535results about "Reconstruction from projection" patented technology

A system and method for using augmented reality device for use during a surgical procedure are described. A system may include an augmented reality device to present a virtual indication, such as a virtual indication of a surgical instrument, a force vector, a direction, or the like. The augmented reality device may present a virtual aspect of a procedure, such as a virtual animation of a step of a procedure. The augmented reality device may present a virtual object, indication, aspect, etc., within a surgical field while permitting the surgical field or aspects of the surgical field to be viewed through the augmented reality display (e.g., presenting virtual objects mixed with real objects).

A pharmaceutical hyperspectral reconstruction method based on a coded aperture snapshot spectral imaging (CASSI) system includes: collecting and processing original pharmaceutical hyperspectral images to obtain augmented pharmaceutical hyperspectral images; performing simulated spatial encoding on the augmented pharmaceutical hyperspectral images to obtain encoded measurement images; performing spectral inverse shift on the encoded measurement images, then performing inverse encoding to obtain inversely encoded three-dimensional hyperspectral images, using the augmented pharmaceutical hyperspectral images as target images, and constructing a training set and a testing set according to the inversely encoded three-dimensional hyperspectral images and the target images; constructing a deep symmetric neural reconstruction network, and training and testing the deep symmetric neural reconstruction network; and deploying a tested deep symmetric neural reconstruction network onto the CASSI system, real-time collecting pharmaceutical measurement images using the snapshot coded imaging system, and performing computational reconstruction on the pharmaceutical measurement images to obtain reconstructed three-dimensional hyperspectral images.

Described herein are systems, methods, and instrumentalities associated with using a multi-layer perceptron (MLP) neural network to process medical images of an anatomical structure. The processing may include padding an input image in accordance with the training of the MLP neural network, splitting the input image (e.g., the padded input image) into patches of a same size, and processing the patches through the MLP neural network over one or more iterations. During an iteration of the processing, the patches may be processed separately and re-combined into an intermediate image before the intermediate image is shifted to concatenate portions of the image that are derived from different patches. This way, global features of the anatomical structure may be learned and used to improve the quality of the image generated by the MLP neural network, without incurring significant computation or memory costs.

A system and method for improving image quality of periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER) imaging include acquiring a plurality of blades of k-space data of a region of interest in a rotational manner around a center of k-space via a magnetic resonance imaging (MRI) scanner from a coil during a PROPELLER sequence, wherein each blade of the plurality of blades of k-space data includes a plurality of parallel phase encoding lines sampled in a phase encoding order. The system and method also include utilizing a deep learning-based multi-level denoising network to denoise each blade of the plurality of blades in an image domain to generate a plurality of denoised blades, to utilize a PROPELLER reconstruction algorithm to generate a denoised-gridded image from the plurality of denoised blades, and to remove individual-based denoising-induced artifacts from the denoised-gridded image to generate a denoised, artifact-free gridded image.

A system includes an attenuation corrector configured to generate Computed Tomography-(CT-) based attenuation correction data from CT image data, a Positron Emission Tomography (PET) reconstructor configured to reconstruct first PET image data based on PET projection data and the CT-based attenuation correction data, an attenuation correction artifact mitigator configured to analyze the first PET image data for a presence of attenuation correction artifact, an inference engine configured to predict attenuation correction data based on non-attenuation corrected PET image data in response to the presence of attenuation correction artifact in the first PET image data, and an attenuation correction data updater configured to generate modified attenuation correction data based on the CT-based attenuation correction data and the predicted attenuation correction data. The PET reconstructor is further configured to reconstruct second PET image data based on the PET projection data and the modified attenuation correction data.

In a creation method of a trained model, a reconstructed image (60) obtained by reconstructing three-dimensional X-ray image data (80) is generated. A projection image (61) is generated from a three-dimensional model of an image element (50) by a simulation. The projection image is superimposed on the reconstructed image to generate a superimposed image (67). A trained model (40) is created by performing machine learning using the superimposed image, and the reconstructed image or the projection image.

A computer device performs feature extraction on two-dimensional medical images included in a three-dimensional medical image, to obtain image features corresponding to the two-dimensional medical images. The three-dimensional medical image are obtained by continuously scanning a target tissue structure. The computer device determines offsets of the two-dimensional medical images in a target direction based on the image features. The computer device performs feature alignment on the image features based on the offsets, to obtain aligned image features. The computer device performs three-dimensional segmentation on the three-dimensional medical image based on the aligned image features, to obtain three-dimensional layer distribution of the target tissue structure in the three-dimensional medical image.

