Respiration rate measuring method based on smart phone

Abstract

The invention discloses a breathing rate measuring method based on a smart phone. The method comprises the following steps: (1) presetting camera parameters, and collecting a user fingertip pulse photoplethysmography (PPG) signal; (2) processing the collected signals through a filtering method and the like to obtain breathing signals; (3) evaluating the respiratory signal quality based on a template matching method, and when the signal quality score is lower than a preset threshold value, not outputting a respiratory rate result and giving an operation suggestion; and (4) carrying out time domain peak valley detection on the high-quality signal to obtain a respiratory rate estimation value. According to the method, the breathing rate of the user is measured by using the rear camera and the flash lamp of the smart phone, additional medical equipment is not needed, the method is simple, convenient, rapid and efficient, the method can be widely applied to general common users, and a simple and feasible scheme is provided for personal health monitoring.

Description

Technical Field

[0001] This invention belongs to the field of medical and health care and smartphone application technology, and relates to a method for measuring respiratory rate based on a smartphone. Background Technology

[0002] Respiratory rate, or the number of breaths per minute, is a key indicator of human physiological state. In cardiovascular disease screening, abnormal changes in respiratory rate may be an early sign of potential problems; in sports rehabilitation, monitoring respiratory rate helps assess exercise intensity and recovery; and in daily health monitoring, it reflects an individual's mental state, stress level, and sleep quality. Therefore, accurate and convenient access to respiratory rate data is of great significance for personal health management.

[0003] For a long time, respiratory rate measurement has primarily relied on specialized medical equipment, such as electrocardiogram (ECG) monitors or dedicated respiratory monitors. While these devices provide highly accurate measurements in clinical settings, they have several limitations that restrict their application in routine health monitoring. Specifically, these devices are often expensive and bulky, making them difficult to carry and use in daily life. Secondly, they are complex to operate, usually requiring assistance from professionals, making it difficult for ordinary users to perform measurements independently. Furthermore, because most are standalone devices, they lack integration capabilities with mobile platforms, hindering real-time data transmission, management, and analysis, making long-term, continuous health tracking difficult.

[0004] In recent years, advancements in sensor, battery, and network technologies have paved the way for the application of smartphones in medical monitoring, enabling them to acquire health-related human activity data in real time and continuously. Conveniently measuring physiological indicators such as heart rate, respiratory rate, and blood pressure using their built-in cameras, microphones, and accelerometers has become an important research direction. Summary of the Invention

[0005] The purpose of this invention is to provide a smartphone-based method for measuring respiratory rate, addressing the problems of high cost, large size, and complex operation of traditional respiratory rate measurement equipment. This invention utilizes the rear camera and flash of a smartphone, employing steps such as image acquisition, signal processing, quality assessment, and peak / valley detection to achieve rapid and accurate measurement of respiratory rate.

[0006] The specific technical solution of this invention is as follows:

[0007] A smartphone-based method for measuring respiratory rate includes the following steps:

[0008] (1) Preset the smartphone camera parameters, gently cover the rear camera with the fingertip of the human finger, and continuously and stably collect the image data of each frame of the finger with the flash.

[0009] (2) Selectively sample pixel data for each frame of image data. Then convert all sampled pixels in the current frame into average brightness information to obtain the original light intensity time series data;

[0010] (3) Normalize and preprocess the original light intensity time series data, and suppress noise by filtering to retain the typical range of respiratory rate and obtain respiratory signal;

[0011] (4) The respiratory signal is evaluated based on the template matching method. When the signal quality score is lower than the preset threshold, the respiratory rate is not calculated and operation suggestions are provided to the user.

[0012] (5) Perform time-domain peak and valley detection on signals that meet the quality score to obtain respiratory rate estimates.

[0013] The present invention has the following beneficial effects:

[0014] (1) This invention makes full use of the existing built-in hardware (camera and flash) of smartphones, eliminating the need to purchase expensive professional medical equipment, significantly reducing the user's measurement costs, and greatly improving the universality and accessibility of the technology.

[0015] (2) The operation process of this invention is simple and intuitive. Users only need to lightly place their fingertip on the rear camera of the mobile phone to complete the measurement, without the need for professional guidance. This simple operation design greatly improves the user experience, making respiratory rate measurement seamlessly integrated into daily life.

[0016] (3) This invention effectively suppresses noise interference caused by factors such as ambient light and finger movement through signal preprocessing and signal quality assessment, ensuring the purity of the respiratory signal. When the signal quality is poor, the system will intelligently provide operation suggestions to avoid inaccurate measurement results, thereby ensuring the accuracy of the respiratory rate estimate and the stability of the measurement process. Attached Figure Description

[0017] Figure 1 This is an overall flowchart of the respiratory rate measurement method based on a smartphone according to the present invention. Detailed Implementation

[0018] To enable those skilled in the art to better understand the technical solution of the present invention, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0019] A smartphone-based method for measuring respiratory rate. (For example...) Figure 1As shown, the specific implementation process is as follows:

[0020] Step 1: Signal Acquisition

[0021] The user gently places their fingertip over the rear camera of their smartphone, ensuring the finger completely covers the camera. Upon initiating the measurement program, the system automatically presets camera parameters, specifically: adaptive maximum sampling frame rate to ensure image acquisition at the highest possible frame rate under varying lighting conditions; and locked automatic exposure to prevent fluctuations in image brightness or color due to changes in light during acquisition, thus ensuring signal stability. Throughout this process, the flash remains constantly on, providing a stable light source for the fingertip. The system continuously and stably acquires image data for each frame of the finger, including photoplethysmography (PPG) signals caused by changes in blood volume in the fingertip skin.

