Abstract: A data generation device includes one or more processors. The processors input input data into a neural network and obtain an inference result of the neural network The processors calculate a first loss and a second loss. The first loss becomes smaller in value as a degree of matching between the inference result and a target label becomes larger. The target label indicates a correct answer of the inference. The second loss is a loss based on a contribution degree to the inference result of a plurality of elements included in the input data and the target label. The processors update the input data based on the first loss and the second loss.

Abstract: A learning device includes one or more processors. The processors acquire input data and a target label indicating a correct answer of inference based on the input data. The processors add noise to at least one of the input data and intermediate layer data of the neural network and perform inference by the neural network with respect to the input data. The noise is based on contributions of a plurality of elements included in the input data with respect to an inference result when the input data is input to a neural network. The processors update parameters of the neural network so that the inference result by the neural network matches the target label.

Abstract: According to one embodiment, a magnetic resonance imaging apparatus includes processing circuitry. The processing circuitry estimates transmission inhomogeneity caused in a transmit RF magnetic field from a first image based on a first signal received by a whole-body coil, and estimates reception inhomogeneity caused in a receive RF magnetic field from the first image and a second image based on a second signal received by a surface coil. The processing circuitry generates a third image, having a resolution higher than a resolution of the first image and a resolution of the second image, based on a third signal received by the surface coil. The processing circuitry corrects the third image by using the estimated transmission inhomogeneity and reception inhomogeneity.

Abstract: A magnetic resonance imaging apparatus acquires first imaging data of a three-dimensional image including a heart, having a plurality of two-dimensional first imaging area data superimposed one on top of another in parallel, and having a resolution at least in one direction different from a resolution in two other directions. A first axis is detected expressed in three dimensions relating to the heart from the three-dimensional first imaging data. A first vector is calculated passing through the first axis and having at least a predetermined resolution. First image data is generated on a plane passing through the first axis and the first vector from the first imaging data and a second axis is detected relating to the heart from the first image data, the second axis being a higher precision axis.

Abstract: A magnetic resonance imaging apparatus according to embodiments includes sequence control circuitry and processing circuitry. The sequence control circuitry performs first imaging and second imaging to a subject. The processing circuitry detects a state of a setting of the subject by using a magnetic resonance image acquired from the first imaging, and causes a display to display supporting information that supports the setting of the subject based on information defected. The sequence control circuitry performs the second imaging on the subject after the supporting information is displayed.

Abstract: An X-ray computed tomography (CT) apparatus according to an embodiment includes an X-ray generator, an X-ray detector and processing circuitry. The X-ray generator irradiates X-rays to a subject. The X-ray detector detects the X-rays that have passed through the subject. The processing circuitry calculates an estimated spectrum based on an irradiation spectrum, an estimated length and information indicating a distortion of a spectrum occurring in a path of the X-rays passing through the subject, the estimated length representing an estimated value of an X-ray transmission length of a material of decomposition target. The processing circuitry determines an X-ray transmission length of the material of decomposition target based on the estimated spectrum and a detected spectrum that is a spectrum after the X-rays have passed through the subject and that is detected by the X-ray detector.

Abstract: According to one embodiment, a magnetic resonance imaging apparatus includes processing circuitry. The processing circuitry acquires a first resonance frequency distribution of a first tissue and a second resonance frequency distribution of a second tissue which is different from the first tissue. The processing circuitry calculates a center frequency of a frequency-selective pulse that suppresses or emphasizes either one of the first tissue and the second tissue in accordance with the first and second resonance frequency distributions. The processing circuitry collects a magnetic resonance signal after the frequency-selective pulse is applied at the calculated center frequency.

Abstract: According to one embodiment, a magnetic resonance imaging apparatus includes processing circuitry. The processing circuitry calculates a static magnetic field correction amount based on a static magnetic field distribution of a first imaging range narrower than a second imaging range. The processing circuitry collects a magnetic resonance (MR) image of the second imaging range under a static magnetic field which is corrected based on the static magnetic field correction amount. The processing circuitry corrects distortion of the collected the MR image.

Abstract: A photon counting X-ray CT apparatus according to an embodiment includes: data acquiring circuitry, and processing circuitry. The data acquiring circuitry is configured to allocate energy measured by signals output from a photon counting detector in response to incidence of X-ray photons to any of a plurality of first energy bins so as to acquire a first data group as count data of each of the first energy bins. The processing circuitry is configured to determine a plurality of second energy bins obtained by grouping the first energy bins in accordance with a decomposition target material that is a material to be decomposed in a imaging region, allocate the first data group to any of the second energy bins so as to generate a second data group, and use the second data group to generate an image representing a distribution of the decomposition target material.

Abstract: According to an embodiment, a learning method of optimizing a neural network, includes updating and specifying. In the updating, each of a plurality of weight coefficients included in the neural network is updated so that an objective function obtained by adding a basic loss function and an L2 regularization term multiplied by a regularization strength is minimized. In the specifying, an inactive node and an inactive channel are specified among a plurality of nodes and a plurality of channels included in the neural network.

Abstract: A magnetic resonance imaging apparatus according to an embodiment includes sequence control circuitry and processing circuitry. The sequence control circuitry conducts, on a subject, first imaging and second imaging that is subsequent to the first imaging. The processing circuitry estimates, based on a magnetic resonance image related to the first imaging and an imaging condition set with regard to the second imaging, information about an image quality in a case in which the second imaging is conducted. The processing circuitry presents, on a display, an estimation result, superimposing the estimation result on the magnetic resonance image. The processing circuitry receives a designation operation on the magnetic resonance image from an operator, and changes a setting of the imaging condition related to the second imaging based on the designation operation.

