System and method for non-contact monitoring of physiological parameters

Abstract

A system and method for monitoring one or more physiological parameters of a subject under free-living conditions is provided. The system includes a camera configured to capture and record a video sequence including at least one image frame of at least one region of interest (ROI) of the subject's body. A computer in signal communication with the camera to receive signals transmitted by the camera representative of the video sequence includes a processor configured to process the signals associated with the video sequence recorded by the camera and a display configured to display data associated with the signals.

What is claimed is: 
     
         1 . A system for monitoring one or more physiological parameters of a subject under free-living conditions, the system comprising:
 a camera configured to capture and record a video sequence including at least one image frame of at least one region of interest (ROI) of the subject's body; and   a computer in signal communication with the camera to receive signals transmitted by the camera representative of the video sequence, the computer including a processor configured to process the signals associated with the video sequence recorded by the camera, and a display configured to display data associated with the signals.   
     
     
         2 . The system of  claim 1  wherein the computer is external to the camera. 
     
     
         3 . The system of  claim 1  wherein the system is configured to monitor one or more of the following physiological parameters: a heart beat; a heart rate (HR), a breathing pattern, a breathing amplitude, a breathing frequency (BF), an exhalation flow rate and/or a pulse transit time (PTT). 
     
     
         4 . The system of  claim 3  wherein the HR and the PTT are detected by tracking an image intensity change of the subject's skin. 
     
     
         5 . The system of  claim 3  wherein the BF and the exhalation flow rate are detected by tracking subtle body movements of the subject associated with breathing. 
     
     
         6 . The system of  claim 1  wherein the processor is configured to:
 select the at least one ROI of the subject's body; 
 detect body movement of the at least one ROI in the video sequence; and 
 determine a breathing pattern based on the detected body movement. 
 
     
     
         7 . The system of  claim 6  wherein the processor is configured to:
 select a region of pixels around an edge of each shoulder of the subject to be the regions of interest (ROIs); 
 determine a derivative of the ROIs along a vertical direction to obtained two differential images of the ROIs; 
 divide the differential image of each selected ROI of the ROIs into a top portion and an equal bottom portion along the edge of a respective shoulder to define an intensity of the top portion as dA, and an intensity of the bottom portion as dB; and 
 determine a vertical movement of the shoulders by: 
 
       
         
           
             
               dI 
               = 
               
                 
                   
                     dA 
                     - 
                     
                       d 
                        
                       
                           
                       
                        
                       B 
                     
                   
                   
                     dA 
                     + 
                     
                       d 
                        
                       
                           
                       
                        
                       B 
                     
                   
                 
                 . 
               
             
           
         
       
     
     
         8 . The system of  claim 7  wherein the processor is configured to:
 calculate dI for every frame of the video sequence; and 
 plot dI against time after applying a low-pass filter with a cut-off frequency of 2 Hz. 
 
     
     
         9 . The system of  claim 1  wherein the camera comprises a plurality of cameras configured to capture video sequences of a plurality of regions of interest of the subject's body. 
     
     
         10 . The system of  claim 1  wherein the camera is housed within a mobile device. 
     
     
         11 . The system of  claim 1  further comprising a light source configured to illuminate the at least one ROI of the subject's body. 
     
     
         12 . The system of  claim 1  further comprising a user interface configured to select at least one ROI and perform signal processing of data associated with the at least one ROI to determine a heart beat and a breathing signal, and display results in real time on the display. 
     
     
         13 . The system of  claim 1  wherein the camera records the video sequence of the subject's face, and the processor is configured to perform a Fast Fourier Transform (FFT) on an intensity signal averaged over a plurality of pixels in the at least one ROI, an FFT spectrum of the at least one ROI representing a heart beat signal as a peak at a frequency corresponding to the heart rate. 
     
     
         14 . The system of  claim 13  wherein the processor is configured to extract a peak amplitude in each pixel of the plurality of pixels and plot a peak amplitude on a colormap to analyze a variation of the heart beat signal in different areas of the subject's face. 
     
     
         15 . The system of  claim 1  wherein the processor is configured to:
 determine a volume flow rate of the exhaled air from the breathing pattern; and 
 determining an energy expenditure based on the determined volume flow rate of the exhaled air. 
 
     
     
         16 . A method for monitoring a breathing pattern of a subject, the method comprising:
 selecting a region of pixels around an edge of each shoulder of the subject to be the regions of interest (ROIs);   determining a derivative of the ROIs along a vertical direction to obtained two differential images of the ROIs;   determining a position of each shoulder by dividing a differential image of each selected ROI into a top portion and an equal bottom portion along the edge of the shoulder, wherein an intensity of the top portion is dA and an intensity of the bottom portion is dB; and   determining a vertical movement of each shoulder for every frame of the video sequence.   
     
     
         17 . The method of  claim 16  further comprising implementing a motion-tracking algorithm to correct motion artifacts, comprising:
 selecting at least one region of interest (ROI); 
 calculating body movement every 100 frames of the video sequence within the top portion and the bottom portion based on a shift in an x direction and a shift in a y direction; 
 updating each of the top portion and the bottom portion with the shift_x and the shift_y; and 
 plotting dI to generate a breathing curve. 
 
     
     
         18 . The method of  claim 17  further comprising determining an exhalation flow rate, wherein an exhaled breath volume is calculated from dI. 
     
     
         19 . The method of  claim 16  further comprising determining a pulse transit time comprising:
 analyzing a time difference of a plurality of PPG signals including a first PTT associated with transit time from the subject's heart to the subject's mouth (t 1 ), a second PTT associated with transit time from the subject's heart to the subject's left palm (t 2 ), and a third PTT associated with transit time from the subject's heart to the subject's right palm (t 3 ); 
 selecting a corresponding ROI selection for each of the first PTT, the second PTT, and the third PTT from the video sequence; and 
 plotting the plurality of PPG signals obtained from the ROI selection to find the time differences of the plurality of PPG signals from different regions of the subject's body in every heart cycle. 
 
     
     
         20 . The method of  claim 19  further comprising determining time differences in PTT among the different regions based on comparing peak locations of the plurality of PPG signals using a linear curve fitting method, comprising:
 selecting a heart beat cycle signal for analysis, wherein a peak location of the selected heart beat cycle signal is estimated by fitting two linear curves L 1  and L 2 , L 1  positioned on a rising edge of the peak location and L 2  is positioned on a falling edge of the peak location; 
 determining an estimated peak location as a point of intersection of the two linear curves L 1  and L 2 ; and 
 determining time differences in PTT by comparing peak locations of the plurality of PPG signals obtained at different body locations.