An image processing method and related system to register projection images (AG, MI) only with respect to a motion of a landmark across said images. The motion of the landmark relates to a motion of a region of interest, ROI. The so registered images (AG, MI) are then subtracted from each other to arrive at a difference image that is locally motion compensated and that represents the ROI at good contrast whilst subtraction artifacts can be avoided.

Systems and methods are provided for performing optical coherence tomography angiography for the rapid generation of en face images. According to one example embodiment, differential interferograms obtained using a spectral domain or swept source optical coherence tomography system are convolved with a Gabor filter, where the Gabor filter is computed according to an estimated surface depth of the tissue surface. The Gabot-convolved differential interferogram is processed to produce an en face image, without requiring the performing of a fast Fourier transform and k-space resampling. In another example embodiment, two interferograms are separately convolved with a Gabor filter, and the amplitudes of the Gabor-convolved interferograms are subtracted to generate a differential Gabor-convolved interferogram amplitude frame, which is then further processed to generate an en face image in the absence of performing a fast Fourier transform and k-space resampling. The example OCTA methods disclosed herein are shown to

The invention relates to a method for identifying an ischaemic region (O.sub.n) of an organ based on anatomical data, wherein the ischaemic region (O.sub.n) is 0.2 to 1 part of the stenosed region at risk (O.sub.z) downstream of the threshold point (P.sub.prog). The size of the ischaemic region (O.sub.n) is proportional to the difference between the indicative value at the threshold point (P.sub.prog) and at the measuring point (P.sub.pom) in the artery. The invention also relates to a system for identifying organ ischaemia, a computer program for identifying organ ischaemia and a computer program product.

Anatomical and functional assessment of coronary artery disease (CAD) using machine learning and computational modeling techniques deploying methodologies for non-invasive Fractional Flow Reserve (FFR) quantification based on angiographically derived anatomy and hemodynamics data, relying on machine learning algorithms for image segmentation and flow assessment, and relying on accurate physics-based computational fluid dynamics (CFD) simulation for computation of the FFR.

An example method includes generating, using a multi-scale block of a convolutional neural network (CNN), a first output image based on an optical coherence tomography (OCT) reflectance image of a retina and an OCT angiography (OCTA) image of the retina. The method further includes generating, using an encoder of the CNN, at least one second output image based on the first output image and generating, using a decoder of the CNN, a third output image based on the at least one second output image. An avascular map is generated based on the third output image. The avascular map indicates at least one avascular area of the retina depicted in the OCTA image.

An image processing apparatus includes an information obtaining unit configured to obtain three-dimensional polarization sensitive tomographic information and three-dimensional motion contrast information about a subject based on tomographic signals of lights having different polarizations, the lights being obtained by splitting a combined light obtained by combining a returned light from the subject illuminated with a measurement light with a reference light corresponding to the measurement light, an obtaining unit configured to obtain a lesion region of the subject using the three-dimensional polarization sensitive tomographic information, and an image generation unit configured to generate an image in which the lesion region is superimposed on a motion contrast image generated using the three-dimensional motion contrast information.

A medical image processing apparatus includes a port, a processor and a display. The port acquires volume data including a subject. The processor generates an image based on the volume data. The display shows the generated image. A pixel value of at least one pixel of the image is defined based on (i) a statistical value of voxel values of voxels in a predetermined range on a virtual ray projected to the volume data and (ii) shading of a contour of the subject at a predetermined position on the virtual ray.

A storage unit stores volume data including a blood vessel region. An extracting unit extracts the blood vessel region from the volume data. A specifying unit specifies a position of a region of interest in the blood vessel region and a deflection direction of a blood vessel region included in the region of interest. A determining unit determines a viewing position and a viewing direction based on the position of the region of interest and the deflection direction. A generating unit generates image data concerning the viewing position and the viewing direction based on the volume data. A display unit displays an image represented by the image data.

In a method and apparatus, magnetic resonance angiography images of an examination volume of a patient are obtained using time-of-flight angiography in a magnetic resonance scanner. By continuous recording, a number of two-dimensional slice images covering the examination volume along an axial direction are acquired in a slice-by-slice layer-wise, such as with overlapping. The slice images are divided into groups of, in each case, a predetermined number of consecutive slice images in the axial direction. A maximum intensity projection image is determined for each group, and the angiography images are determined as the maximum intensity projection images and/or dependent on the maximum intensity projection images.

A system and method are provided for medical imaging that includes acquiring, during a common imaging acquisition process, rotational, x-ray volume image data and x-ray tomosynthesis image data from a subject. The method includes reconstructing a time-resolved three-dimensional (3D) image volume from the rotational, x-ray volume image data and producing a four-dimensional (4D) image series of the subject with resolved overlapping features by selectively combining the time-resolved 3D image volume and the x-ray tomosynthesis imaging data.