[0022] Step 2: Pixel Sampling and Processing

[0023] For each frame of image data acquired, this invention employs a single-column interval pixel sampling method. This method processes only one of two adjacent pixels in each column of pixel data to reduce computation and improve processing efficiency. Simultaneously, all sampled pixels in each frame are converted into average brightness information. Specifically, if the pixel is in YUV format, the Y component is extracted and the mean is calculated; if the pixel is in RGB format, the Y component is calculated first, followed by the mean, to obtain the original light intensity time series data.

[0024] Step 3: Signal Preprocessing

[0025] The raw light intensity time series data obtained in step two is normalized. To unify the sampling rate and facilitate subsequent processing, the raw light intensity time series data is interpolated and resampled to 150Hz based on the original image frame timestamps. Subsequently, a third-order bandpass filter is used to process the resampled signal. The passband frequency range of this filter is set to 0.1 - 0.5 Hz, which covers the typical value of normal adult respiratory rate (6-30 breaths / minute), effectively removing high-frequency noise and low-frequency baseline drift to obtain a respiratory signal containing respiratory rate information.

[0026] Step 4: Signal Quality Assessment

[0027] To ensure the reliability of the final results, this invention performs a quality assessment of the signal before calculating the respiratory rate. For the processed respiratory signal, a template-matching-based method is used to evaluate its quality score. This method segments the respiratory signal into independent stroke waveforms, normalizes the amplitude and time of each stroke waveform, calculates its correlation coefficient with a preset median template, and takes the median of the correlation coefficients. Calculate the SQI score of the respiratory signal. ,in The median of the correlation coefficient between the beat waveform and the median template; and These are the preset minimum and maximum thresholds for the correlation coefficient. Based on the evaluation results, if the signal quality score is lower than the preset threshold (e.g., 60 points), the signal quality is considered poor, and subsequent respiratory rate calculations will not be performed. In this case, the system will provide the user with operational suggestions, such as adjusting finger position or changing the measurement environment, to improve signal acquisition quality.

[0028] Step 5: Peak Detection and Respiratory Rate Calculation

[0029] Once the signal quality score reaches a preset threshold, time-domain peak-valley detection is performed to extract an estimated respiratory rate. The specific steps are as follows:

[0030] (1) Within a preset measurement time, the main signal is subjected to peak or trough detection to obtain the time points of all valid peaks or troughs;

[0031] (2) Calculate the time interval between all adjacent valid peaks or troughs. ;

[0032] (3) Perform statistical analysis on all time intervals to obtain the average respiratory cycle. ;

[0033] (4) Calculate the estimated respiratory rate based on the mean respiratory cycle. bpm.

Claims

1. A method for measuring respiratory rate based on a smartphone, characterized in that, Includes the following steps: (1) Preset the smartphone camera parameters, gently cover the rear camera with the fingertip of the human finger, and continuously and stably collect the image data of each frame of the finger with the flash. (2) Selectively sample pixel data for each frame of image data; and convert all sampled pixels in the current frame into average brightness information to obtain the original light intensity time series data; (3) Normalize and preprocess the original light intensity time series data, and suppress noise by filtering to retain the typical range of respiratory rate and obtain respiratory signal; (4) The respiratory signal is evaluated based on the template matching method. When the signal quality score is lower than the preset threshold, the respiratory rate is not calculated and operation suggestions are provided to the user. (5) Perform time-domain peak and valley detection on signals that meet the quality score to obtain respiratory rate estimates.

2. The method for measuring respiratory rate based on a smartphone according to claim 1, characterized in that, In step (1), the camera parameters of the smartphone are adaptive maximum sampling frame rate, and locked automatic exposure and automatic white balance.

3. The method for measuring respiratory rate based on a smartphone according to claim 1, characterized in that, In step (2), the selective sampling method is single-column interval pixel sampling, that is, only one pixel data is processed in two adjacent pixel data in a column.

4. The method for measuring respiratory rate based on a smartphone according to claim 1, characterized in that, In step (2), all sampled pixels in the frame are converted into average brightness information. Specifically, if the pixel is in YUV format, the Y component is extracted and the mean is calculated. If the pixel is in RGB format, the Y component is calculated first and then the mean is calculated.

5. The method for measuring respiratory rate based on a smartphone according to claim 1, characterized in that, In step (3), the preprocessing step is as follows: first, the original light intensity time series data is interpolated and resampled to a fixed sampling rate according to the original image frame timestamp, and then a third-order bandpass filter is used to process the resampled signal.

6. The method for measuring respiratory rate based on a smartphone according to claim 1, characterized in that, In step (4), the quality assessment method is the template matching method, which includes: segmenting the respiratory signal into independent beat waveforms, normalizing the amplitude and time of the beat waveforms, and calculating the correlation coefficient between each normalized waveform and a preset median template to obtain the quality assessment index.

7. The method for measuring respiratory rate based on a smartphone according to claim 1, characterized in that, The method for extracting respiratory rate estimates using peak-valley detection technology is as follows: (1) Within the preset measurement time, the main signal is detected for peaks or troughs to obtain the time points of all valid peaks or troughs; (2) Calculate the time interval between all adjacent valid peaks or troughs. ; (3) Perform statistical analysis on all time intervals to obtain the average respiratory cycle. ; (4) Calculate the estimated respiratory rate based on the mean respiratory cycle. bpm.

Images