Abstract: Various methods and systems are provided for training a denoising system for a digital imaging system. The denoising system can be a deep learning denoising system formed as a blind or non-blind denoising system in which the training dataset provided to the denoising system includes a noisy image formed with simulated noise added to a clean digital image, and a reference image formed of the clean image having residual noise added thereto, where the residual noise is a fraction of the simulated noise used to form the noisy image. The use of the residual noise within the reference image of the training dataset teaches the DL network in the training process to remove less than all the noise during subsequent inferencing of digital images from the digital imaging system. By leaving selected amounts of noise in the digital images, the denoiser can be tuned to improve image attributes and texture.

Abstract: This disclosure proposes to speed up computation time of a convolutional neural network (CNN) by leveraging information specific to a pre-defined region, such as a breast in mammography and tomosynthesis data. In an exemplary embodiment, a method for an image processing system is provided, comprising, generating an output of a trained convolutional neural network (CNN) of the image processing system based on an input image, including a pre-defined region of the input image as an additional input into at least one of a convolutional layer and a fully connected layer of the CNN to limit computations to input image data inside the pre-defined region; and storing the output and/or displaying the output on a display device.

Abstract: Various methods and systems are provided for denoising images. In one example, a method includes obtaining an input image and a noise map representing noise in the input image, generating, from the noise map and based on a calibration factor, a strength map, entering the input image and the strength map as input to a denoising model trained to output a denoised image based on the input image and the strength map, and displaying and/or saving the denoised image output by the denoising model.

Abstract: Systems/techniques that facilitate improved neural network inferencing efficiency with fewer parameters are provided. In various embodiments, a system can access a medical image on which an artificial intelligence task is to be performed. In various aspects, the system can facilitate the artificial intelligence task by executing a neural network pipeline on the medical image, thereby yielding an artificial intelligence task output that corresponds to the medical image. In various instances, the neural network pipeline can include respective skip connections from the medical image, prior to any convolutions, to each convolutional layer in the neural network pipeline.

Abstract: An image processing system and method is provided for correcting artefacts within a three-dimensional (3D) volume reconstructed from a plurality of two-dimensional (2D) projection images of an object. The system and method is implemented on an imaging system having a processing unit operable to control the operation of a radiation source and a detector to generate a plurality of 2D projection images. The system also includes a memory connected to the processing unit and storing processor-executable code that when executed by the processing unit operates to reconstruct the 3D volume from the plurality of 2D projection images, the 3D volume defined in a pseudo parallel geometry based on a zero angle from the plurality of 2D projection images, wherein reconstructing the 3D volume comprises reconstructing a 3D virtual object defined in a pseudo parallel geometry based on the zero angle from the plurality of 2D projection images, and correcting the 3D virtual object to form the 3D volume.

Abstract: A tomosynthesis machine allows for faster image acquisition and improved signal-to-noise by acquiring a projection attenuation data and using machine learning to identify a subset of the projection attenuation data for the production of thinner slices and/or higher resolution slices using machine learning.

Abstract: A convolutional neural network (CNN) is employed in an automated feature and/or anomaly detection system for analyzing images provided by X-ray imaging system. The automated detection system operates in a manner that reduces the number of full-resolution CNN convolution layers required in order to speed up the network inference and learning processes for the detection system. To do so, the detection system utilizes as an input a more compact representation of the tomographic data to alleviate the CNN memory footprint and computation time issues in prior art X-ray systems.

Abstract: Various systems are provided for non-uniform thickness and/or sampling of slabs of the breast to present DBT acquisitions. A method for generating a patient image as a set of slabs representing an imaged object, the method comprising acquiring a tomosynthesis projection, reconstructing a series of slab images, each slab representing a portion of a breast, and a plurality of slabs of non-uniform thickness and/or non-uniform sampling in a 3D reconstructed domain defined by x-, y-, and z-axes.

Abstract: This disclosure proposes to speed up computation time of a convolutional neural network (CNN) by leveraging information specific to a pre-defined region, such as a breast in mammography and tomosynthesis data. In an exemplary embodiment, a method for an image processing system is provided, comprising, generating an output of a trained convolutional neural network (CNN) of the image processing system based on an input image, including a pre-defined region of the input image as an additional input into at least one of a convolutional layer and a fully connected layer of the CNN to limit computations to input image data inside the pre-defined region; and storing the output and/or displaying the output on a display device.

Abstract: The present invention pertains to a system for projecting a pattern of interest onto a retinal area of a human eye, containing a carrying frame for being worn by a patient, a camera for capturing an image, a projector device for projecting a pulsed light beam reflecting a pattern of interest into a human eye, and a processor device being in communication with the camera and the projector device, wherein the processor device is adapted for converting the image captured by the camera into the pattern of interest being basis for the pulsed light beam, wherein the projector device and the camera are attached to the carrying frame.

