Abstract: A method of training a supervised machine-learning algorithm to determine a sleep state of a subject. The method uses training data and training labels in respect of a plurality of subjects derived from video images of the subjects. The training data comprises: at least one measure of subject movement; and at least one cardiorespiratory parameter of the subject. The method comprises training the supervised machine learning algorithm using the training data and the training labels.

Abstract: A method and apparatus for monitoring a human or animal subject in a room using video imaging of the subject and analysis of the video image to detect and quantify movement of the subject and to derive an estimate of vital signs such as heart rate or breathing rate. The method includes techniques for de-correlating global intensity variations such as sunlight changes, compensating for noise, eliminating areas not of interest in the image, and quickly and automatically finding regions of interest for detecting subject movement and estimating vital signs. A logic machine is used for interpreting detected movement of the subject, and an artificial neural network is used to calculate a confidence measure for the vital signs estimates from signal quality indices. The confidence measure may be used with a normal density filter to output estimates of the vital signs.

Abstract: A method and apparatus for monitoring a human or animal subject in a room using video imaging of the subject and analysis of the video image to detect and quantify movement of the subject and to derive an estimate of vital signs such as heart rate or breathing rate. The method includes techniques for de-correlating global intensity variations such as sunlight changes, compensating for noise, eliminating areas not of interest in the image, and quickly and automatically finding regions of interest for detecting subject movement and estimating vital signs. A logic machine is used for interpreting detected movement of the subject, and an artificial neural network is used to calculate a confidence measure for the vital signs estimates from signal quality indices. The confidence measure may be used with a normal density filter to output estimates of the vital signs.

Abstract: A method and apparatus for monitoring a human or animal subject in a room using video imaging of the subject and analysis of the video image to detect and quantify movement of the subject and to derive an estimate of vital signs such as heart rate or breathing rate. The method includes techniques for de-correlating global intensity variations such as sunlight changes, compensating for noise, eliminating areas not of interest in the image, and quickly and automatically finding regions of interest for detecting subject movement and estimating vital signs. A logic machine is used for interpreting detected movement of the subject, and an artificial neural network is used to calculate a confidence measure for the vital signs estimates from signal quality indices. The confidence measure may be used with a normal density filter to output estimates of the vital signs.

Abstract: A method and apparatus for monitoring a human or animal subject in a room using video imaging of the subject and analysis of the video image to detect and quantify movement of the subject and to derive an estimate of vital signs such as heart rate or breathing rate. The method includes techniques for de-correlating global intensity variations such as sunlight changes, compensating for noise, eliminating areas not of interest in the image, and quickly and automatically finding regions of interest for detecting subject movement and estimating vital signs. A logic machine is used for interpreting detected movement of the subject, and an artificial neural network is used to calculate a confidence measure for the vital signs estimates from signal quality indices. The confidence measure may be used with a normal density filter to output estimates of the vital signs.

Abstract: In order to detect gross subject movement in a video image in a way which is not sensitive to illumination change, for example illumination changes caused by movement of shadows or sunlight, spaced pairs of image frames are selected from a video sequence and sub-divided into cells, and spatial frequency analysis is performed in each cell. The magnitude of the spatial frequency components in corresponding cells in the two selected image frames are compared. If the number of cells with high magnitude difference is high then the video image is determined as containing gross subject movement whereas if the number of cells with high magnitude differences is low, the sequence is determined as not containing gross movement, though it may contain illumination changes or no or fine movement.

Abstract: A method and apparatus for extracting a breathing rate estimate from video images of a respiring subject. Signals corresponding to the spatial coordinates of feature points tracked through the video sequence are filtered and excessively large changes are attenuated to reduce movement artefacts. The signals are differentiated and signals which correlate most strongly with other signals are selected. The selected signals are subject to principal component analysis and the best quality of the top five principal components is selected and its frequency is used to calculate and output a breathing rate estimate. The method is particularly suitable for detecting respiration in subject in secure rooms where the video image is of substantially the whole room and the subject is only a small part of the image, and maybe covered or uncovered.

Abstract: A method and apparatus for detecting fine movement of a subject in video images, and for distinguishing over noise and other image artefacts. The video images are processed to detect movement of image features through the sequence and to calculate how spatially distributed those moving features are across the image. The movement tracks of the features may be subject to principal component analysis and a spatial dispersion measure calculated by the product of the distance between tracked image features and the contributions of those image features to the most significant principal components. If the spatial dispersion measure is high then this is indicative of feature movement being dispersed widely across the image, whereas if it is low, it is indicative of the main feature movements being concentrated in one part of the image, and thus more likely to represent subject movement than noise.

Abstract: A method and apparatus for estimating heart rate of a subject from a video image of the subject. Regions of interest are generated by: detecting and tracking feature points through the video image sequence, triangulating the feature points and generating square regions of interest corresponding to the in-circles of the triangles; or, according to size and location probability distributions which are defined to have a high probability for image areas away from strong intensity gradients and which generate good quality signals. In an alternative embodiment, the intensity variations from the square regions of interest through the frame sequence are taken as time series signals and those signals which have a strong peak in the power spectrum are selected and subject to principal component analysis. The principal component with a highest signal quality is selected and its frequency is found and used to estimate the heart rate.

