CN114209299A

CN114209299A - Human body physiological parameter detection channel selection method based on IPPG technology

Info

Publication number: CN114209299A
Application number: CN202111505574.0A
Authority: CN
Inventors: 董立泉; 徐歌; 孔令琴; 原静; 赵跃进; 刘明; 惠梅
Original assignee: Beijing Institute of Technology BIT
Current assignee: Beijing Institute of Technology BIT
Priority date: 2021-12-10
Filing date: 2021-12-10
Publication date: 2022-03-22
Anticipated expiration: 2041-12-10
Also published as: CN114209299B

Abstract

The invention relates to a method for selecting a human body physiological parameter detection channel based on an IPPG technology, belonging to the field of physiological signal detection. The invention illuminates the skin tissues of different parts of a human body by using the visible light source, and simultaneously acquires videos containing pulse information in corresponding illuminated areas by using the CCD camera. IPPG signals in three channels of RGB are extracted from a video through image data processing. And (3) establishing a classification model for IPPG signal disease diagnosis of the RGB three channels through a classification algorithm, and respectively inputting IPPG signals of the RGB three channels of the test set into the pre-established training set classification models. And respectively predicting the health condition of each channel subject of RGB in the test set according to the RGB channel IPPG signal disease classification model. And comparing the accuracy of classifying the diseases of each channel of the test set RGB, and selecting the channel with the highest accuracy as a diagnosis channel of the cardiovascular diseases. The invention selects the optimal IPPG signal channel as the channel for disease classification aiming at different types of diseases, thereby improving the accuracy of disease diagnosis.

Description

Human body physiological parameter detection channel selection method based on IPPG technology

Technical Field

The invention relates to a method for selecting a human body physiological parameter detection channel based on an IPPG (Internet protocol packet) technology, in particular to a method for selecting a disease diagnosis channel based on an imaging type photoplethysmography (IPPG), and belongs to the field of physiological signal detection.

Background

The pulse wave information of the human body contains various physiological parameters related to cardiovascular states, such as heart rate, heart rate variability, blood pressure, blood oxygen, respiratory rate and the like. Since the pulse wave contains a lot of physiological and pathological features of human body, the diagnosis of physiological and pathological states based on the variation of the waveform and amplitude of the pulse wave has become one of the most fundamental and challenging scientific hot problems.

In recent years, with the development of imaging sensor technology, imaging photoplethysmography (IPPG) has been widely used for detecting physiological parameters of human body due to its characteristics of non-contact, non-invasive measurement, low cost, simple operation, etc. IPPG technology usually uses visible light as a light source to illuminate a human body, uses an imaging sensor to obtain a video of a slight change in skin color caused by the intensity of reflected light absorbed by blood and tissues of the human body, and then extracts an IPPG signal from the video by an image processing technology. And rapidly separating a plurality of physiological parameter characteristics related to the cardiovascular state in the IPPG signal, and further judging the cardiovascular health state according to the physiological parameter characteristics. Visible light is used as a light source, IPPG signals can be extracted from RGB channels when video image processing is carried out, and the IPPG signals have the highest signal-to-noise ratio in a green channel, so that the green channel is often used as a diagnosis channel for evaluating simple cardiovascular states such as heart rate, heart rate variability and the like. However, the pathological mechanisms of different diseases such as diabetes and heart disease affect the cardiovascular system differently, and the pathological features thereof are obviously different in the representation of the RGB three channels of the IPPG signal. The depth of arrival at the tissue is also different for different wavelengths of light, the red channel being able to reach the deepest layers of the tissue, followed by green and blue light. When the IPPG signals generated by the light with different wavelengths return to the CCD imaging sensor, the carried deep information of the tissues is also different. Therefore, the IPPG signal of the green channel is used as a diagnostic channel for cardiovascular diseases, and the physiological and pathological features of the RGB channel are ignored, so that the accuracy of diagnosing the health status of patients with different cardiovascular diseases is not high. For the state evaluation of the complex cardiovascular related diseases, the optical physiological characteristics and the actual pathological characteristics of the IPPG signal in the RGB channels are combined, and the optimal IPPG signal channel is selected comprehensively as the diagnosis channel of the cardiovascular related diseases.

Disclosure of Invention

The invention aims to solve the problem that the diagnosis accuracy of the health state is not high when different cardiovascular diseases are diagnosed and classified based on a green channel IPPG signal at present. According to the method, when disease classification is carried out based on IPPG signals, the IPPG signals of the optimal color channels are selected as the channels for disease diagnosis aiming at different types of cardiovascular diseases, namely the IPPG signals of the optimal color channels are selected to correctly distinguish healthy people and disease subjects, and the accuracy of disease classification diagnosis is improved. The invention is suitable for the selection of IPPG channels of different cardiovascular diseases and is not limited to a specific disease. The invention avoids the disease state mistaking of the patient caused by the low-precision IPPG signal channel disease classification effect, and assists doctors to better diagnose the disease.

