Rationale and objectives: The authors investigated the diagnostic efficacy of compression of real-time ultrasound (US) examinations.
Methods: Low- and high-compression recordings (9:1 and 15:1, respectively) of examinations were generated by using Joint Photographic Experts Group algorithms. Seven radiologists used a five-level response scale to answer questions about the presence, number, and location of focal fibroid tumors in 67 randomly sorted uterine examinations. The images were viewed after no, low, and high compression. Results were evaluated by using multipatient, multireader receiver operating characteristic jack-knife analysis.
Results: Given the reduction in the US digital video rates from 74 Mbit/sec for uncompressed images to 8 Mbit/sec for low compression and 4.7 Mbit/sec for high compression, there were no statistically significant differences in accuracy between the compression schemes. Confidence intervals suggested that the sample size was adequate.
Conclusion: Compressed images with compression ratios of 9:1 and 15:1 were diagnostically equivalent to uncompressed images of uterine fibroid tumors.