Abstract: A method and apparatus for monitoring the health and safety of a subject in a room such as a secure room based on video images of the subject. The images are analysed to characterise the movement of the subject as gross movement, fine movement or no movement. In the case of gross movement, no vital signs of the subject are estimated and a display indicates that the subject is moving, but no vital signs are available. In the absence of gross movement, vital signs of the subject such as heart rate or breathing rate are estimated from the video images of the subject, for example by detecting and analysing photoplethysmogram signals in the video images, and the vital signs are displayed. Alerts may be generated if the vital signs are out of the normal physiological range. If vital signs cannot be detected in the video images but the movement of the subject is characterised as fine movement, the display shows that no vital signs are being estimated, but that the subject is moving.

Abstract: A method and apparatus for monitoring a human or animal subject in a room using video imaging of the subject and analysis of the video image to detect and quantify movement of the subject and to derive an estimate of vital signs such as heart rate or breathing rate. The method includes techniques for de-correlating global intensity variations such as sunlight changes, compensating for noise, eliminating areas not of interest in the image, and quickly and automatically finding regions of interest for detecting subject movement and estimating vital signs. A logic machine is used for interpreting detected movement of the subject, and an artificial neural network is used to calculate a confidence measure for the vital signs estimates from signal quality indices. The confidence measure may be used with a normal density filter to output estimates of the vital signs.

Abstract: A method and apparatus for monitoring a human or animal subject in a room using video imaging of the subject and analysis of the video image to detect and quantify movement of the subject and to derive an estimate of vital signs such as heart rate or breathing rate. The method includes techniques for de-correlating global intensity variations such as sunlight changes, compensating for noise, eliminating areas not of interest in the image, and quickly and automatically finding regions of interest for detecting subject movement and estimating vital signs. A logic machine is used for interpreting detected movement of the subject, and an artificial neural network is used to calculate a confidence measure for the vital signs estimates from signal quality indices. The confidence measure may be used with a normal density filter to output estimates of the vital signs.

Abstract: A method and apparatus for monitoring a human or animal subject in a room using video imaging of the subject and analysis of the video image to detect and quantify movement of the subject and to derive an estimate of vital signs such as heart rate or breathing rate. The method includes techniques for de-correlating global intensity variations such as sunlight changes, compensating for noise, eliminating areas not of interest in the image, and quickly and automatically finding regions of interest for detecting subject movement and estimating vital signs. A logic machine is used for interpreting detected movement of the subject, and an artificial neural network is used to calculate a confidence measure for the vital signs estimates from signal quality indices. The confidence measure may be used with a normal density filter to output estimates of the vital signs.

Abstract: A method and apparatus for detecting fine movement of a subject in video images, and for distinguishing over noise and other image artefacts. The video images are processed to detect movement of image features through the sequence and to calculate how spatially distributed those moving features are across the image. The movement tracks of the features may be subject to principal component analysis and a spatial dispersion measure calculated by the product of the distance between tracked image features and the contributions of those image features to the most significant principal components. If the spatial dispersion measure is high then this is indicative of feature movement being dispersed widely across the image, whereas if it is low, it is indicative of the main feature movements being concentrated in one part of the image, and thus more likely to represent subject movement than noise.

Abstract: In order to detect gross subject movement in a video image in a way which is not sensitive to illumination change, for example illumination changes caused by movement of shadows or sunlight, spaced pairs of image frames are selected from a video sequence and sub-divided into cells, and spatial frequency analysis is performed in each cell. The magnitude of the spatial frequency components in corresponding cells in the two selected image frames are compared. If the number of cells with high magnitude difference is high then the video image is determined as containing gross subject movement whereas if the number of cells with high magnitude differences is low, the sequence is determined as not containing gross movement, though it may contain illumination changes or no or fine movement.

Abstract: A method and apparatus for monitoring the health and safety of a subject in a room such as a secure room based on video images of the subject. The images are analysed to characterise the movement of the subject as gross movement, fine movement or no movement. In the case of gross movement, no vital signs of the subject are estimated and a display indicates that the subject is moving, but no vital signs are available. In the absence of gross movement, vital signs of the subject such as heart rate or breathing rate are estimated from the video images of the subject, for example by detecting and analysing photoplethysmogram signals in the video images, and the vital signs are displayed. Alerts may be generated if the vital signs are out of the normal physiological range. If vital signs cannot be detected in the video images but the movement of the subject is characterised as fine movement, the display shows that no vital signs are being estimated, but that the subject is moving.

Abstract: A method and apparatus for estimating heart rate of a subject from a video image of the subject. Regions of interest are generated by: detecting and tracking feature points through the video image sequence, triangulating the feature points and generating square regions of interest corresponding to the in-circles of the triangles; or, according to size and location probability distributions which are defined to have a high probability for image areas away from strong intensity gradients and which generate good quality signals. In an alternative embodiment, the intensity variations from the square regions of interest through the frame sequence are taken as time series signals and those signals which have a strong peak in the power spectrum are selected and subject to principal component analysis. The principal component with a highest signal quality is selected and its frequency is found and used to estimate the heart rate.

Abstract: A method and apparatus for extracting a breathing rate estimate from video images of a respiring subject. Signals corresponding to the spatial coordinates of feature points tracked through the video sequence are filtered and excessively large changes are attenuated to reduce movement artefacts. The signals are differentiated and signals which correlate most strongly with other signals are selected. The selected signals are subject to principal component analysis and the best quality of the top five principal components is selected and its frequency is used to calculate and output a breathing rate estimate. The method is particularly suitable for detecting respiration in subject in secure rooms where the video image is of substantially the whole room and the subject is only a small part of the image, and maybe covered or uncovered.