Beamforming
The Beamforming Algorithm reconstructs ultrasound images from RF or IQ channel data using beamforming and, optionally, coherent compounding.
Input
The algorithm is invoked with ultrasound channel data (RF or IQ) as input, this can be either one or more images (for offline processing) or a stream (for online processing). The input data must contain a PulseEchoSequence describing the acquisition parameters. The semantics of the input data are determined by the ChannelDataLayout. For focused imaging, each frame in the input data is processed individually. For ultrafast imaging, the algorithm can perform coherent compounding or process each frame as a separate slice.
Output
The output is a pre-scan converted ultrasound image or image set. For focused imaging and ultrafast imaging with coherent compounding, the output is a 2D image for each input frame. For ultrafast imaging without compounding, the output is a 3D image, where each slice corresponds to a single PulseEchoEvent in the input data.
Description
The Beamforming Algorithm reconstructs ultrasound images from channel data by applying delay-and-sum beamforming. The algorithm uses the PulseEchoSequence and, if available, FrameGeometry metadata to determine the geometry and spatial sampling of the output image. If no geometry metadata is present, a default geometry is derived from the acquisition sequence.
For ultrafast imaging, the region of interest for beamforming is determined by UltrafastBeamformingRois if present, or otherwise by the sequence. The algorithm supports both focused and ultrafast imaging modes, and can optionally perform coherent compounding to improve image quality.
The output image size and scaling can be configured. The algorithm is designed to work efficiently with large channel datasets and supports parallel processing of frames.
The Controller for the Beamforming Algorithm provides the following options:
F-number: Sets the F-number of the transducer, influencing the beam width and focusing.
Coherent Compounding: Enables or disables coherent compounding during beamforming.
Output Size: Desired output image size in x and y directions, corresponding to the number of rays and number of samples along the rays in the resulting pre-scan converted image. For focused imaging, the number of rays corresponds to the number of focused beams and the configured x value in this parameter is ignored.
Output Value Scaling: Scaling factor for output values to avoid saturation for integer output types.
Note
The algorithm automatically determines the geometry and region of interest based on the input data and metadata.
Note
For ultrafast imaging, the output format depends on whether coherent compounding is enabled.