Abstract: Techniques are described for generating reformatted views of a three-dimensional (3D) anatomy scan using deep-learning estimated scan prescription masks. According to an embodiment, a system is provided that comprises a memory that stores computer executable components, and a processor that executes the computer executable components stored in the memory. The computer executable components comprise a mask generation component that employs a pre-trained neural network model to generate masks for different anatomical landmarks depicted in one or more calibration images captured of an anatomical region of a patient. The computer executable components further comprise a reformatting component that reformats 3D image data captured of the anatomical region of the patient using the masks to generate different representations of the 3D image data that correspond to the different anatomical landmarks.

Abstract: Systems and techniques that facilitate data diversity visualization and/or quantification for machine learning models are provided. In various embodiments, a processor can access a first dataset and a second dataset, where a machine learning (ML) model is trained on the first dataset. In various instances, the processor can obtain a first set of latent activations generated by the ML model based on the first dataset, and a second set of latent activations generated by the ML model based on the second dataset. In various aspects, the processor can generate a first set of compressed data points based on the first set of latent activations, and a second set of compressed data points based on the second set of latent activations, via dimensionality reduction. In various instances, a diversity component can compute a diversity score based on the first set of compressed data points and second set of compressed data points.

Abstract: A medical imaging system includes at least one medical imaging device providing image data of a subject and a processing system programmed to generate a plurality of training images having simulated medical conditions by blending a pathology region from a plurality of template source images to a plurality of target images. The processing system is further programmed to train a deep learning network model using the plurality of training images and input the image data of the subject to the deep learning network model. The processing system is further programmed to generate a medical image of the subject based on the output of the deep learning network model.

Abstract: A computer-implemented method of automatically labeling medical images is provided. The method includes clustering training images and training labels into clusters, each cluster including a representative template having a representative image and a representative label. The method also includes training a neural network model with a training dataset that includes the training images and the training labels, and target outputs of the neural network model are labels of the medical images. The method further includes generating a suboptimal label corresponding to an unlabeled test image using the trained neural network model, and generating an optimal label corresponding to the unlabeled test image using the suboptimal label and representative templates.

Abstract: Methods and systems are provided for generating a normative medical image from an anomalous medical image. In an example, the method includes receiving an anomalous medical image, wherein the anomalous medical image includes anomalous data, mapping the anomalous medical image to a normative medical image using a trained generative network of a generative adversarial network (GAN), wherein the anomalous data of the anomalous medical image is mapped to normative data in the normative medical image. In some examples, the method may further include displaying the normative medical image via a display device, and/or utilizing the normative medical image for further image analysis tasks to generate robust outcomes from the anomalous medical image.

Abstract: Methods and systems are provided for reconstructing images from measurement data using one or more deep neural networks according to a decimation strategy. In one embodiment, a method for reconstructing an image using measurement data comprises, receiving measurement data acquired by an imaging device, selecting a decimation strategy, producing a reconstructed image from the measurement data using the decimation strategy and one or more deep neural networks, and displaying the reconstructed image via a display device. By decimating measurement data to form one or more decimated measurement data arrays, a computational complexity of mapping the measurement data to image data may be reduced from O(N4), where N is the size of the measurement data, to O(M4), where M is the size of an individual decimated measurement data array, wherein M<N.

Abstract: Methods and systems are provided for reconstructing images from measurement data using one or more deep neural networks according to a decimation strategy. In one embodiment, a method for reconstructing an image using measurement data comprises, receiving measurement data acquired by an imaging device, selecting a decimation strategy, producing a reconstructed image from the measurement data using the decimation strategy and one or more deep neural networks, and displaying the reconstructed image via a display device. By decimating measurement data to form one or more decimated measurement data arrays, a computational complexity of mapping the measurement data to image data may be reduced from O(N4), where N is the size of the measurement data, to O(M4), where M is the size of an individual decimated measurement data array, wherein M<N.

Abstract: Methods and systems are provided for determining diagnostic-scan parameters for a magnetic resonance (MR) diagnostic-scan, from MR calibration images, enabling acquisition of high-resolution diagnostic images of one or more anatomical regions of interest, while bypassing acquisition of localizer images, increasing a speed and efficiency of MR diagnostic-scanning. In one embodiment, a method for a magnetic resonance imaging (MRI) system comprises, acquiring a magnetic resonance (MR) calibration image of an imaging subject, mapping the MR calibration image to a landmark map using a trained deep neural network, determining one or more diagnostic-scan parameters based on the landmark map, acquiring an MR diagnostic image according to the diagnostic-scan parameters, and displaying the MR diagnostic image via a display device.

Abstract: Methods and systems are provided for reconstructing images from measurement data using one or more deep neural networks according to a decimation strategy. In one embodiment, a method for reconstructing an image using measurement data comprises, receiving measurement data acquired by an imaging device, selecting a decimation strategy, producing a reconstructed image from the measurement data using the decimation strategy and one or more deep neural networks, and displaying the reconstructed image via a display device. By decimating measurement data to form one or more decimated measurement data arrays, a computational complexity of mapping the measurement data to image data may be reduced from O(N4), where N is the size of the measurement data, to O(M4), where M is the size of an individual decimated measurement data array, wherein M<N.

Abstract: Methods and systems are provided for determining scan settings for a localizer scan based on a magnetic resonance (MR) calibration image. In one example, a method for magnetic resonance imaging (MRI) includes acquiring an MR calibration image of an imaging subject, mapping, by a trained deep neural network, the MR calibration image to a corresponding anatomical region of interest (ROI) attribute map for an anatomical ROI of the imaging subject, adjusting one or more localizer scan parameters based on the anatomical ROI attribute map, and acquiring one or more localizer images of the anatomical ROI according to the one or more localizer scan parameters.