Technology is described for handling truncated data in iterative reconstruction. A method comprises iterating on a volume of an object including a non-truncated part based on image data and at least one truncated part representing deficiently imaged data. The volume is represented by voxels. The iterating includes regularizing the non-truncated part of the volume using a first regularizer, and regularizing the truncated part of the volume using a second regularizer different from the first regularizer.

Systems and methods include acquisition of an anatomical image of an object, acquisition of molecular imaging data of the object at the plurality of photon detectors, reconstruction of a functional image based on the molecular imaging data, input of the anatomical image and the functional image to a trained neural network to generate a second functional image, and presentation of the second functional image.

Systems and methods for detecting multiple events during nuclear imaging scans, and for reconstructing images based on the detected events, are disclosed. In some embodiments, an image scanning system scans a subject, and generates a signal characterizing a detection event. The system generates sampled data based on sampling the at least one signal. Further, the system applies a trained machine learning process to the sampled data. Based on the application of the trained machine learning process, the system generates pulse data characterizing a plurality of decoupled pulses. For example, the pulse data may characterize pulse energy values of each of the decoupled pulses, and corresponding times for each of the pulses. Further, the method includes transmitting the pulse data to generate time-coincident pairs for image reconstruction.

A system for automating a magnetic resonance imaging (MRI) pipeline for a patient is provided. The system includes one or more processors that are configured to receive information that includes patient data and access one or more trained LLMs. The one or more processors are further configured to apply the patient data to the one or more trained LLMs to generate an MRI protocol that includes one or more pulse sequences and pulse sequence parameters. The one or more processors are further configured to store the MRI protocol.

This invention relates to methods and systems relating to medical diagnostics. In particular, the invention relates to a method of processing data from a medical diagnostic system which comprises a detector arrangement configured to detect interaction of high energy particles with tissue. The method comprises receiving diagnostic data associated with interaction of high energy particles with tissue, wherein the diagnostic data is generated as a result of interaction of the high energy particles and the tissue being detected by the detector arrangement of the medical diagnostic system. The method further comprises processing the received diagnostic data with a trained machine-based classifier in order to facilitate and / or assist a medical diagnosis, wherein the trained machine-based classifier is trained at least with digital diagnostic data generated by a digital twin of the medical diagnostic system, wherein the digital twin is a computer-implemented simulation of at least part of the medical diagnostic system.

An image is reconstituted by iterative approximation, a PET event updated image is produced by updating a current image using a PET event, a Compton event updated image is produced by updating the current image using a Compton event, the PET event updated image and the Compton event updated image that have been independently produced are weighted and added together, and the current image is updated using an image obtained by addition processing. In this way, PET events and Compton events, which have different properties, can be used in combination to efficiently and stably reconstitute images, improving image quality.

Systems and methods for fast reconstruction of motion-resolved magnetic resonance images using space-time-coil convolutional networks are disclosed. The system can receive a plurality of k-space data sets. The system can detect a motion signal therefrom. The system can classify the k-space data sets according to states of the motion signals. The system can resolve the k-space data set to Euclidean space images. The system can resolve the Euclidean space images to a combined Euclidian space image. For example, the system can use a convolutional network that exploits spatial, temporal and coil correlations without k-space data consistency to minimize computation time.

A computer-implemented system for foreign object detection in a scene. The system includes a memory-suppress diffusion network module adapted to reconstruct a reconstructed image from an encoded image, and a contrastive dissimilarity network adapted to combine the input image and the reconstructed image to predict an anomaly map for the input image. The encoded image is based on an input image, and the memory-suppress diffusion network module and the contrastive dissimilarity network are trained using only normal, real images. The system leverages only normal images in training and does not compromise the detection performance at the inference stage.

The present disclosure relates to a 2D-3D medical image registration method. The method includes obtaining a preoperative CT image and an intraoperative Xray image of a target bone block, inputting the preoperative CT image and the intraoperative Xray image into a depth-based learning-based regression network, and roughly estimating an initial spatial posture of the target bone block by using the regression network; and adjusting a projection of the preoperative CT image based on the initial spatial posture to generate a DRR image; and inputting the DRR image, the intraoperative Xray image, and the preoperative CT image into a pre-trained correspondence point relationship estimation network, estimating a feature point corresponding relationship between the DRR image and the Xray image by using the corresponding point relationship estimation network, and optimizing the initial spatial posture according to the feature point corresponding relationship to obtain an optimized spatial posture.