The purpose of the invention is realized by the following technical scheme.

A method for selecting a human body physiological parameter detection channel based on an IPPG technology comprises the following steps:

step 1, collecting a video of a pulse beating part of a subject;

step 2, processing the video collected in the step 1 to obtain an IPPG signal;

1) selecting a region containing a face image aiming at a first frame image of the acquired face video of the subject, and selecting [ a ] in a rectangular matrix₁,b₁]Taking the position of any pixel point as an interested area extracted by the IPPG signal; wherein, a₁Is height, b₁Is a width, a₁And b₁The size of the image collected by the CCD camera is smaller;

2) calculating the pixel mean value of the region of interest to obtain an original IPPG signal;

3) carrying out RGB channel separation on the original IPPG signal, and extracting IPPG signals of RGB three channels;

4) preprocessing the IPPG signals of the three channels to achieve the purpose of removing the influence of non-physiological parameter characteristics and obtain preprocessed IPPG signals of the RGB three channels;

step 3, repeating the steps 1 and 2, and collecting different healthy people and disease patients for multiple times to obtain a plurality of preprocessed IPPG signals of RGB three channels; the disease patient is a cardiovascular disease patient;

step 4, marking the IPPG signals obtained in the step 3 to distinguish the IPPG signals of healthy people and disease patients; namely, the label of the IPPG signal of a healthy person is marked as 0, and the label of the IPPG signal of a disease patient is marked as 1; then as a data set;

and 5, dividing the data set of each channel of RGB into a training set and a test set respectively, wherein the training sets of the channels of RGB are used for training classification models, inputting the test sets into the classification models trained by each channel respectively for classification, comparing the classification result of each subject in the test set of each channel with the real label value, and determining the IPPG signal channel with the highest classification accuracy as the disease diagnosis channel.

Advantageous effects

1. The invention discloses a method for selecting a human body physiological parameter detection channel based on an IPPG technology, which realizes the selection of a disease diagnosis channel based on the difference of IPPG signal physiological characteristics of healthy people and disease patients under different color channels.

2. The invention is suitable for obtaining IPPG signals of different parts of a human body and selecting IPPG signal color channels aiming at different types of cardiovascular diseases.

3. The invention realizes the selection of color channels for diagnosing different diseases simply and conveniently by utilizing an IPPG non-contact optical detection mode without wound, improves the accuracy of disease diagnosis, can take the result as the basis for assisting doctors to judge professionally, and avoids inaccurate classification effect precision of diseases and ill conditions caused by non-efficient IPPG signal color channel selection.

Drawings

Fig. 1 is a schematic view of a face video acquisition of a method for selecting a human physiological parameter detection channel based on an IPPG technique according to an embodiment;

FIG. 2 is a general flowchart of a method for selecting a human physiological parameter detection channel based on the IPPG technology provided by the embodiment;

fig. 3 is an algorithm flowchart of a method for selecting a human physiological parameter detection channel based on the IPPG technique according to an embodiment.

Detailed Description

To make the objects, advantages and features of the present invention more apparent, a method and an apparatus for selecting a disease diagnosis channel based on IPPG technology according to the present invention will be described in detail with reference to the accompanying drawings and specific embodiments. It should be noted that: the accompanying drawings, which are in a simplified form and are not to scale, are included for purposes of illustrating embodiments of the invention in a convenient and clear manner, and are incorporated in and constitute a part of the specification; the acquisition part of the IPPG signal is not limited to the face of a human body, and other parts of the human body capable of extracting pulse waves are also applicable. The invention is not limited to the classification of certain cardiovascular diseases from healthy people, and is applicable to the classification of diseases related to physiological parameter characteristic changes caused by the cardiovascular diseases. The classification algorithm employed in the present invention is not limited to a certain binary classification algorithm.

Example 1 of the present invention includes healthy persons and diabetic patients as subjects.

A human body physiological parameter detection channel selection method based on an IPPG technology is characterized in that a face video acquisition schematic diagram is shown in figure 1, and a general flow chart is shown in figure 2.

Step 1, illuminating skin tissues of different parts of a human body by using a visible light source, and acquiring videos containing pulse information in corresponding illumination areas by using a CCD (charge coupled device) camera;

step 1-1, starting a light source and a camera:

the subject sits still on a chair, the skin tissues of different parts of the human body are illuminated by a visible light source, and simultaneously a CCD camera is started. A polarizer is arranged in front of the visible light source, and an analyzer is arranged in front of the CCD camera. The polarizer and analyzer remove the effects of specularly reflected light. The skin tissues of different parts of the human body refer to human body parts such as the face, the arms and the fingers, and the pulse waves can be extracted. The embodiment is described with a human face video as a part from which an IPPG signal is extracted.

