Abstract: Systems and methods facilitate privacy-preserving data curation in a federated learning system by transmitting a portion of a potential data sample to a remote location. The portion is inspected for quality to rule out data samples that do not satisfy data curation criteria. The remote examination focuses on checking the region of interest but maintains privacy as the examination is unable to parse any other identifiable subject information such as face, body shape etc. because pixels or voxels outside the portion are not included. The examination results are sent back to the collaborators so that inappropriate data samples can be excluded during federated learning rounds.

Abstract: Systems and methods facilitate privacy-preserving data curation in a federated learning system by transmitting a portion of a potential data sample to a remote location. The portion is inspected for quality to rule out data samples that do not satisfy data curation criteria. The remote examination focuses on checking the region of interest but maintains privacy as the examination is unable to parse any other identifiable subject information such as face, body shape etc. because pixels or voxels outside the portion are not included. The examination results are sent back to the collaborators so that inappropriate data samples can be excluded during federated learning rounds.

Abstract: Systems and Methods for adaptive aggregation in a federated learning model. An aggregation server sends global model weights to all chosen collaborators for initialization. Each collaborator updates the model weights for certain epochs and then sends the updated model weights back to the aggregation server. The aggregation server adaptively aggregates the updated model weights using at least a computed model divergence value and sends the aggregated model weight to collaborators.

Abstract: An apparatus and method for tracking the position and orientation of one or more objects in three dimensional space is disclosed. One or more tracked sensor units are each connected with a respective object. Each tracked sensor unit includes one or more light sources and an inertial measurement unit. One or more position sensitive detector tracking devices track the position of the tracked sensor units. Each position sensitive detector tracking device includes a plurality of position sensitive detector sensors combined with optical lenses that focus light from a larger field of view onto each position sensitive detector sensor. The position and orientation of each object in three-dimensional space is calculated from the output of the inertial measurement unit of the respective tracked sensor unit and the output of the one or more position sensitive detector tracking devices in response to light emitted from the one or more light sources of the respective tracked sensor unit.

Abstract: An apparatus and method for tracking the position and orientation of one or more objects in three dimensional space is disclosed. One or more tracked sensor units are each connected with a respective object. Each tracked sensor unit includes one or more light sources and an inertial measurement unit. One or more position sensitive detector tracking devices track the position of the tracked sensor units. Each position sensitive detector tracking device includes a plurality of position sensitive detector sensors combined with optical lenses that focus light from a larger field of view onto each position sensitive detector sensor. The position and orientation of each object in three-dimensional space is calculated from the output of the inertial measurement unit of the respective tracked sensor unit and the output of the one or more position sensitive detector tracking devices in response to light emitted from the one or more light sources of the respective tracked sensor unit.

Abstract: Methods and systems for luggage visualization perform virtual unpacking by visually moving an object image away from its original pose. A scanned 3D volume is segmented guided by a confidence measure to create a label volume whose voxels specify the detected object IDs. The luggage dataset and the label volume are visualized by volume rendering. Using an automatic coloring algorithm, any pair of objects whose projections are adjacent in an image are assigned distinct hues. A layered framework efficiently renders a scene mixed with packed luggage, animated unpacking objects, and already unpacked objects put aside for further inspection. A GPU is used to automatically select objects that are not blocked by others and can be unpacked.

Abstract: Methods, apparatuses, and systems are provided for live capturing of light map image sequences for image-based lighting of medical data. Patient volume scan data for a target area is received over time by a processor. Lighting environment data for the target area is captured over time by a camera. The camera transmits the lighting environment data to the processor over time. The processor lights the patient volume scan data with the lighting environment data into lighted volume data over time. The processor renders an image of the lighted volume data over time.

Abstract: Methods, apparatuses, and systems are provided for live capturing of light map image sequences for image-based lighting of medical data. Patient volume scan data for a target area is received over time by a processor. Lighting environment data for the target area is captured over time by a camera. The camera transmits the lighting environment data to the processor over time. The processor lights the patient volume scan data with the lighting environment data into lighted volume data over time. The processor renders an image of the lighted volume data over time.

Abstract: A method for Monte Carlo volume rendering in accordance with the present teachings includes: (a) tracing a plurality of light rays into a scene containing volumetric data, the light rays configured for simulating global illumination; (b) randomizing the scattering location and direction of the plurality of light rays through the volume, wherein a common sequence of random numbers is used in order for the scattering direction of each of the plurality of randomized scattered light rays to be substantially parallel; (c) computing at least one trilinearly interpolated and shaded sample along each of the plurality of randomized scattered light rays based on stored volumetric data, wherein at least a portion of the stored volumetric data used in at least a portion of the computing is configured for coherent access; and (d) rendering the volume with global illumination based on a plurality of iterations of the tracing, the randomizing, and the computing. Systems for Monte Carlo volume rendering are described.

Abstract: A method for Monte Carlo volume rendering in accordance with the present teachings includes: (a) tracing a plurality of light rays into a scene containing volumetric data, the light rays configured for simulating global illumination; (b) randomizing the scattering location and direction of the plurality of light rays through the volume, wherein a common sequence of random numbers is used in order for the scattering direction of each of the plurality of randomized scattered light rays to be substantially parallel; (c) computing at least one trilinearly interpolated and shaded sample along each of the plurality of randomized scattered light rays based on stored volumetric data, wherein at least a portion of the stored volumetric data used in at least a portion of the computing is configured for coherent access; and (d) rendering the volume with global illumination based on a plurality of iterations of the tracing, the randomizing, and the computing. Systems for Monte Carlo volume rendering are described.