A method comprising receiving a plurality of images from a multi-modal imaging system; generating a plurality of filtered measurements by performing multi-modal spectral fusion of the plurality of images; and generating, using a physics-informed neural network (PINN) trained based on one or more physical principles associated with X-ray attenuation or scattering, a reconstructed object image based on the plurality of filtered measurements, wherein generating the reconstructed object image comprises (i) generating, using the PINN, a system matrix for an X-ray imaging forward model by refining one or more coefficients of the system matrix based on a physics-informed loss function, and (ii) generating, using the X-ray imaging forward model and based on the plurality of filtered measurements, the reconstructed object image.

Systems and methods include determination of an anatomical image of an object, input of the anatomical image to a trained neural network to generate a synthetic functional image, acquisition of molecular imaging data of the object based on acquisition parameters, reconstruction of a functional image based on the molecular imaging data, determination of a difference between the functional image and the synthetic functional image, change of one of the acquisition parameters based on the difference, acquisition of second molecular imaging data of the object based on the changed acquisition parameters, and reconstruction of a second functional image based on the second molecular imaging data.

A cross-regional and cross-view learning (C2RV) framework is provided for sparse-view reconstruction in cone-beam computed tomography (CBCT) by advantageously leveraging cross-region and cross-view feature learning to enhance representation of a point in 3D space before estimating an attenuation coefficient of the point. Specifically, multi-scale 3D volumetric representations (MS-3DV) are first introduced, where features are obtained by back-projecting multi-view features at different scales to the 3D space. Explicit MS-3DV enable cross-regional learning in the 3D space, providing richer information that helps better identify different internal anatomy structures. Hence, features of the point can be queried in a hybrid way, i.e. multi-scale voxel-aligned features from MS-3DV and multi-view pixel-aligned features from projections. Instead of considering queried features equally, scale-view cross-attention (SVC-Att) is used to adaptively learn aggregation weights by self-attention and cross-attention. Finally, multi-scale and multi-view features are aggregated to estimate the attenuation coefficient.

Techniques for generating magnetic resonance (MR) images of a subject from MR data obtained by a magnetic resonance imaging (MRI) system, the techniques including: obtaining input MR data obtained by imaging the subject using the MRI system; generating a plurality of transformed input MR data instances by applying a respective first plurality of transformations to the input MR data; generating a plurality of MR images from the plurality of transformed input MR data instances and the input MR data using a non-linear MR image reconstruction technique; generating an ensembled MR image from the plurality of MR images at least in part by: applying a second plurality of transformations to the plurality of MR images to obtain a plurality of transformed MR images; and combining the plurality of transformed MR images to obtain the ensembled MR image; and outputting the ensembled MR image.

A system for facilitating detection of defects in materials is configurable to: (i) access a set of input images comprising a plurality of cross-sectional images providing a representation of a component; and (ii) process the set of input images using an off-nominal anomaly detection model by, for each particular cross-sectional image of the plurality of cross-sectional images: (a) generate a set of patch embedding vectors comprising a patch embedding vector for each image patch of the particular cross-sectional image; (b) generate a set of difference metrics based on the set of patch embedding vectors and a probabilistic representation of a nominal component; and (c) determine a set of anomaly scores for the particular cross-sectional image based on the set of difference metrics.

A method, apparatus, and computer-readable storage medium for controlling X-ray computed tomography (CT) imaging. A first set of projection data is acquired in a first CT scan of an object with a CT imaging apparatus. The first CT image data is reconstructed from the first set of projection data. X-ray tube current modulation information is determined for a second CT scan of the object, based on a noise propagation model between X-ray projection data and CT image data, and using, as inputs, the obtained first set of projection data, information indicating an imaging region-of-interest (ROI) for the second CT scan, and a target image quality level in the imaging ROI. The second CT scan of the object is obtained based on the obtained X-ray tube current modulation information.

An image registration system and a method for a femur surgery relates to the field of surgery image registration are provided, before a surgery, DRR projections at different angles are carried out of a three-dimensional femur CT image, femoral coordinate systems and femoral characteristic points are respectively generated and selected on a the two-dimensional DRR projection images and two-dimensional X-ray images. Based on the femoral coordinate systems and the femoral characteristic points, the two-dimensional DRR projection images and the two-dimensional X-ray images are registered, so that the image registration precision is improved, a two-dimensional DRR projection image in the same femoral posture as a femoral posture during a two-dimensional X-ray image scanning can be accurately obtained, which can obtain a real surgical path for surgical navigation, ensures the precision of surgery-assisted navigation, and further improves the surgery quality.

