Analyzing ultrasound images
Measurements of the diameter of human blood vessels are often used to determine and treat abnormal vascular functions. In such measurements, the boundaries of ultrasonic images are typically monitored over a period of time and recorded along with a corresponding electrocardiogram...
Measurements of the diameter of human blood vessels are often used to determine and treat abnormal vascular functions. In such measurements, the boundaries of ultrasonic images are typically monitored over a period of time and recorded along with a corresponding electrocardiogram (EKG) trace on a video recorder. Later, manual analysis on a frame-by-frame basis is performed to manually measure the blood-vessel diameter using electronic calipers. Not only is this process inefficient, but it also results in a high inter- and intraobserver variability.
To overcome this variability, Icon Technologies (Victoria Park, Western Australia) has developed an automated imaging system. It improves the speed, precision, and objectivity of the analysis of videotapes containing ultrasound images that have captured changes in blood-vessel diameter.
To digitize the original images, the videotape is replayed through a SR-S388E time-code video recorder from JVC (Wayne, NJ) under RS-232 serial control. Using LabVIEW software from National Instruments (NI; Austin, TX), a driver library for the video recorder implements the most common control commands. Images are then acquired from tape to a 450-MHz Dual-Pentium III PC via an NI PCI-1407 monochrome acquisition card at rates of up to 17 frames/s.
During imaging-system operation, a technician uses the PC mouse to outline the general area on the screen that includes the blood vessel of interest, specify the starting and ending time codes, and then initiate the analysis procedure. After this procedure is activated, the system requires no further input from the technician. Video frames are acquired continuously between the two defined time codes on the tape, and the application uses predefined masks to look for the EKG trace in different regions of the image.
Using image-processing algorithms developed with NI LabVIEW, the EKG trace is auto-thresholded, and a peak-detection algorithm is applied to determine if the frame occurred within a specified window of the first positive wave (R-wave) in the EKG. Frames that do not meet this criterion are discarded, and a new frame is acquired and tested. Frames that do meet this criterion are passed to the second stage of the image-processing process, where a technician-defined region of interest is thresholded to define the vessel boundaries.
According to Mark Trotman, managing director of Icon Technologies, the entire analysis of the EKG R-wave and the blood-vessel diameter is accomplished via a sequence of three image-processing tools from the IMAQ Vision Toolkit. Measurement of the blood-vessel diameters is objective and reproducible with 6.7% precision.