Step 1-2, video acquisition:

and collecting diffuse reflection light imaging videos containing pulse information in corresponding face illumination areas by using a CCD camera. The duration of the face video captured by the CCD camera is set to 30 seconds (or more than 30 seconds), the frame rate of the camera is 30fps, and the size of the captured image is 1920 × 1000. The subject remains relatively stationary during the photographing.

Step 2, processing the video collected in the step 1 to obtain an IPPG signal:

step 2-1, aiming at the first frame image of the collected face video of the subject, selecting a region containing the face image, and selecting [ a ] in a rectangular matrix₁,b₁]Taking the position of any pixel point as an interested area extracted by the IPPG signal; wherein, a₁Is height, b₁Is a width, a₁Size greater than 300 pixels, b₁Size greater than 300 pixels, a₁And b₁The size is smaller than the size of the collected image;

and 2-2, calculating a pixel mean value aiming at the region of interest selected in the step 2-1 to obtain an original IPPG signal.

IPPG compositionMatrix [ a ] in image device₁,b₁]Each pixel value within a region can be calculated by equation (1):

C(x,y)＝I×(ρ_s(t)+ρ_d(t))+V_n (1)

wherein, C (x, y) represents the light intensity value corresponding to the pixel with the coordinate of (x, y); i represents the light intensity of the light source; rho_s(t) and ρ_d(t) respectively representing a specular reflection coefficient and a diffuse reflection coefficient; v_nRepresenting quantization noise of the image sensor.

Removal of quantization noise V of image sensor by equation (2)_nI.e. all pixel averaging is performed on each frame of image.

Wherein the content of the first and second substances,

representing the average light intensity of all pixels on a frame of image. The polarizer and the analyzer remove non-physiological parameter light intensity information related to specular reflection, and all the non-physiological parameter light intensity information under the time sequence t is obtained according to a formula (3)

The sets constitute the IPPG signal.

And 2-3, performing RGB channel separation on the original IPPG signals in the step 2-2, and extracting IPPG signals of three RGB channels. A flow chart of the algorithm is shown in fig. 3.

And 2-4, preprocessing the IPPG signals of the three channels in the step 2-3 to obtain preprocessed IPPG signals of the RGB three channels. The preprocessing of the IPPG signals of the three RGB channels is to adopt a band-pass filter, a trend removing technology and the like to remove the influence of high-frequency noise, baseline drift and other non-physiological parameter characteristics.

Step 3, repeating the steps 1 and 2, and collecting different healthy people and diabetics for multiple times to obtain a plurality of preprocessed IPPG signals of RGB three channels;

the data volume of the IPPG signals of the RGB three channels of the healthy person and the diabetic patient is more than 100 groups, wherein the data volume of the IPPG signals of the diabetic patient is about 1/3 of the data volume of the IPPG signals of the healthy subject.

Step 4, marking the IPPG signals of the RGB three channels obtained in the step 3 to distinguish the IPPG signals of healthy people and diabetic patients; namely, the label of the IPPG signal of a healthy person is marked as 0, and the label of the IPPG signal of a diabetic patient is marked as 1; then as a data set;

step 4-1, marking the IPPG signals of the RGB three channels obtained in the step 3 to distinguish the IPPG signals of healthy people and diabetic patients; namely, the label of the IPPG signal of a healthy person is marked as 0, and the label of the IPPG signal of a diabetic patient is marked as 1;

and 4-2, respectively forming data sets by the RGB three-channel IPPG signals obtained in the step 4-1 and the labels, wherein the data sets are an R channel data set, a G channel data set and a B channel data set.

Step 5-1, randomly dividing the IPPG signal data sets of the three RGB channels obtained in the step 4-2 into training sets and test sets according to the same proportion, and establishing classification models for the three groups of training sets by adopting the same classification algorithm; the division ratio of the training set to the test set is 4: 1; the classification algorithm may be, but is not limited to, two classification algorithms such as DNN/deep fm/SVM/RF.

And 5-2, inputting the test set of each RGB channel into the classification models of the three RGB channels established in the step 5-1, and predicting the health state of the subject in each test set by using the classification models of the three RGB channels.

And 5-3, comparing the predicted health state of each subject in the RGB three-channel test set with the real label value to obtain the accuracy of judging the health states of the RGB three groups of subjects. The accuracy index adopts the following formula

Accuracy Accuracy_dIndicating the proportion of healthy and diabetic patients that were correctly predicted. Wherein TP_dCorrectly predicting the health status of a subject as a healthy person; TN (twisted nematic)_dIndicating that the health status of the subject was correctly predicted for diabetic patients: FP_dIndicating that the health status of the subject was erroneously predicted to be a healthy person: FN (FN)_dIndicating that the health status of the subject was incorrectly predicted to be diabetic.

