A method of producing a patient-specific bone plate, the method comprising: (a) generating a virtual 3D patient-specific reconstructed bone model from a plurality of virtual 3D bone component part models, the plurality of virtual 3D bone component part models being associated with a respective plurality of bone component parts of a bone of a patient; (b) selecting a bone plate template most closely resembling the virtual 3D patient-specific reconstructed bone model; (c) generating a tangible 3D patient-specific reconstructed bone model from the virtual 3D patient-specific reconstructed bone model; (e) obtaining a tangible bone plate template associated with the selected bone plate template most closely resembling the virtual 3D patient-specific reconstructed bone model; and (f) forming a patient-specific bone plate by conforming the tangible bone plate template to the tangible 3D patient-specific reconstructed bone model.

A medical image processing apparatus according to an embodiment includes processing circuitry. The processing circuitry is configured: to perform registration between one contrast image and each of a plurality of mask images; to calculate a plurality of matching degrees between the plurality of mask images and the one contrast image registered with each other, on the basis of the plurality of mask images and the one contrast image registered with each other; and to determine a difference between one of the mask images corresponding to a maximum matching degree among the plurality of matching degrees and the one contrast image, as one subtraction image corresponding to the one contrast image.

Systems and methods for enhanced diagnosis of transthyretin-related cardiac amyloidosis in a subject are disclosed. The systems and methods may use both SPECT imaging data as well as an anatomical imaging data, such as computed tomography (CT) data, to produce a combined image. Within the combined image, the radiotracer uptake between two volumes of interests are compared, one of which may represent the left ventricle of the subject and the other may represent the blood pool retention of the subject. Combining the anatomical imaging data with SPECT data enables better anatomical delineation and helps in avoiding areas with coronary or lymph node calcifications and overlying soft tissue and bony pathologies.

A method for registering a three-dimensional image data record of a target region of a patient with a two-dimensional x-ray image is provided. The method includes selecting at least one rigid reference structure with an associated contour; establishing a two-dimensional gradient x-ray image and a three-dimensional gradient data record of the image data record; finding a neighborhood in the gradient x-ray image from a plurality of neighborhoods extending about test points for a plurality of contour points; establishing local two-dimensional displacement information by comparison of the contour points with the associated comparison points; establishing movement parameters of a three-dimensional movement model describing a movement of the target region between recording of the image data record and the x-ray image from the local two-dimensional displacement information; and establishing a registration transformation describing the registration by correcting the test transformation based on the movement parameters.

An extra-oral dental imaging apparatus for obtaining an image from a patient has a radiation source; a first digital imaging sensor that provides, for each of a plurality of image pixels, at least a first digital value according to a count of received photons that exceed at least a first energy threshold; a mount that supports the radiation source and the first digital imaging sensor on opposite sides of the patient's head; a computer in signal communication with the digital imaging sensor for acquiring a first two-dimensional image from the first digital imaging sensor; and a second digital imaging sensor that is alternately switched into place by the mount and that provides image data according to received radiation.

A PET/CT imaging system is provided. The imaging system includes a PET detection system having a plurality of detector rings and an axial gap between at least two adjacent detector rings within the plurality of detector rings. The imaging system includes a CT system having an x-ray generator and a CT detection system positioned within the axial gap between the at least two detector rings. The system is configured to collect PET data and CT data on the same volume of interest substantially simultaneously.

A method and apparatus is provided to iteratively reconstruct a computed tomography (CT) image using a hybrid pre-log and post-log iterative reconstruction method. A pre-log formulation is applied to values of the projection data that are less than a threshold (e.g., X-ray intensities corresponding to high absorption trajectories). The pre-log formulation has better noise modeling and better image quality for reconstructed images, but is slow to converge. Projection data values above the threshold are processed using a post-log formulation, which has fast convergence but poorer noise handling. However, the poorer noise handling has little effect on high value projection data. Thus, the hybrid pre-log and post-log method provides improved image quality by more accurately modeling the noise of low count projection data, without sacrificing the fast convergence of the post-log method, which is applied to high-count projection data.

The present invention relates to deep learning implementations for medical imaging. More particularly, the present invention relates to a method and system for indicating whether additional medical tests are required after analysing an initial medical screening, in substantially real-time. Aspects and/or embodiments seek to provide a method and system for recommending additional medical tests, in substantially real-time, based on analysing an initial medical scan, with the use of deep learning.

Apparatus and methods are described including, using a computer processor (28), automatically identifying whether a given pixel (111) within an image corresponds to a portion of an object. A set of concentric circles (132a-c) that are disposed around the pixel are sampled, and a first function is applied to each of the circles such that the circles are defined by a first set of rotationally invariant descriptors. A second function is applied to the set of circles to generate a second set of descriptors, each of which represents a difference between respective pairs of the circles. A third function is applied such that the second set of descriptors becomes rotationally invariant. The processor identifies whether the given pixel corresponds to the portion of the object, based upon the first and second sets of rotationally invariant descriptors. Other applications are also described.

The invention relates to an orthodontic diagnostic method wherein at least one initial two-dimensional X-ray image (1) of a first zone (2) of a head (3) is taken. Then at least one three-dimensional X-ray image (4) of a second zone (5) of a dental situation is taken, and the three-dimensional X-ray image (4) is combined with the initial two-dimensional X-ray image (1) using a registration process in order to obtain a full image (8).