INCREASING ACCURACY OF A PHYSIOLOGICAL SIGNAL OBTAINED FROM A VIDEO OF A SUBJECT

摘要

What is disclosed is a system and method for increasing the accuracy of physiological signals obtained from video of a subject being monitored for a desired physiological function. In one embodiment, image frames of a video are received. Successive batches of image frames are processed. For each batch, pixels associated with an exposed body region of the subject are isolated and processed to obtain a time-series signal. If movement occurred during capture of these image frames that is below a pre-defined threshold level then parameters of a predictive model are updated using this batch's time-series signal. Otherwise, the last updated predictive model is used to generate a predicted time-series signal for this batch. The time-series signal is fused with the predicted time-series signal to obtain a fused time-series signal. The time-series signal for each batch is processed to obtain a physiological signal for the subject corresponding to the physiological function.

主权项

1. A method for increasing the accuracy of physiological signals obtained from a video of a subject being monitored for a desired physiological function, the method comprising:
receiving image frames of a video captured by a video imaging device of a subject being monitored for a desired physiological function, said image frames being acquired of at least a portion of an exposed body region of said subject where a physiological signal corresponding to said desired physiological function can be registered by said video imaging device; defining a size N of a batch of image frames such that Nmin≦N≦Nmax, where Nmin is a minimum size of a batch of image frames and Nmax is a maximum size of a batch of image frames; and for each batch of image frames of size N:
processing the current batch of image frames to isolate pixels associated with said body region;processing said isolated pixels to obtain a time-series signal for this batch;determining whether a movement occurred during video acquisition of this batch of image frames; andin response to said movement being below a predetermined threshold level, updating the parameters of a predictive model using said obtained time-series signal, otherwise:
using said updated predictive model to generate a predicted time-series signal for the current batch; andfusing said time-series signal with said predicted time-series signal to obtain a fused time-series signal for the current batch.