And 5-4, comparing the accuracy of judging the health state of the subjects in the RGB three groups of test sets, and determining an IPPG signal channel with the highest classification accuracy as a diabetes diagnosis channel.

Through the steps, the IPPG signal channel most suitable for diabetes is selected as a diagnosis channel of diseases, and the diagnosis accuracy of the diseases is improved.

In a specific embodiment, the method for selecting the human physiological parameter detection channel based on the IPPG technology is used for a diabetic subject patient, and the optimal IPPG signal color channel is a red channel. The pulse wave is influenced by the viscosity of blood, the absorption and scattering effect of tissues, the elasticity of the artery wall, etc. when the ventricle passes to the artery vessel, and is finally revealed by the IPPG signal. The blood of a diabetic is affected by hyperglycemia for a long time, the influence degrees of different layers of skin tissues by the blood sugar are inconsistent, and the depths of signal channels with different colors reaching the tissues are inconsistent, so that the physiological parameter information carried by the IPPG signal in the RGB three channels is greatly different due to the fact that the physiological parameter information comprises the diabetes pathological information and the tissue optical information. The red light is the light penetrating the tissue to the deepest depth in the visible light, the carried physiological information is the most abundant, and the classification of the diabetic patients and the healthy people by adopting the IPPG signal of the red channel can obtain more accurate disease classification results, thereby having important significance for the analysis of further pathological information of the patients.

In example 2 of the present invention, a healthy person and a heart disease patient are used as subjects, and a facial image thereof is used as a portion from which an IPPG signal is extracted.

Step 1 and step 2 are the same as in example 1;

step 3, repeating the steps 1 and 2, and collecting different healthy people and heart disease patients for multiple times to obtain a plurality of preprocessed IPPG signals of RGB three channels;

the data volume of the IPPG signals of the RGB three channels of the healthy person and the heart disease patient is more than 100 groups, and the data volume of the IPPG signals of the heart disease patient is about 1/3 of the data volume of the IPPG signals of the healthy subject.

Step 4, marking the IPPG signals of the RGB three channels obtained in the step 3 to distinguish the IPPG signals of healthy people and heart disease patients; namely, the label of the IPPG signal of a healthy person is marked as 0, and the label of the IPPG signal of a heart disease patient is marked as 1; then as a data set;

step 4-1, marking the IPPG signals of the RGB three channels obtained in the step 3 to distinguish the IPPG signals of healthy people and patients with heart diseases; namely, the label of the IPPG signal of a healthy person is marked as 0, and the label of the IPPG signal of a heart disease patient is marked as 1;

Accuracy Accuracy_hIndicating the proportion of healthy and heart disease patients that were correctly predicted. Wherein TP_hCorrectly predicting the health status of a subject as a healthy person; TN (twisted nematic)_hIndicating that the health status of the subject was correctly predicted for a heart disease patient: FP_hIndicating that the health status of the subject was erroneously predicted to be a healthy person: FN (FN)_hIndicating that the health status of the subject was erroneously predicted to be a cardiac patient.

And 5-4, comparing the accuracy of judging the health state of the subjects in the RGB three groups of test sets, and determining an IPPG signal channel with the highest classification accuracy as a heart disease diagnosis channel.

In a specific embodiment, the method for selecting the human physiological parameter detection channel based on the IPPG technology can obtain that the optimal color channel for distinguishing the heart disease subject from the healthy subject is a green color channel. The heart beat of the heart disease patient has certain obstacles, and the pulse wave signals transmitted from the ventricles are influenced by the heart beat in the RGB channels to the same extent. Although the depth of the signal channels of different colors reaching the tissue is not consistent, the difference of pathological information carried by the IPPG signal is consistent in the embodiment of RGB three channels, and only the difference of intensity exists. The green channel has the highest response in the CCD imaging device due to the high signal-to-noise ratio, and is less influenced by motion artifacts than the red channel, so that more accurate disease diagnosis results can be obtained by classifying patients with heart diseases and healthy people by using the IPPG signal of the green channel.

Although the present invention has been described in detail with reference to the above embodiments, it should be understood by those skilled in the art that: modifications and equivalents may be made thereto without departing from the spirit and scope of the invention and it is intended to cover in the claims the invention as defined in the appended claims.

Claims

1. A method for selecting a human body physiological parameter detection channel based on an IPPG technology is characterized by comprising the following steps: the method comprises the following steps:

step 1, collecting a video of a pulse beating part of a subject;

step 2, processing the video collected in the step 1 to obtain an IPPG signal;