Because of the enormous size of uncompressed medical ultrasound studies, transmission and storage are currently limited to individual still-frame images. Motion ultrasound in an analog format is not well stored, retransmitted, manipulated, accessed, or displayed. Thus, ultrasound is mostly a procedure for local use only, even though it would clearly be desirable to share motion studies across an enterprise. Placing motion ultrasound into a digital format would greatly improve storage efficiency, data access, and consultative telemedicine applications. Research into a new approach to coding ultrasound video has produced an effective algorithm that performs substantially better than traditional video coding algorithms. Initial experiments indicate the possibility of capturing, compressing, storing and transmitting real-time ultrasound at 30 frames per second over standard ISDN lines. This project will continue to refine the algorithm to achieve greater efficiency and quality, to adjust coding parameters to different classes of ultrasound studies, and to determine how well the algorithm performs when compared to standard encoding technologies. The algorithm will eventually be integrated into the currently marketed Windows NT-based teleradiology software, WinRad. We expect to gain further experience that can be transferred to the compression of other types of moving medical imaging studies.
Any application in which monitoring of real-time ultrasound studies from distant sites is needed. This includes fixed stations such as between referral hospitals and outlying clinics, or mobile stations, as between ambulance to hospital, or battlefront to combat support hospital. The greatest value of this technology, however, is in archiving cardiac ultrasound, in a compact, inexpensive digital format for store-and- forward applications.
