Abstract: A deep learning-based continuous federated learning network system is provided. The system includes a global site comprising a global model and a plurality of local sites having a respective local model derived from the global model. The plurality of model tuning modules having a processing system are provided at the plurality of local sites for tuning the respective local model. The processing system is programmed to receive incremental data and select one or more layers of the local model for tuning based on the incremental data. Finally, the selected layers are tuned to generate a retrained model.

Abstract: A computer implemented method is provided. The method includes establishing, via multiple processors, a continuous federated learning framework including a global model at a global site and respective local models derived from the global model at respective local sites. The method also includes retraining or retuning, via the multiple processors, the global model and the respective local models without sharing actual datasets between the global site and the respective local sites but instead sharing synthetic datasets generated from the actual datasets.

Abstract: A pulse sequence generation computing device for a magnetic resonance imaging (MRI) system includes a processor in communication with a memory device. The processor is programmed to receive a pulse sequence including a plurality of gradient pulses and provide a pulse sequence threshold function corresponding to an acoustic noise reduction level. For each gradient pulse in the pulse sequence, the processor is programmed to determine an amplitude and a slew rate of the gradient pulse, determine a threshold amplitude and slew rate of the gradient pulse, and compare the determined amplitude and slew rate to the threshold amplitude and slew rate. If either the determined amplitude or slew rate exceeds the threshold amplitude or slew rate, the processor adjusts at least one of the amplitude and the slew rate of the gradient pulse to an amplitude and a slew rate as defined by the pulse sequence threshold function.

Abstract: The subject matter discussed herein relates to multi-modal image alignment to facilitate biopsy procedures and post-biopsy procedures. In one such example, prostate structures (or other suitable anatomic features or structures) are automatically segmented in pre-biopsy MR and pre-biopsy ultrasound images. Thereafter, pre-biopsy MR and pre-biopsy ultrasound contours are aligned. To account for non-linear deformation of the imaged anatomic structure, a patient-specific transformation model is trained via deep learning based at least in part on the pre-biopsy ultrasound images. The pre-biopsy ultrasound images that are overlaid with the pre-biopsy MR contours and based off the deformable transformation model are then aligned with the biopsy ultrasound images. Such real-time alignment using multi-modality imaging techniques provides guidance during the biopsy and post-biopsy system.

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: The present disclosure relates to automatically determining respiratory phases (e.g., end-inspiration/expiration respiratory phases) in real time using ultrasound beamspace data. The respiratory phases may be used subsequently in a therapy or treatment (e.g., image-guided radiation-therapy (IGRT)) for precise dose-delivery. In certain implementations, vessel bifurcation may be tracked and respiration phases determined in real time using the tracked vessel bifurcations to facilitate respiration gating of the treatment or therapy.

Abstract: The subject matter discussed herein relates to the automatic, real-time registration of pre-operative magnetic resonance imaging (MRI) data to intra-operative ultrasound (US) data (e.g., reconstructed images or unreconstructed data), such as to facilitate surgical guidance or other interventional procedures. In one such example, brain structures (or other suitable anatomic features or structures) are automatically segmented in pre-operative and intra-operative ultrasound data. Thereafter, anatomic structure (e.g., brain structure) guided registration is applied between pre-operative and intra-operative ultrasound data to account for non-linear deformation of the imaged anatomic structure. MR images that are pre-registered to pre-operative ultrasound images are then given the same nonlinear spatial transformation to align the MR images with intra-operative ultrasound images to provide surgical guidance.

Abstract: The subject matter discussed herein relates to multi-modal image alignment to facilitate biopsy procedures and post-biopsy procedures. In one such example, prostate structures (or other suitable anatomic features or structures) are automatically segmented in pre-biopsy MR and pre-biopsy ultrasound images. Thereafter, pre-biopsy MR and pre-biopsy ultrasound contours are aligned. To account for non-linear deformation of the imaged anatomic structure, a patient-specific transformation model is trained via deep learning based at least in part on the pre-biopsy ultrasound images. The pre-biopsy ultrasound images that are overlaid with the pre-biopsy MR contours and based off the deformable transformation model are then aligned with the biopsy ultrasound images. Such real-time alignment using multi-modality imaging techniques provides guidance during the biopsy and post-biopsy system.

Abstract: An RF coil array for use in MRI is disclosed, which includes a plurality of transmit coil elements and a plurality of RF power amplifiers. Each RF power amplifier is integrated with at least one transmit coil element for driving the at least one transmit coil element. An MRI transmit array is also disclosed, which includes a plurality of RF transmitters for generating a plurality of RF signals, the RF coil array above-mentioned, and a DC voltage source for providing a DC voltage to the plurality of transmit coil elements. The RF coil array further includes an RF shield for shielding the plurality of transmit coil elements from interacting with magnet cryostat and gradient coil elements. The plurality of RF power amplifiers are connected with respective RF transmitters and configured for power amplification of the RF signals from the respective RF transmitters, and the plurality of transmit coil elements are configured for transmitting respective amplified RF signals.