Abstract: The present disclosure provides, in certain implementations, a rule-based or deep learning-based approach capable of assessing diagnostic utility of images in near real time with respect to acquisition. Correspondingly, an automated implementation of such an algorithm on the scanner would, in fact, emulate the doctor himself rating images in real time, and reduce the number of unneeded re-scans and recalls. In one aspect of the present invention it was found that diagnostic utility of an image is not an absolute measure, but instead depends upon the reading radiologist and the scan indication (i.e., the purpose of the scan). Therefore, adapting the threshold (probability of an imaging volume to be deemed good) as a function of reading radiologist and scan indication can result in decreasing the number of re-scans and recalls.

Abstract: Methods and systems are provided for generating a normative medical image from an anomalous medical image. In an example, the method includes receiving an anomalous medical image, wherein the anomalous medical image includes anomalous data, mapping the anomalous medical image to a normative medical image using a trained generative network of a generative adversarial network (GAN), wherein the anomalous data of the anomalous medical image is mapped to normative data in the normative medical image. In some examples, the method may further include displaying the normative medical image via a display device, and/or utilizing the normative medical image for further image analysis tasks to generate robust outcomes from the anomalous medical image.

Abstract: A method for automated evaluation of motion correction is presented. The method includes identifying one or more regions of interest in each of a plurality of images corresponding to a subject of interest. Furthermore, the method includes selecting valid voxels in each of the one or more regions of interest in each of the plurality of images. The method also includes computing a similarity metric, a dispersion metric, or both the similarity metric and the dispersion metric for each region of interest in each of the plurality of images. Additionally, the method includes generating a similarity map, a dispersion map, or both the similarity map and the dispersion map based on the similarity metrics and the dispersion metrics corresponding to the one or more regions of interest.

Abstract: The subject matter discussed herein relates to systems and methods for generating a clinical outcome based on creating a task-specific model associated with processing raw image(s). In one such example, input raw data is acquired using an imaging system, a selection input corresponding to a clinical task is received, and a task-specific model corresponding to the clinical task is retrieved. Using the task-specific model, the raw data is mapped onto an application specific manifold. Based on the mapping of the raw data onto the application specific manifold the clinical outcome is generated, and subsequently providing the clinical outcome for review.

Abstract: The present disclosure relates to use of a workflow for automatic prescription of different radiological imaging scan planes across different anatomies and modalities. The automated prescription of such imaging scan planes helps ensure contiguous visualization of the different landmark structures. Unlike prior approaches, the disclosed technique determines the necessary planes using the localizer images itself and does not explicitly segment or delineate the landmark structures to perform plane prescription.

Abstract: The present disclosure provides, in certain implementations, a rule-based or deep learning-based approach capable of assessing diagnostic utility of images in near real time with respect to acquisition. Correspondingly, an automated implementation of such an algorithm on the scanner would, in fact, emulate the doctor himself rating images in real time, and reduce the number of unneeded re-scans and recalls. In one aspect of the present invention it was found that diagnostic utility of an image is not an absolute measure, but instead depends upon the reading radiologist and the scan indication (i.e., the purpose of the scan). Therefore, adapting the threshold (probability of an imaging volume to be deemed good) as a function of reading radiologist and scan indication can result in decreasing the number of re-scans and recalls.

Abstract: An imaging system and method are disclosed. An MR image and measured B0 field map of a target volume in a subject are reconstructed, where the MR image includes one or more bright and/or dark regions. One or more distinctive constituent materials corresponding to the bright regions are identified. Each dark region is iteratively labeled as one or more ambiguous constituent materials. Susceptibility values corresponding to each distinctive and iteratively labeled ambiguous constituent material is assigned. A simulated B0 field map is iteratively generated based on the assigned susceptibility values. A similarity metric is determined between the measured and simulated B0 field maps. Constituent materials are identified in the dark regions based on the similarity metric to ascertain corresponding susceptibility values. The MRI data is corrected based on the assigned and ascertained susceptibility values. A diagnostic assessment of the target volume is determined based on the corrected MRI data.

Abstract: A system and method for detecting motion is presented. The system and method includes identifying a region of interest in the plurality of images corresponding to a subject of interest. Furthermore, the system and method includes determining signal characteristics corresponding to the region of interest. In addition, the system and method includes generating a composite signal, where the composite signal comprises an aggregate of the signal characteristics corresponding to the region of interest. The system and method also includes analyzing the composite signal to detect motion in the region of interest.

Abstract: A method for synchronization of a longitudinal data set from a subject includes receiving a first ensemble registration estimate having a first reference image corresponding to a first image ensemble and receiving a second image ensemble different from the first image ensemble. The method includes determining a second reference image based on the second image ensemble and the first reference image. Further, the method includes determining a second ensemble registration estimate based on the first ensemble registration estimate, the second reference image, the first image ensemble and the second image ensemble using an optimization technique. The method further includes generating a synchronized image ensemble corresponding to the first image ensemble and the second image ensemble based on the second ensemble registration estimate. The method also includes determining a medical condition of the subject by a medical practitioner based on the synchronized image ensemble.

Abstract: The disclosed approach employs a generic methodology for transforming individual modality specific multi-parametric data into data, e.g., maps or images, which provides direct insight into the underlying physiology of the tissue. This may facilitate better clinical evaluation of the disease data as well as help non-imaging technologists and scientist to directly correlate imaging findings with basic biological phenomenon being studied with imaging.