High Frame Rate Acquisition

Overview
The Sonix systems can be configured to run at very high frame rates, the methods described below show how to perform high frame rate imaging.

Small Sector
For 6cm imaging depth, a single RF line can be acquired at 10 kHz. Thus, 10 lines can be acquired at 1 kHz. To implement this using the research software interface:
 * Run the Sonix software in the B/RF collection mode
 * Use an Ulterius SDK client to connect to Exam and set the Sector parameter to 5-10%. The frame rate will be displayed on the screen
 * Setting the line density can also be used to adjust the spacing between lines (64,128,256...)
 * Alternatively, the Texo SDK can be used by itself to have full control over the number of lines, spacing between lines, and location of the sector



Trade offs: Easy to implement
 * No hardware overhead
 * Long observation time
 * Typically very small field of view
 * Acquisition lag exists between individual lines in the sector.

Synchronization
By dividing the image into small sectors each sector can be acquired at high frame rates. By synchronizing the excitation and the start of the image acquisition for each sector, the acquisition delays can be removed (this step is generally called re-phasing). This way the data can be acquired at a virtual high frame rates. The above settings (10 lines per sector) results in virtual frame rate of 1000 frames per second (fps). Using one line per sector will result in 10,000 fps.

Sonix supports both input synchronization and output synchronization schemes. Thus, in one approach, (i) Sonix itself can be used as a synchronizer. This way, at the beginning of the acquisition for each sector, it can send a pulse (output synch). This pulse can be used to restart the exciter. In another approach (ii) Sonix can wait for an external pulse before starting the acquisition for each sector (input synch). This way, an external synchronizer is required to control both Sonix and the exciter. To implement both of these techniques on Sonix machine, a custom sequencer needs to be implemented. The Texo SDK can be used for this.



Trade offs:
 * External hardware overhead is required
 * Observation time can be limited (number of scan lines that can be programmed for the sequencer is limited)
 * Both transducer and phantom/tissue should be stationary during the experiment
 * Acquisition lag still exists between individual lines in each sector