Siemens Medical Solutions USA, Inc.

Systems and methods for generating pseudo-masks are provided. 1) a first and a second medical image of an anatomical structure of a patient and 2) a mask of the anatomical structure in the second medical image are received. The second medical image is aligned to the first medical image according to a deformation field. The mask of the anatomical structure in the second medical image is modified based on the deformation field to generate a pseudo-mask of the anatomical structure in the first medical image. The pseudo-mask of the anatomical structure in the first medical image is output.

For transducers with a chip-on-array arrangement, the dematching layer extends beyond a footprint of the array, allowing for connection of the grounding plane without sidewall metalization. The flexible circuit material is tiled, reducing thermal deformation, reducing cost, and increasing process yield. The dematching layer extension and the tiled flexible circuit may be used together or individually in a given transducer.

For SPECT imaging, the reconstruction optimizes separately for the image or object and the regularizer, such as using alternating direction method of multipliers (ADMM). This framework provides for SPECT focused plug and play (PnP). Various denoisers may be used. The result is reconstruction specific to SPECT that has the hybrid (e.g., physics model with denoiser) benefits while minimizing complication.

For ultrasound-based proxy estimation with an ultrasound scanner, one or more ultrasound measurements are converted to a biomarker of a different modality, such as providing a measurement of a standard biomarker from histology or a lab test assay. The conversion uses a user-selected model, allowing for selection of one of various models relating ultrasound measurements to the biomarker. The ultrasound scan sequence may be optimized to the one or more biomarkers and the user-selected model. This improvement in use of ultrasound measurements relative to standard biomarkers may more likely result in acceptance of the less costly and more easily obtained ultrasound measures in diagnosis.

For cardiac flow detection in echocardiography, by detecting one or more valves, sampling planes or flow regions spaced from the valve and/or based on multiple valves are identified. A confidence of the detection may be used to indicate confidence of calculated quantities and/or to place the sampling planes.

Systems and methods for reconstructing medical images are disclosed. Measurement data from positron emission tomography (PET) data, and measurement data from an anatomy modality, such as magnetic resonance (MR) data, is received from an image scanning system. A PET image is generated based on the PET measurement data, and a plurality of anatomy images are generated based on the anatomy measurement data. The plurality of anatomy images include voxels of differing spatial resolutions. A multi-spatial image reconstruction process is applied to the PET measurement data and the plurality of anatomy images. Based on the application of the image reconstruction process, image volume data characterizing a reconstructed medical image is generated. The reconstructed medical image may be displayed to a medical professional for diagnosis.

Provided is a medical imaging scanner system such as a PET scanner system having a patient bed pallet that is provided with at least one multi-axis motion sensor for monitoring the movement of the patient bed pallet and detecting any deviation in the orientation of the patient bed pallet from a predetermined desired orientation during the movement.

An ultrasound scanner scans a patient. The scan data is checked for an optimal view as a subpart of the scan. Where the field of view does not include the optimal view, the scan is repeated with a different spatial extent. Where the field of view does include the optimal view, then the scan data for that optimal view is used to generate an image. Both the view and the overall or broader scan are controlled together so that a sonographer may place the transducer at an approximate location to still provide a precise view.

For coarse-slice resolution upsampling, a machine-learned model or network infers data or samples along or between slices (e.g., creates extra slices) from the 2D MRI data (e.g., from the stack of slices). Rapid scanning may be used to create the stack of slices with poor slice-to-slice resolution, and a 3D volume with better resolution across slices is provided by inference from the stack of slices.

For shear wave imaging with ultrasound, the apparent pulse repetition frequency is increased by combining displacements from different lateral locations. Different combinations based on different shear wave velocities and corresponding time shifts and/or attenuations and corresponding scalings are tested to find a smooth displacement profile for the combination. Once the smooth displacement profile is found, the corresponding shear wave velocity is estimated or determined.