Abstract: The subject matter discussed herein relates to the automatic, real-time registration of pre-operative magnetic resonance imaging (MRI) data to intra-operative ultrasound (US) data (e.g., reconstructed images or unreconstructed data), such as to facilitate surgical guidance or other interventional procedures. In one such example, brain structures (or other suitable anatomic features or structures) are automatically segmented in pre-operative and intra-operative ultrasound data. Thereafter, anatomic structure (e.g., brain structure) guided registration is applied between pre-operative and intra-operative ultrasound data to account for non-linear deformation of the imaged anatomic structure. MR images that are pre-registered to pre-operative ultrasound images are then given the same nonlinear spatial transformation to align the MR images with intra-operative ultrasound images to provide surgical guidance.

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: A method is provided. The method includes acquiring simultaneously multiple magnetic resonance (MR) images and multiple ultrasound images of an anatomical region of a subject over a scanned duration. The method also includes training an unsupervised deep learning-based deformable registration network. This training includes training a MR registration subnetwork based on the multiple MR images to generate MR deformation and transformation vectors, training an ultrasound registration subnetwork based on the multiple ultrasound images to generate ultrasound deformation and transformation vectors, and training a MR-to-ultrasound subnetwork based the multiple MR images and the multiple ultrasound images to generate MR-to-ultrasound deformation and transformation vectors between corresponding pairs of MR images and ultrasound images at each time point.

Abstract: A synchronization system based on wireless or limited cable interconnection is disclosed, which includes a central unit comprising a delay compensation module, and RF transmission channels. At least one channel comprises a clock controller and a synchronization controller. The central unit transmits a controlled clock signal and a controlled synchronization signal to the at least one channel, and receives a clock echo signal and a synchronization echo signal. The delay compensation module estimates a clock phase compensation based on the controlled clock signal transmitted and the clock echo signal received, and a synchronization delay compensation based on the controlled synchronization signal transmitted and the synchronization echo signal received. The clock controller and the synchronization controller adjusts respectively a clock signal and a synchronization signal received from the at least one channel based on the clock phase compensation and the synchronization delay compensation.

Abstract: The present disclosure relates to the classification of images, such as medical images using machine learning techniques. In certain aspects, the technique may employ a distance metric for the purpose of classification, where the distance metric determined for a given image with respect to a homogenous group or class of images is used to classify the image.

Abstract: A method is provided. The method includes acquiring simultaneously multiple magnetic resonance (MR) images and multiple ultrasound images of an anatomical region of a subject over a scanned duration. The method also includes training an unsupervised deep learning-based deformable registration network. This training includes training a MR registration subnetwork based on the multiple MR images to generate MR deformation and transformation vectors, training an ultrasound registration subnetwork based on the multiple ultrasound images to generate ultrasound deformation and transformation vectors, and training a MR-to-ultrasound subnetwork based the multiple MR images and the multiple ultrasound images to generate MR-to-ultrasound deformation and transformation vectors between corresponding pairs of MR images and ultrasound images at each time point.

Abstract: The present disclosure relates to the classification of images, such as medical images using machine learning techniques. In certain aspects, the technique may employ a distance metric for the purpose of classification, where the distance metric determined for a given image with respect to a homogenous group or class of images is used to classify the image.

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: The present disclosure describes non-invasive approaches for delineating and characterizing tissue using MR imaging over a range of treatment levels. By way of example, tumor tissue may be distinguished and delineated from other tissue, such as muscle tissue. Further, tumor tissue may be characterized as malignant or benign using such approaches.

Abstract: A synchronization system based on wireless or limited cable interconnection is disclosed, which includes a central unit comprising a delay compensation module, and RF transmission channels. At least one channel comprises a clock controller and a synchronization controller. The central unit transmits a controlled clock signal and a controlled synchronization signal to the at least one channel, and receives a clock echo signal and a synchronization echo signal. The delay compensation module estimates a clock phase compensation based on the controlled clock signal transmitted and the clock echo signal received, and a synchronization delay compensation based on the controlled synchronization signal transmitted and the synchronization echo signal received. The clock controller and the synchronization controller adjusts respectively a clock signal and a synchronization signal received from the at least one channel based on the clock phase compensation and the synchronization delay compensation.

Abstract: An RF coil array for use in MRI is disclosed, which includes a plurality of transmit coil elements and a plurality of RF power amplifiers. Each RF power amplifier is integrated with at least one transmit coil element for driving the at least one transmit coil element. An MRI transmit array is also disclosed, which includes a plurality of RF transmitters for generating a plurality of RF signals, the RF coil array above-mentioned, and a DC voltage source for providing a DC voltage to the plurality of transmit coil elements. The RF coil array further includes an RF shield for shielding the plurality of transmit coil elements from interacting with magnet cryostat and gradient coil elements. The plurality of RF power amplifiers are connected with respective RF transmitters and configured for power amplification of the RF signals from the respective RF transmitters, and the plurality of transmit coil elements are configured for transmitting respective amplified RF signals.