Devices, systems, and methods herein relate to non-invasive monitoring of a patient. These systems and methods may receive one or more image signals corresponding to a skin of the patient, process the one or more image signals using a first machine learning model, and predict a physiological parameter based on the processed one or more image signals using a second machine learning model.

Data derivable from remotely detected electromagnetic radiation (16) emitted or reflected by a subject (12) is processed. The data includes physiological information. An input signal is received and indicative entities are transmitted. The indicative entities being indicative of physiological information representative of at least one vital parameter (17; 150) in a subject (12) of interest, wherein the indicative entities are detected under consideration of at least one defined descriptive model (114) describing a relation between physical skin appearance characteristics and a corresponding representation in the input signal such that non-indicative side information represented by non-indicative entities in the input signal is substantially undetectable in a resulting transmitted signal. The at least one vital parameter (17; 150) is detected from the transmitted signal including the indicative entities. The at least one vital parameter (17; 150) is extracted under consideration of detected skin color properties representing circulatory activity.

The present application discloses a method of adjusting a parameter, the parameter being used to derive a physiological characteristic of an individual from an image of the user, the method comprising the steps of: obtaining the parameter for the individual; obtaining a corresponding parameter for a plurality of other individuals within a cohort of the individual; comparing the parameter for the individual with a statistically significant parameter for the plurality of other individuals; and adjusting the parameter for the individual in accordance with the difference between the parameter for the individual and the statistically significant parameter for the plurality of other individuals.

The present invention relates to a device, system and a method for generating a photoplethysmographic image carrying vital sign information of a subject. To provide an increased validity and robustness against motion, in particular against ballistocardiographic motion, the proposed device comprises an input interface (30) for obtaining image data of a skin region of a subject in at least two different wavelength channels, said image data comprising two or more image frames acquired by detecting light transmitted through or reflected from the skin region over time, wherein said image data comprise wavelength-dependent reflection or transmission information in said at least two different wavelength channels, a combination unit (31) for combining, per pixel or group of pixels and per time instant, image data values of said at least two different wavelength channels to obtain a time-variant pulse signal per pixel or group of pixels, and an image generation unit (32) for generating a photoplethysmographic image from a property of the respective pulse signals in a time window including at least two image frames.

The present invention relates to a system and a related method for extracting physiological information from remotely detected electromagnetic radiation. The system comprises an interface configured for receiving a data stream comprising image data representing an observed overall region comprising at least one subject of interest; a partitioning unit configured for defining a plurality of sub regions in the overall region; and a classifier configured for classifying the plurality of sub regions into at least one indicative type of region and at least one auxiliary type of region, wherein the at least one indicative type of region comprises at least one indicative region of interest at least partially representing the subject of interest. Preferably, the at least one auxiliary type of region comprises at least one reference region. More preferably, the system further comprises a data processor configured for processing at least one sub region classified as region of interest, thereby obtaining vital information.

The invention relates to a method for measuring the human heartbeat rate, comprising the steps of: 1) obtaining a video of the measured subject; 2) parsing the video into a series of image frames; 3) arranging the multiple generated image frames in sequential order; 4) detecting the position of the region of the face in each image; 5) obtaining the facial fluctuation frequency of the measured object according to the change of the position of the face region in the image frame system.

An embodiment in accordance with the present invention provides a method for non-invasively determining the functional severity of coronary artery stenosis. The method includes gathering patient-specific data related to concentration of a contrast agent within a coronary artery of a patient using a coronary computed tomography angiography scan (CCTA). The patient-specific data is used to calculate a patient-specific transluminal attenuation gradient for the coronary artery of the patient. The patient specific transluminal attenuation gradient is used to determine an estimate of a coronary flow velocity, pressure gradient, loss coefficient, coronary flow reserve, and/or fractional flow reserve for the patient. Coronary flow velocity, pressure gradient, loss coefficient, coronary flow reserve, and fractional flow reserve can then be used to estimate the functional severity of coronary artery stenosis.

An apparatus, method and computer program is described comprising: receiving video data for a scene; determining a movement measurement for at least some of a plurality of subframes of the video data; weighting the movement measurements to generate a plurality of weighted movement measurements, wherein the weighting is dependent on the subframe; and generating a movement indication for the scene from a combination of some or all of the weighted movement measurements.

In one embodiment, a method of amplifying temporal variation in at least two images comprises examining pixel values of the at least two images. The temporal variation of the pixel values between the at least two images can be below a particular threshold. The method can further include applying signal processing to the pixel values.

A system for monitoring one or more biological parameters of an individual, the system comprising: an illumination unit comprising at least one coherent light source configured for providing at least one coherent optical illumination beam and for directing said beam onto a selected inspection region on body of said individual; a light collection unit comprising at least a first imaging unit, said first imaging unit comprising a lens unit and a detector array and configured for collecting light returning from said inspection region to thereby generate at least one sequences of image data pieces associated with said inspection region, at a selected sampling rate; a control unit comprising at least one processor, the control unit is configured for receiving input data comprising said at least one sequence of image data piece and for processing the image data pieces for generating data indicative of one or more parameters of the individual. The processing comprises determining contrast variations in image data comprising image data indicative of one or more speckle patterns formed in light returning from illumination spots in said inspection region.