A method and apparatus for vascular assessment are disclosed. The apparatus, in some embodiments, receives, from a medical imaging device, a medical image of a coronary vessel tree of a subject and calculates a plurality of geometric measurements associated with individual portions of a vascular segment of the coronary vessel tree. The apparatus also determines a plurality of resistances associated with the plurality of geometric measurements associated with the individual portions of the vascular segment and determines a plurality of pressure drops across the individual portions of the vascular segment based on the determined resistances and a calculated or estimated blood flow. The apparatus further calculates based on the plurality of pressure drops, a functional index indicative of a presence or an absence of a stenosis within the vascular segment.

In part, the disclosure relates to methods, and systems suitable for evaluating image data from a patient on a real time or substantially real time basis using machine learning (ML) methods and systems. Systems and methods for improving diagnostic tools for end users such as cardiologists and imaging specialists using machine learning techniques applied to specific problems associated with intravascular images that have polar representations. Further, given the use of rotating probes to obtain image data for OCT, IVUS, and other imaging data, dealing with the two coordinate systems associated therewith creates challenges. The present disclosure addresses these and numerous other challenges relating to solving the problem of quickly imaging and diagnosis a patient such that stenting and other procedures may be applied during a single session in the cath lab.

A radiographic imaging apparatus (100) is configured to generate movement maps (30) of pixels (21) belonging to a first image (11) based on the first image (11) and a second image (12) captured at different times, to move a pixel (21) of the first image (11) based on a smoothed movement map (30a) in which high-frequency components of the movement maps (30) have been suppressed in a spatial direction and generate a deformed image (11a), and to combine the deformed image (11a) and the second image (12).

The invention relates to systems and methods for predicting ischemic brain tissue fate from multi-phase CT-angiography. More specifically, systems and methods are provided that enable meaningful prediction of core, penumbra and perfusion from mCTA images using software that has been trained via machine learning to interpret mCTA images.

The present approach relates to determining a reference value based on image data that includes a non-occluded vascular region (such as the ascending aorta in a cardiovascular context). This reference value is compared on a pixel-by pixel basis with the CT values observed in the other vasculature regions. With this in mind, and in a cardiovascular context, the determined FFR value for each pixel is the ratio of CT value in the vascular region of interest to the reference CT value.

Minimizing image sensor input/output pads in a pulsed fluorescence imaging system is disclosed. A system includes an emitter for emitting pulses of electromagnetic radiation and an image sensor comprising a pixel array for sensing reflected electromagnetic radiation. The system includes a plurality of bidirectional pads comprising an output state for issuing data and an input state for receiving data. The system includes a controller configured to synchronize timing of the emitter and the image sensor. The system is such that at least a portion of the pulses of electromagnetic radiation emitted by the emitter comprises electromagnetic radiation having a wavelength from about 770 nm to about 790 nm and/or from about 795 nm to about 815 nm.

A method for extracting a major vessel region from a vessel image by a processor may comprise the steps of: extracting an entire vessel region from a vessel image; extracting a major vessel region from the vessel image on the basis of a machine learning model which extracts a major vessel region; and revising the major vessel region by connecting separated vessel portions on the basis of the entire vessel region.

Systems and techniques that facilitate automated localization of large vessel occlusions are provided. In various embodiments, an input component can receive computed tomography angiogram (CTA) images of a patient's brain. In various embodiments, a localization component can determine, via a machine learning algorithm, a location of a large vessel occlusion (LVO) in the patient's brain based on the CTA images. In various instances, the location of the LVO can comprise a laterality and an occlusion site. In various aspects, the laterality can indicate a right side or a left side of the patient's brain, and the occlusion site can indicate an internal carotid artery (ICA), an M1 segment of a middle cerebral artery (MCA) or an M2 segment of an MCA. In various cases, a visualization component can generate and display to a user a three-dimensional maximum intensity projection (MIP) reconstruction of the patient's brain based on the CTA images to facilitate visual verification of the LVO by the user.

An ultrasound imaging system configured to assess a blood flow rate through a target vessel. The ultrasound imaging system includes an ultrasound probe having an ultrasound array configured to capture ultrasound image of the target vessel and a doppler array configured to detect the fluid flow through a region of interest of the target vessel. Logic operations of a console of the system and methods include determining a region of interest of the ultrasound image, calculating a percentage of the blood vessel occupied by the vascular access device, utilizing a data training set to predict a blood flow rate after placement of the vascular access device based on a blood flow rate prior to placement of the vascular device, and utilizing a data training set to predict a blood flow rate downstream of the vascular access device based on a blood flow rate upstream of the vascular device.

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.