Abstract: A method and system for integrating radiological and pathological information for cancer diagnosis, therapy selection, and monitoring is disclosed. A radiological image of a patient, such as a magnetic resonance (MR), computed tomography (CT), positron emission tomography (PET), or ultrasound image, is received. A location corresponding to each of one or more biopsy samples is determined in the at least one radiological image. An integrated display is used to display a histological image corresponding to the each biopsy samples, the radiological image, and the location corresponding to each biopsy samples in the radiological image. Pathological information and radiological information are integrated by combining features extracted from the histological images and the features extracted from the corresponding locations in the radiological image for cancer grading, prognosis prediction, and therapy selection.

Abstract: Robotic navigation is provided for nuclear probe imaging. Using a three-dimensional scanner (19), the surface of a patient is determined (42). A calibrated robotic system positions (48) a nuclear probe about the patient based on the surface. The positioning (48) may be without contacting the patient and the surface may be used in reconstruction to account for spacing of the probe from the patient. By using the robotic system for positioning (48), the speed, resolution and/or quality of the reconstructed image may be predetermined, user settable, and/or improved compared to manual scanning. The reconstruction (52) may be more computationally efficient by providing for regular spacing of radiation detection locations within the volume.

Abstract: A method for processing medical imaging data includes: (a) selecting a subset of medical imaging data to be processed, wherein the medical imaging data is stored in a cloud-based storage system; (b) choosing a processing algorithm to apply to the selected subset of medical imaging data, wherein the chosen processing algorithm is stored in the cloud-based storage system; (c) executing the chosen processing algorithm in the cloud-based storage system to generate a processing result; and (d) displaying the processing result to a client via a user interface. Systems for processing medical imaging data are described.

Abstract: Multiple users are supported with an ultrasound server. Tiling of images may be used to limit transmission and/or bandwidth. By transmitting parts of images that change and avoiding transmission of other parts, wireless and processing bandwidth may be optimized. On the server side, separate instances are used for scanning each patient or for each of the multiple transducer probes being used. Dynamic assignment of shared resources based on use of the transducer probes may provide further optimization. From an overall perspective, the server may beamform from data received by a transducer probe based on controls routed from a separate tablet used as a display and user input.

Abstract: Methods and systems for luggage visualization perform virtual unpacking by visually moving an object image away from its original pose. A scanned 3D volume is segmented guided by a confidence measure to create a label volume whose voxels specify the detected object IDs. The luggage dataset and the label volume are visualized by volume rendering. Using an automatic coloring algorithm, any pair of objects whose projections are adjacent in an image are assigned distinct hues. A layered framework efficiently renders a scene mixed with packed luggage, animated unpacking objects, and already unpacked objects put aside for further inspection. A GPU is used to automatically select objects that are not blocked by others and can be unpacked.

Abstract: A method and system for integrating radiological and pathological information for cancer diagnosis, therapy selection, and monitoring is disclosed. A radiological image of a patient, such as a magnetic resonance (MR), computed tomography (CT), positron emission tomography (PET), or ultrasound image, is received. A location corresponding to each of one or more biopsy samples is determined in the at least one radiological image. An integrated display is used to display a histological image corresponding to the each biopsy samples, the radiological image, and the location corresponding to each biopsy samples in the radiological image. Pathological information and radiological information are integrated by combining features extracted from the histological images and the features extracted from the corresponding locations in the radiological image for cancer grading, prognosis prediction, and therapy selection.

Abstract: A nuclear probe and ultrasound transducer are interconnected, such as being in a same hand-held housing. The interconnection aligns the coordinate systems in a known spatial relationship. The ultrasound data is used to detect transducer offset or change in position without a tracking sensor. The radiation detected by the nuclear probe may be reconstructed into an image based on the detected transducer position since the nuclear probe moves with the ultrasound transducer. Both anatomical and functional imaging may be provided together without the complications of calibration and tracking. Where a therapeutic transducer is included, therapy may also be provided. The anatomical and functional information identifies the regions for treatment. The same device, already positioned correctly based on the functional and anatomical imaging, is then used for treatment with high intensity focused ultrasound.

Abstract: For cloud-based computer assisted detection, hierarchal detection is used, allowing detection on data at progressively greater resolutions. Detected locations at coarser resolutions are used to limit the data transmitted at greater resolutions. Data is only transmitted for neighborhoods around the previously detected locations. Subsequent detection using higher resolution data refines the locations, but only for regions associated with previous detection. By limiting the number and/or size of regions provided at greater resolutions based on the previous detection, the progressive transmission avoids transmission of some data. Additionally, or alternatively, lossy compression may be used without or with minimal reduction in detection sensitivity.

Abstract: The present invention discloses a method and system for deploying software, e.g., software modules designed for acquisition and/or post-processing of images related to a particular imaging agent. The modules can be remotely accessed from a software server.

Abstract: A mobile device (160) for medical image analysis is disclosed. The mobile device (160) includes a display (162), a communication module (218), a memory (204) configured to store processor-executable instructions (224) and a processor (202) in communication with the display (162), the communication module (218) and the memory (204). The processor (202) being configured to execute the processor-executable instructions (224) to implement a compression routine to generate a compressed representation of a medical image stored in the memory (204), transmit the compressed representation to a remote device (110) via the communication module (218), receive segmented results from the remote device (110), wherein the segmented results are derived from a reconstruction of the compressed representation generated at the remote device (110), and present, via the display (162), a segmented medical image based on the received segmented results.