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.

A framework for registering anatomical to functional images using deep learning. In accordance with one aspect, the framework extracts features by applying an anatomical image and a corresponding functional image as input to a first trained convolutional neural network. A deformation field is estimated by applying the extracted features as input to a second trained convolutional neural network. The deformation field may then be applied to the anatomical image to generate a registered anatomical image.

An image processing apparatus according to an embodiment includes a processor. The processor acquires medical image data. The processor estimates the structure of a target organ depicted in the medical image data. The processor determines, for each region of the target organ, the degree of certainty representing the accuracy of the estimation result of the structure of the target organ depicted in the medical image data. The processor causes at least information indicating whether user's editing based on the degree of certainty is necessary or information representing a region where user's editing is restricted to be displayed on a display image based on the medical image data.

Embodiments include methods and systems for determining a sensitivity of a patient's blood flow characteristic to anatomical or geometrical uncertainty. For each of one or more of individuals, a sensitivity of a blood flow characteristic may be obtained for one or more uncertain parameters. An algorithm may be trained based on the sensitivities of the blood flow characteristic and one or more of the uncertain parameters for each of the plurality of individuals. A geometric model, a blood flow characteristic, and one or more of the uncertain parameters of at least part of the patient's vascular system may be obtained for a patient. The sensitivity of the patient's blood flow characteristic to one or more of the uncertain parameters may be calculated by executing the algorithm on the blood flow characteristic of at least part of the patient's vascular system, and one or more of the uncertain parameters.

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.

A method and apparatus for vascular assessment are disclosed. The apparatus, in some embodiments, receives a plurality of 2-D angiographic images of a portion of a vasculature of a subject, and processes the images to produce a stenotic model over the vasculature. The stenotic model has measurements of the vasculature at one or more locations along vessels of the vasculature. The apparatus, in some embodiments, determines a flow characteristic of the stenotic model and calculates an index indicative of vascular function, based, at least in part, on the flow characteristic in the stenotic model.

The systems and methods can accurately and efficiently determine a myocardial risk from a lesion disposed along a coronary segment using hemodynamic characteristic(s) associated with one or more sections of the corresponding lesion site. The method may include segmenting one or more lesion sites disposed along at least one arterial segment of the one or more arterial segments of the coronary model into one or more sections. Each lesion site includes a lesion. The method may include determining one or more characteristics for at least one section using at least the one or more characteristics associated with the at least one arterial segment. The one or more characteristics for the at least one section including hemodynamic force characteristic(s) (e.g., wall shear stress (WSS)). The method may include determining one or more risk indices for each lesion site using at least the hemodynamic force characteristic(s) for the at least one section.

A medical data processing device according to an embodiment includes processing circuitry. The processing circuitry acquires estimated data related to a state of a biological organ at a first timing, and actually measured data indicating the state of the biological organ at the first timing. The processing circuitry calculates a parameter based on the estimated data and the actually measured data. The processing circuitry estimates a state of the biological organ in a predetermined time phase at a second timing different from the first timing based on the parameter and the estimated data in the predetermined time phase.

Systems and methods of predicting future medical events are based on the processing of medical image. The prediction of premature birth and estimation of gestational age based on ultrasound images are presented as illustrative examples. The new abilities to estimate the probability of future medical events, before they otherwise could be predicted, provides new avenues for the development of preventative treatments.

The present invention relates to the field of medical imaging in the absence of contrast agents. In one form, the invention relates to the field of imaging vessels, particularly blood vessels such as the pulmonary vasculature and is suitable for use as a technique for detecting pulmonary embolism (PE), such as acute PE. Embodiments of the present invention provide improved image processing techniques having the capability to extract and use image data to overcome the need for contrast agents to distinguish between different types of tissue. Furthermore, it has also been realised that the image data accessed by the improved image processing can be used to identify irregularities in vessels.