The present invention relates to the field of medical monitoring, and in particular non-contact monitoring of one or more physiological parameters in a region of a patient during surgery. Systems, methods, and computer readable media are described for generating a pulsation field and/or a pulsation strength field of a region of interest (ROI) in a patient across a field of view of an image capture device, such as a video camera. The pulsation field and/or the pulsation strength field can be generated from changes in light intensities and/or colors of pixels in a video sequence captured by the image capture device. The pulsation field and/or the pulsation strength field can be combined with indocyanine green (ICG) information regarding ICG dye injected into the patient to identify sites where blood flow has decreased and/or ceased and that are at risk of hypoxia.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for classifying a patient. In one aspect, a method comprises: receiving multi-modal data characterizing a patient, wherein the multi-modal data comprises a respective feature representation for each of a plurality of modalities; processing the multi-modal data characterizing the patient using an encoder neural network to generate an embedding of the multi-modal data characterizing the patient; determining a respective classification score for each patient category in a set of patient categories based on the embedding of the multi-modal data characterizing the patient; and classifying the patient as being included in a corresponding patient category from the set of patient categories based on the classification scores.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for jointly training an encoder neural network and a decoder neural network. In one aspect, a method comprises: updating current values of a set of encoder parameters and current values of a set of decoder parameters using gradients of a reconstruction loss function that measures an error in a reconstruction of multi-modal data from a training example, wherein: the reconstruction loss function comprises a plurality of scaling factors that each scale a respective term in the reconstruction loss function that measures an error in the reconstruction of a corresponding proper subset of feature dimensions of the multi-modal data from the training example.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating multi-modal data archetypes. In one aspect, a method comprises obtaining a plurality of training examples, wherein each training example corresponds to a respective patient and includes multi-modal data, having a plurality of feature dimensions, that characterizes the patient; jointly training an encoder neural network and a decoder neural network on the plurality of training examples; and generating a plurality of multi-modal data archetypes that each correspond to a respective dimension of a latent space, comprising, for each multi-modal data archetype: processing a predefined embedding that represents the corresponding dimension of the latent space using the decoder neural network to generate multi-modal data, having the plurality of feature dimensions, that defines the multi-modal data archetype.

A computational methodology for noninvasively assessing the severity of arterial stenosis and predicting the therapeutic outcome of interventional treatment for stenosis assessed as severe, mild, or in between based on patient's CT/MRI imaging data, ultrasound test data, and physio-pathological material properties. The method includes two major parts. The steps in the first part comprise receiving medical data, segmenting the anatomical three-dimensional geometry of the stenosed artery, setting up boundary conditions at inlet and outlets using the ultrasound velocity waveforms together with 3-element WinKessel model, and computing pulsatile pressure waveforms proximal and distal to the existing stenosis for TPI. The steps in the second part comprise of varying the VR of the stenosis virtually from 0% to 95% with an increment of 5%, computing TPI for each level of VR, establishing the functional relation between TPI and VR, identifying the two thresholds of VR.sub.mild and VR.sub.severe on TPI-VR curve, determining the severity of the existing stenosis by comparing VR.sub.existing with VR.sub.mild and VR.sub.severe concurrently and predicting the outcome of the lesion (TPI) improvement after an interventional treatment such as stenting for the existing stenosis.

Methods for characterizing fluids from a patient. A time series of images of a conduit are received, and a conduit image region in the images is identified. A flow type of the fluids passing through the conduit may be classified as one of air, laminar liquid, and turbulent liquid by evaluating an air-liquid boundary of the fluid. A volumetric flow rate of the fluids in the conduit is estimated. The volumetric flow rate may be based on the classified flow type. A concentration of a blood component of the fluids passing through the conduit may be estimated based on the images. A proportion of the fluid that is blood may also be determined, and a volume of blood that has passed through the conduit within a predetermined period of time may be estimated based on the estimated total volumetric flow rate and the determined proportion.

Systems and methods are disclosed for determining anatomy directly from raw medical acquisitions using a machine learning system. One method includes obtaining raw medical acquisition data from transmission and collection of energy and particles traveling through and originating from bodies of one or more individuals; obtaining a parameterized model associated with anatomy of each of the one or more individuals; determining one or more parameters for the parameterized model, wherein the parameters are associated with the raw medical acquisition data; training a machine learning system to predict one or more values for each of the determined parameters of the parametrized model, based on the raw medical acquisition data; acquiring a medical acquisition for a selected patient; and using the trained machine learning system to determine a parameter value for a patient-specific parameterized model of the patient.

In some embodiments, an apparatus for imaging blood within a target region of tissue includes an imaging device configured to output image data associated with light received by the imaging device having a first and second spectral ranges, wherein the absorptivity by blood of light having the first spectral range is less than the absorptivity by blood of light having the second spectral range, and a controlling element configured to capture the image data associated with light received by the imaging device and to process the captured image data associated with light having the first spectral range and the captured image data associated with light having the second spectral range to generate compound image data associated with an amount of blood within the target region of tissue.

The present disclosure provides systems and methods to receiving OCT or IVUS image data frames to output one or more representations of a blood vessel segment. The image data frames may be stretched and/or aligned using various windows or bins or alignment features. Arterial features, such as the calcium burden, may be detected in each of the image data frames. The arterial features may be scored. The score may be a stent under-expansion risk. The representation may include an indication of the arterial features and their respective score. The indication may be a color coded indication.

Systems and methods are disclosed for assessing the quality of medical images of at least a portion of a patient's anatomy, using a computer system. One method includes receiving one or more images of at least a portion of the patient's anatomy; determining, using a processor of the computer system, one or more image properties of the received images; performing, using a processor of the computer system, anatomic localization or modeling of at least a portion of the patient's anatomy based on the received images; obtaining an identification of one or more image characteristics associated with an anatomic feature of the patient's anatomy based on the anatomic localization or modeling; and calculating, using a processor of the computer system, an image quality score based on the one or more image properties and the one or more image characteristics.