Recognition Based Medical Imaging System: Methods and Systems for Enhanced Medical Image Acquisition and Reconstruction Using Recognition Physics
OwnerJonathan Washburninventor
Life of this patent
Estimating the term…
Abstract
A recognition-based medical imaging framework is disclosed that computes a bounded recognition coverage value from a quality signal under a scaling parameter Xopt, where Xopt is set to phi over pi or calibrated, and uses the same value to drive real-time acquisition control and reconstruction weighting, the modality-agnostic system for computed tomography magnetic resonance imaging ultrasound positron emission tomography and digital radiography runs on CPU GPU or FPGA within vendor safety limits and achieves reduced dose or scan time at matched image quality or improved signal-to-noise or contrast-to-noise at matched dose with outputs recorded in standards-compliant images and metadata.
Claims20
A method for medical imaging comprising receiving raw acquisition data and scanner or patient metadata from an imaging device, computing a recognition coverage value for a region of interest from a quality signal r using a recognition coverage function Fcov parameterized by a recognition scaling parameter Xopt, adaptively selecting or modulating one or more acquisition parameters during the exam based on the recognition coverage value, reconstructing an image using the same recognition coverage value to weight at least one of frequency-domain filters spatially varying regularization and voxel-wise operations, and outputting a medical image with associated metadata that records the acquisition parameters the recognition coverage value and reconstruction settings.
A medical imaging system comprising a modality sensor assembly configured to acquire raw data and scanner or patient metadata, a controller configured to compute a recognition coverage value from a quality signal r using a recognition coverage function Fcov parameterized by a recognition scaling parameter Xopt and to adapt one or more acquisition parameters based on the recognition coverage value during the exam, and a reconstruction engine configured to use the same recognition coverage value to weight at least one of frequency-domain filters spatially varying regularization and voxel-wise operations to produce a medical image, the system further comprising safety interlocks enforcing rate limits and bounds and an interface configured to export the medical image together with metadata that records the acquisition parameters the recognition coverage value and reconstruction settings.
A non-transitory computer-readable medium storing instructions that when executed by one or more processors cause the processors to receive raw acquisition data and scanner or patient metadata from an imaging device, compute a recognition coverage value from a quality signal r using a recognition coverage function Fcov parameterized by a recognition scaling parameter Xopt, adapt one or more acquisition parameters during the exam based on the recognition coverage value, reconstruct an image using the same recognition coverage value to weight at least one of frequency-domain filters spatially varying regularization and voxel-wise operations, and output a medical image with associated metadata that records the acquisition parameters the recognition coverage value and reconstruction settings.
Dependencies are inferred from the claim text.