An X-ray CT system and a method of processing medical images are provided that enable combining of images with reduced effect of the differences in coordinates of the pixels in the overlapped areas of a plurality of constituent images. The X-ray CT system includes a processor and a synthesizer. Based on coordinates of first pixels in a first image of a first three-dimensional region of the subject and coordinates of second pixels in a second image of a second three-dimensional region of the subject, the processor combines the first pixels with the second pixels on a one-for-one basis within a predetermined range in the rostrocaudal direction. The synthesizer generates third pixels relative to the first pixels and the second pixels and generates a third image that includes the third pixels.

The image processing device generates an operation image by adding or subtracting an original image and a standard deviation image which maps the standard deviation for the pixels configuring the original image. In this operation image, images of the structures seen in parts in the original image other than metal pieces are erased. Consequently, structures that are not metal pieces appearing in the original image in a whitish color, for example, do not appear in the operation image. If such an operation image is subjected to binarization processing in which the metal pieces appearing in a whitish color, for example, are extracted, since accurate graph cut processing is then possible, images originating from structures that are not metal pieces do not appear in the resulting image.

This application relates to a method of measuring a volume of an organ. In one aspect, the method includes acquiring a plurality of captured images of the organ and photographing metadata and preprocessing the plurality of images to acquire a plurality of image patches of a specified size. The method may also include inputting the plurality of image patches into a three-dimensional (3D) convolutional neural network (CNN)-based neural network model and estimating an organ region corresponding to each of the plurality of image patches. The method may further include measuring a volume of the organ by using an area of the estimated organ region and the photographing metadata. The method may further include measuring an uncertainty value of the 3D CNN-based neural network model and uncertainty values of the plurality of images based on a result of estimating by the 3D CNN-based neural network model.

A method and system for converting low-dose mammographic images with much noise into higher quality, less noise, higher-dose-like mammographic images, using of a trainable nonlinear regression (TNR) model with a patch-input-pixel-output scheme, which can be called a call pixel-based TNR (PTNR). An image patch is extracted from an input mammogram acquired at a reduced x-ray radiation dose (lower-dose), and pixel values in the patch are entered into the PTNR as input. The output of the PTNR is a single pixel that corresponds to a center pixel of the input image patch. The PTNR is trained with matched pairs of mammograms, inputting low-dose mammograms together with corresponding desired standard x-ray radiation dose mammograms (higher-dose), which are ideal images for the output images. Through the training, the PTNR learns to convert low-dose mammograms to high-dose-like mammograms. Once trained, the trained PTNR does not require the higher-dose mammograms anymore. When a new reduced x-ray radiation dose (low dose) mammogram is entered, the trained PTNR would output a pixel value similar to its desired pixel value, in other words, it would output high-dose-like mammograms or “virtual high-dose” mammograms where noise and artifacts due to low radiation dose are substantially reduced, i.e., a higher image quality. With the “virtual high-dose” mammograms, the detectability of lesions and clinically important findings such as masses and microcalcifications can be improved.

Provided is a dynamic analysis device that includes a hardware processor and a storage that stores setting information on image associated information and a dynamic analysis including dynamic analyses. The image associated information is linked to the dynamic analysis among predetermined dynamic analyses of multiple types. The hardware processor acquires a dynamic image to be subject to the dynamic analysis and the image associated information of the dynamic image, specifies a type of the dynamic analysis that is applied to the dynamic image based on the acquired image associated information and the setting information stored in the storage, executes the dynamic analysis of the specified type to generate an analysis result, and outputs the generated analysis result.

A method is for acquiring an X-ray image of a region of interest of an examinee using an X-ray system and a displaceable patient table for positioning the examinee. In an embodiment, the method includes: selecting the region of interest; acquiring, section-by-section, successive image sections in relation to the region of interest, the acquiring, for each successive image section of the successive image sections, including moving the X-ray source and the X-ray detector along a common acquisition direction, moving the patient table counter to the common acquisition direction, determining a respective essentially strip-shaped detection area within the detection zone for a respective image section of the successive image sections, and detecting the respective image section by way of the determined detection area and the X-ray source, to acquire the respective image section; and generating a composite X-ray image of the region of interest from the respective successive image sections.

In an X-ray generator an X-ray tube includes an anode and a cathode and is energized with at least a first high voltage potential. A dynamic damper with a frequency dependent impedance is interposed between the X-ray tube and a source of the high voltage potential. The impedance of the dynamic damper increases with an increase in frequency associated with tube-spit. In an X-ray generator with resonant switching to provide a first kV level and a second kV level to the X-ray tube, the impedance of the dynamic damper is low at the operational frequency of the resonant switch to promote energy recovery when the resonant switch operates to provide a first kV level to the X-ray tube.

A method and apparatus is provided to decompose spectral computed tomography (CT) projection data into material components using singles-counts and total-counts projection data. The singles-counts projection data more accurately solves the material decomposition problem, but can produce multiple results only one of which is correct. The total-counts projection data generates a unique result, but is less precise. The total-counts projection data is used to disambiguate the multiple results of the singles-counts projection data providing a unique results that is also precise. The unique and precise material decomposition can be achieved by limiting a search region for the singles-counts result to a neighborhood surrounding the total-counts result, choosing a singles-counts result that is closest to the total-counts result, choosing a singles-counts result that minimizes a total-counts cost function, or using a combined cost function that includes a singles-counts projection data and energy-integrated projection data.

An information processing method of an embodiment is a processing method of information acquired by imaging performed by a medical image diagnostic apparatus, the information processing method includes the steps of: on the basis of first subject data acquired by the imaging performed by the medical image diagnostic apparatus, acquiring noise data in the first subject data; on the basis of second subject data acquired by the imaging performed by a medical image diagnostic modality same kind as the medical image diagnostic apparatus and the noise data, acquiring synthesized subject data in which noises based on the noise data are added to the second subject data; and acquiring a noise reduction processing model by machine learning using the synthesized subject data and third subject data acquired by the imaging performed by the medical image diagnostic modality.

Provided are an X-ray imaging apparatus that is capable of tracking a position of an object of interest using a Kalman filter so as to reduce the amount of X-ray radiation exposure of a subject, calculating covariance indicative of accuracy of the tracing, and controlling a collimator so that the position of the object of interest and calculated covariance may be correlated with a position and an area of a region into which X-rays are radiated, and a method of controlling the X-ray imaging apparatus.