A method for generating a motion-corrected MR image dataset of a subject includes: acquiring k-space data of an MR image of a subject in an imaging sequence; acquiring at least two low-resolution scout images of the subject interleaved with the k-space data of the imaging sequence; comparing the scout images with one another in order to detect and / or to estimate subject motion between the scout images; and reconstructing a motion-corrected MR image dataset from the k-space data acquired in the imaging sequence. The reconstruction process includes: estimating the motion trajectory of the subject by comparing the k-space data with at least one of the low-resolution scout images; and estimating the motion-corrected image dataset using the estimated motion trajectory, wherein the estimations involve minimizing the data consistency error between the acquired k-space data and a forward model described by an encoding operator.

Disclosed herein are systems, methods, and instrumentalities associated with magnetic resonance (MR) image reconstruction. An under-sampled MR image may be reconstructed through an iterative process (e.g., over multiple iterations) based on a machine-learning (ML) model. The ML model may be obtained through a reinforcement learning process during which the ML model may be used to predict a correction to an input MR image of at least one of the multiple iterations, apply the correction to the input MR image to obtain a reconstructed MR image, determine a reward for the ML model based on the reconstructed MR image, and adjust the parameters of the ML model based on the reward. The reward may be determined using a pre-trained reward neural network and the ML model may also be pre-trained in a supervised manner before being refined through the reinforcement learning process.

A deep-learning based framework to assess the quality of medical image auto-segmentation is described. According to an example, a computer-implemented method comprises receiving segmentation masks generated, via one or more segmentation models, from medical image data depicting an anatomical region of a subject, wherein each of the segmentation masks depicts a different anatomical structure of a set of different anatomical structures included in the anatomical region. The method further comprises generating reconstructed versions of the segmentation masks based on application of a multi-channel reconstruction model to the segmentation masks, wherein the reconstructed versions correspond to optimized versions of the segmentation masks. The method further comprises determining an assessment of quality of the segmentation masks based on comparison of the segmentation masks to the reconstructed versions, generating output data regarding the assessment of quality, and rendering the output data via an electronic output device.

For controlling reconstruction in emission tomography, the quality of data for detected emissions and / or the application controls the settings used in reconstruction. For example, a count density of the detected emissions is used to control the number of iterations in reconstruction to more likely avoid over and under fitting. The count density may be adaptively determined by re-binning through pixel size adjustment to find a smallest pixel size providing a sufficient count density. As another example, the detected data may have poor quality due to motion or high body mass index (BMI) of the patient, so the reconstruction is set to perform differently (e.g., less smoothing for high motion or a different number of iterations for high BMI). The quality of the data may be used in conjunction with the application or task for imaging the patient to control the reconstruction.

A method for training an algorithm for machine learning for a correction of recordings obtained by an imaging apparatus. A number of output images are recorded. Moreover a smaller number of first, high-quality scattered radiation images are simulated from the output images. A corresponding number of second, low-quality scattered radiation images is further simulated, wherein the simulation is undertaken with a number of photons reduced by at least an order of magnitude. The algorithm is trained with the second, low-quality scattered radiation images as input data and the first, high-quality scattered radiation images as output data.

A technique is disclosed for analyzing and displaying the extent to which the images and structures inferred by a physically seeded multiscale network correspond to genuine resolution improvement through noise insensitive point spread function deconvolution, and the extent to which they correspond to the hallucination of realistic looking structures with realistic frequency contents. A relative modulation transfer function can be computed, which can represent the distribution of frequency components in a particular reconstruction (e.g., a volumetric reconstruction from high-resolution data) that are not robustly recovered by a different reconstruction (e.g., a volumetric reconstruction via processing of low-resolution data with a trained neural network). The high-frequency portion of these frequency components can represent hallucinations introduced by a trained neural network, and can be leveraged to filter the different reconstruction prior to further use.

A magnetic resonance imaging (MRI) recovery method using a score-based diffusion model and an apparatus thereof are provided. The MRI recovery method using the score-based diffusion model, which is performed by a computer, is implemented, including training a continuous time-dependent score function with denoising score matching and sampling data from a conditional distribution given the measurements, leveraging the learned score function, and recovering an image.