Abstract: An image processing method applied by a medical imaging device is proposed, including at least one camera and at least one display screen on which is displayed a medical image showing a region of interest of a patient, the method being characterized in that the region of interest is determined by the following steps, the screen and the camera being laid out so as to be facing a practitioner simultaneously, acquisition by the camera of at least one tracking image containing the eyes of the practitioner, analysis of the tracking image so as to localize in the medical image at least one pixel of interest contemplated by the practitioner on the screen from the orientation of the eyes of the practitioner, processing the medical image so as to select the region of interest from the localized pixel of interest.

Abstract: An image processing method applied by a medical imaging device is proposed, including at least one camera and at least one display screen on which is displayed a medical image showing a region of interest of a patient, the method being characterized in that the region of interest is determined by the following steps, the screen and the camera being laid out so as to be facing a practitioner simultaneously, acquisition by the camera of at least one tracking image containing the eyes of the practitioner, analysis of the tracking image so as to localize in the medical image at least one pixel of interest contemplated by the practitioner on the screen from the orientation of the eyes of the practitioner, processing the medical image so as to select the region of interest from the localized pixel of interest.

Abstract: A method is provided for monitoring the radiation dose in a body to be imaged using an X-ray imaging device comprising an X-ray emitting source, a detector, a processing unit and a display. The method comprises exposing the body to be imaged to radiation doses to acquire initial 2D images; computing a 3D model of the body in relation to the initial 2D images using the processing unit; applying a model of the interactions between matter and radiation to the 3D model of the body using the processing unit; calculating, using the processing unit, a dose map of the distribution of the accumulated radiation dose in the 3D model for parameters that define the conditions of X-ray exposure; and displaying the 3D model of the body oriented according to the position of the emitting source with the dose map.

Abstract: A method for monitoring a radiation dose applied to a patient being exposed or having been exposed to radiation from a medical imaging system, the medical imaging system comprising an X-ray source, the X-ray source describing a trajectory in a coordinate system of reference of the medical imaging system having an origin, the method comprising determining the current coordinates, in the coordinate system of reference of the medical imaging system, of a current region being exposed or having been exposed to radiation from the X-ray source, of an envelope of a patient model, in relation to a current position of the X-ray source, determining, from the current coordinates, a current coordinate system of reference having as origin the current position of the X-ray source, the current coordinate system of reference being centered on the current region, and displaying a representation of the current region centered on the current coordinate system of reference.

Abstract: An image processing method for interventional imaging in which a region of interest of a patient is viewed. The method comprises acquiring a succession of images of a region of interest of the patient. The method also comprises detecting and tracking, on the successive images, at least one surgical instrument introduced inside the region of interest of the patient, in order to isolate said instrument therein; and comparing two successive images on which the surgical instrument has been isolated in order to identify at least one common shape therein. The method further comprises estimating the displacement of said common shape between both of these successive images; and re-alignment processing of the different successive images depending on the thereby determined estimations of displacements, these displacement estimations being considered as corresponding to the displacement caused by the physiological movement of the patient with the exception of any other movement.

Abstract: A method to process radiological images is provided. The method comprises partitioning a radiological image of a region to be treated into a superimposition of layers, the region to be treated comprising at least one first structure and a second structure, wherein one layer solely comprises part of the first structure to be isolated from the remainder of the image, the layer solely comprising that part of the first structure to be isolated from the remainder of the image being determined by means of a parametric model of the first structure. The method further comprises determining an image of the region to be treated from the layering thus obtained, in which the isolated part of the first structure is omitted.

Abstract: The disclosure generally relates to dual-energy imaging, and in particular, techniques to produce and process dual-energy images using a dual-energy imaging system. One embodiment provides a method for generating at least one image of a region of interest in a patient, the method comprising: obtaining at least two radiological images of the region of interest identified with at least one marker arranged on and/or around the patient, wherein a first image is acquired with a first X-ray energy and a second image is acquired with a second X-ray energy; and determining a final radiological image of the region of interest by linearly combining the two radiological images to obtain an image without the markers.

Abstract: A treatment process of radiological images is provided. The process comprises obtaining at least one set of images. For each set, the process comprises: segmenting an at least one first image to obtain an at least one first segmented image and to detect a plurality of arteries of the region of interest and an at least one second image to obtain an at least one second segmented image and to detect and isolate the tool; defining in the at least one first segmented image a plurality of lines, wherein each line defines an artery; determining, from the second segmented image and the defined lines, an artery of interest corresponding to the artery in which a tool has been inserted; and applying a quantitative analysis algorithm of coronary lesions to the artery of interest to detect lesion of the artery of interest.

Abstract: A method for controlling the emission of an X-ray imaging device configured to take images of a patient's body, into which a medical instrument has been inserted, is provided. The method comprises: processing at least one image of the patient's body taken by the X-ray imaging device to extract information representing the instrument, wherein the at least one image comprises the instrument; and adapting X-ray emission parameters of the X-ray imaging device, depending on the information extracted from the at least one image, to minimize the X-ray dose emitted towards the patient's body.

Abstract: A method for processing radiological images is provided. The method includes extracting at least two image portions from at least two radiological images of an area to be treated. The method also includes defining an area in a final image for each extracted image portion. The method further includes laying out the extracted image portions in the final image so that each area of the final image comprises an extracted image portion.