Abstract: According to one embodiment, an MRI apparatus includes imaging control circuitry that performs shimming imaging for collecting a first MR signal, and multi-slice imaging for collecting a second MR signal along with radiation of a non-region-selective prepulse, and processing circuitry that generates static magnetic field distributions of the slices, determines a first center frequency of an RF pulse corresponding to each slice and a second center frequency of the prepulse based on the static magnetic field distribution, and determines an order of slices for collecting the second MR signal in accordance with the first and/or second center frequencies, wherein the imaging control circuitry performs the multi-slice imaging in accordance with the order and the first and second center frequencies.

Abstract: According to one embodiment, a magnetic resonance imaging apparatus includes an interface, processing circuitry, and imaging control circuitry. The interface inputs, to a locator image, a position-indicating region indicating a position in a displayed cross section. The processing circuitry determines a collection direction relating to multi-slice imaging based on a static magnetic field distribution relating to the position-indicating region and said position-indicating region. The imaging control circuitry performs the multi-slice imaging in the collection direction to a plurality of slices in an imaging region which includes at least the position-indicating region.

Abstract: According to one embodiment, a magnetic resonance imaging apparatus includes processing circuitry calculating 0-order shimming value for correcting 0-order components of inhomogeneity of a static magnetic field of a collection region in multi-slice collection for each of slices in the collection region, first-order shimming values for correcting first-order components of the inhomogeneity for each of the slices in the collection region, and multiple-order shimming values for correcting second or higher-order components of the inhomogeneity over the entire of the collection region, by using a distribution of the static magnetic field in the collection region, and imaging control circuitry performing the multi-slice collection to the collection region by using the 0-order, first-order, and multiple-order shimming values.

Abstract: An image processing apparatus according to an embodiment includes processing circuitry. The processing circuitry is configured to detect a region of body fluid flowing in a subject from time-series images acquired by scanning an imaging area including a tagged region to which a tagging pulse is applied and imaging the imaging area; generate a plurality of display images in which the detected body fluid region is displayed in a display mode determined based on a positional relation between the body fluid region and a boundary line, the boundary line determined based on the tagged region; and output time-series display images including the plurality of display images to be displayed on a display.

Abstract: According to one embodiment, a magnetic resonance imaging apparatus includes processing circuitry. The processing circuitry estimates transmission inhomogeneity caused in a transmit RF magnetic field from a first image based on a first signal received by a whole-body coil, and estimates reception inhomogeneity caused in a receive RF magnetic field from the first image and a second image based on a second signal received by a surface coil. The processing circuitry generates a third image, having a resolution higher than a resolution of the first image and a resolution of the second image, based on a third signal received by the surface coil. The processing circuitry corrects the third image by using the estimated transmission inhomogeneity and reception inhomogeneity.

Abstract: A photon counting X-ray CT apparatus according to an embodiment includes: data acquiring circuitry, and processing circuitry. The data acquiring circuitry is configured to allocate energy measured by signals output from a photon counting detector in response to incidence of X-ray photons to any of a plurality of first energy bins so as to acquire a first data group as count data of each of the first energy bins. The processing circuitry is configured to determine a plurality of second energy bins obtained by grouping the first energy bins in accordance with a decomposition target material that is a material to be decomposed in a imaging region, allocate the first data group to any of the second energy bins so as to generate a second data group, and use the second data group to generate an image representing a distribution of the decomposition target material.

Abstract: According to one embodiment, a medical signal processing apparatus includes processing circuitry. The processing circuitry inputs a medical signal to a learned model configured to output one of the following: a corrected signal that is corrected so as to reduce a pattern of the medical signal, the pattern appearing at a location shifted by a known shift amount in a known direction; pattern-related information relating to the pattern; or disease information relating to the medical signal. The processing circuitry outputs, by using the direction and the shift amount, one of the corrected signal, the pattern-related information, or the disease information.

Abstract: A magnetic resonance imaging apparatus according to an embodiment includes processing circuitry. The processing circuitry generates a plurality of cross-sectional images for setting a sectional position to be collected in main imaging based on a characteristic portion of a target detected in three-dimensional data. The processing circuitry lists the cross-sectional images on a display and superimposes a mark corresponding to the characteristic portion on at least one of the cross-sectional images. The processing circuitry receives a setting operation to determine the sectional position. The processing circuitry causes, when the mark is selected in the setting operation, a cross-sectional image to be emphasized a sectional position of which is defined using the characteristic portion corresponding to the mark among the listed cross-sectional images. The processing circuitry performs main imaging based on the sectional position after the setting operation.

Abstract: A magnetic resonance imaging apparatus according to an embodiment includes processing circuitry. The processing circuitry generates a plurality of cross-sectional images for setting a sectional position to be collected in main imaging based on a characteristic portion of a target detected in three-dimensional data. The processing circuitry lists the cross-sectional images on a display and superimposes a mark corresponding to the characteristic portion on at least one of the cross-sectional images. The processing circuitry receives a setting operation to determine the sectional position. The processing circuitry causes, when the mark is selected in the setting operation, a cross-sectional image to be emphasized a sectional position of which is defined using the characteristic portion corresponding to the mark among the listed cross-sectional images. The processing circuitry performs main imaging based on the sectional position after the setting operation.