This invention describes methods to compute coronary physiology indexes using a high precision registration model, which consists of acquiring coronary angiography images of coronary vessels, performing intravascular imaging, and registering the coronary angiography images with intravascular images to create a high precision registration model, based upon which the coronary flow, fractional flow reserve (FFR) and index of microcirculation resistance (IMR) can be computed. The methods described in this invention to compute coronary flow, FFR, IMR are based on both coronary angiography and intravascular images, and the accuracy is better than those derived from coronary angiography alone or intravascular imaging alone, and have high practical values.

A method of processing a 3D OCT dataset is present. A method according to some embodiments of the present invention includes obtaining an OCT data from the 3D OCT dataset; obtaining an OCTA data from the 3D OCT dataset; performing segmentation for at least one boundary on the OCT data; processing the OCTA data in a region of interest to create at least one image representation by assigning a value to each pixel of each of the image representation; and displaying at least one image representation.

Systems and methods are provided for assessing patient blood flow using video image processing. According to one aspect, a method of analyzing at least one blood flow characteristic of a patient includes capturing a video including a plurality of frames of an arterio-venous (AV) fistula on the patient; amplifying motion in the video to produce a motion-amplified video; determining a difference in intensity between consecutive frames in the motion-amplified video to produce a time-function of an amplitude of the optic flow representing movement in an area of interest on the patient; and determining the at least one blood flow characteristic of the patient based on the time-function.

A method and system for determining fractional flow reserve (FFR) for a coronary artery stenosis of a patient is disclosed. In one embodiment, medical image data of the patient including the stenosis is received, a set of features for the stenosis is extracted from the medical image data of the patient, and an FFR value for the stenosis is determined based on the extracted set of features using a trained machine-learning based mapping. In another embodiment, a medical image of the patient including the stenosis of interest is received, image patches corresponding to the stenosis of interest and a coronary tree of the patient are detected, an FFR value for the stenosis of interest is determined using a trained deep neural network regressor applied directly to the detected image patches.

Provided is a highly reliable technique for the evaluation of ischemic conditions. A preferred embodiment is a nuclear medicine measurement protocol in which the administration of a radiopharmaceutical agent and radiation measurement are performed twice at rest and under stress. In the nuclear medicine measurement protocol, radiation collection is performed without radiopharmaceutical agent administration before the second radiopharmaceutical agent administration, and the result is used to correct the nuclear medicine measurement result after the second radiopharmaceutical agent administration.

Devices, systems, and methods for evaluating blood flow with vascular perfusion imaging are disclosed. In an embodiment, a medical system is disclosed. One embodiment of the medical system comprises a perfusion imaging system configured to obtain perfusion imaging data associated with movement of contrast through a vessel of a patient, a graphical user interface, and a medical processing unit in communication with the perfusion imaging system and the graphical user interface. The medical processing unit is configured to receive a first set of perfusion imaging data from the perfusion imaging system, determine at least one parameter representative of the movement of the contrast through the vessel of the patient, generate a first graphical representation of the first set of perfusion imaging data and the at least one parameter determined based on the first set of perfusion imaging data, and output the first graphical representation to the graphical user interface.

Embodiments of the invention provide a method, system and computer program product for artificially intelligent ejection fraction determination. In a method for artificially intelligent ejection fraction determination, a neural network is loaded into memory of a computer, that has been trained with different sets of cardiac imaging data acquired during imaging of a ventricle for different hearts and a known ejection fraction for each of the sets. Then, a contemporaneous set of imaging data is acquired of a ventricle of a heart and the contemporaneous set of imaging data is provided to the neural network. Finally, an ejection fraction determination output by the neural network is displayed in a display of the computer without tracing a ventricle boundary of the heart.

Methods and systems are provided for generating quantitative computed tomography (CT) angiographic images using imaging systems that acquire a set of projection views forming CT angiographic image data of a patient using a projection duration of less than 50 milliseconds. The method may include producing a composite image from the CT angiographic image data that indicates an attenuation value at each composite image pixel of the patient, backprojecting each projection view in the CT angiographic image data and weighting a value backprojected into at image pixel by an attenuation value of a corresponding pixel in the composite image and summing backprojected values for each image pixel to produce a CT image of the patient. The method may further include performing a scatter correction and determining at least one of a flow direction or a velocity of flow with a vessel in the patient to provide the quantitative CT angiographic images.

According to one embodiment, a structuring circuitry temporarily structures a dynamical model of analysis processing based on the time-series medical image. The identification circuitry identifies a latent variable of the dynamical model so that at least one of a prediction value of a blood vessel morphology and a prediction value of a bloodstream based on the temporarily structured dynamical model is in conformity with at least one of an observation value of the blood vessel morphology and an observation value of the bloodstream measured in advance. The analysis circuitry analyzes the dynamical model to which the identified latent variable is allocated.

The present disclosure is directed to a method and system for automatically predicting a blood flow feature based on a medical image. The method may include acquiring, by a processor, image patches and a vessel related feature of a vessel tree. Then, the blood flow feature of the vessel tree may be calculated, by the processor, using a learning network based on both the image patches and the vessel related feature of the vessel tree. The learning network includes a multi-model neural network and a tree structure recurrent neural network connected in series. The method and system of present disclosure can perform a quick and accurate prediction for the blood flow feature, such as FFR, of the vessel tree of a target object (such as certain site of human body or animal body) based on both the medical images and vessel related features of the vessel tree of the target object. The predicted FFR may assist the user in pathological diagnosis or other treatment of the target object.