting of tone, blood pressure, expiratory volume, and oxygen saturation (pSO2) .

The method further comprises determining at least one additional physiological signal from at least the PTT ;
14. The method of claim 13 , wherein the physiological signal is selected from the group consisting of tone, blood pressure, expiratory volume, and oxygen saturation (pSO2) .

2. The method of claim 1 , further comprising denoising and normalizing pulse signals from the finger and facial skin before calculating the PTT.

filtering the pulse signal ;
performing PPG style signal construction;
determining heart rate (HR) from the PPG-style signal construction;
Calculating the PTT from the HR;
Calculating blood pressure from the PTT;
16. The method of claim 15 , further comprising:

A system for calculating pulse transition time ( PTT ) of a subject, the system comprising:
a camera for acquiring optical data from the subject's face and fingertips;
A user computing device for receiving optical data from the camera, the user computing device comprising a processor and a memory for storing a plurality of instructions, the processor comprising:
analyzing the optical data to select data related to the subject 's face and fingertips ;
detecting optical data from the skin of the face and the skin of the fingertips;
determining a time series from the light data by collecting the light data until an elapsed period is reached, and then calculating the time series from the light data collected for the elapsed period;
a user computing device executing the instructions to calculate the PTT from the time series;
A system comprising:
The memory is configured to store a predefined native instruction code set, and the processor is configured to receive a corresponding primitive instruction selected from the predefined native instruction code set stored in the memory. configured to perform a predefined set of basic actions in response;
The memory includes a first machine code set selected from the native instruction code set relating to analyzing the light data to select data related to the subject's face and light data from the facial skin. a second machine code set selected from the native instruction code set relating to detecting the optical data; and determining a time series from the optical data by collecting the optical data until an elapsed period is reached; a third machine code set selected from the native instruction code set for calculating the time series from optical data collected for an elapsed time period; and a third machine code set selected from the native instruction code set for calculating a physiological signal from the time series. a fourth machine code set selected from the code sets;
The memory includes a fifth set of machine codes selected from the native instruction code sets relating to detecting the light data from the skin of the face, comprising determining a plurality of facial boundaries; a sixth set of machine codes selected from the set of native instruction codes relating to selecting facial boundaries; and a set of machine codes selected from the set of native instruction codes relating to applying a histogram analysis to video data from the face. 7 machine code sets are further stored,
The memory includes a ninth machine code set selected from the native instruction code set relating to analyzing the optical data to select data related to a fingertip of the subject and optical data from the skin of the fingertip. a tenth machine code set selected from the native instruction code set relating to detecting the optical data; and determining a time series from the optical data by collecting the optical data until an elapsed period is reached; an eleventh machine code set selected from the native instruction code set, the machine code set being related to calculating the time series from optical data collected for an elapsed time period; a twelfth machine code set selected from the instruction code set;
The memory includes a thirteenth machine code set selected from the native instruction code set relating to detecting the optical data from the skin of the fingertip, comprising determining a plurality of fingertip boundaries; a fourteenth machine code set selected from the native instruction code set relating to selecting a fingertip boundary; and a fourteenth machine code set selected from the native instruction code set relating to applying a histogram analysis to video data from the fingertip. The system further stores 15 machine code sets .

The memory further comprises an eighth machine code set selected from the native instruction code set relating to applying a multi-parameter convolutional neural network (CNN) to video data from the face to determine the face boundaries. 18. The system of claim 17 , wherein the system stores :

The memory further comprises a sixteenth machine code set selected from the native instruction code set relating to applying a multi-parameter convolutional neural network (CNN) to video data from the fingertip to determine the fingertip boundary. 20. The system of claim 18 , wherein the system stores :

18. The system of claim 17 , wherein the fingertip is pressed against the camera to acquire optical data , such that only skin detection rather than fingertip detection is performed.

the camera comprises a mobile phone camera; the optical data is obtained as video data from the mobile phone camera ; the user computing device comprises a mobile communication device; the mobile phone camera comprises a rear camera; 18. The system of claim 17 , wherein a fingertip is placed over the camera to acquire the video data .

22. The system of claim 21 , further comprising a flash associated with the mobile phone camera to provide light for acquiring the light data.