TY - JOUR
T1 - Comparison of MPEG digital image compression with super VHS for diagnostic echocardiographic readings
AU - Brar, Rupinder
AU - Yurow, Gary
AU - Neumann, Alex
AU - Stamos, Thomas
AU - Garcia, Mario
AU - Stoddard, Marcus
AU - Cherian, P. K.
AU - Thomas, James D.
AU - Soble, Jeffrey S.
PY - 1997
Y1 - 1997
N2 - Image compression is important to digital echocardiography, given network and storage constraints, but may adversely affect image quality. We investigated the diagnostic quality of the MPEG-1 (Motion Picture Experts Group) digital video compression standard relative to superVHS video (sVHS). Methods: 60 consecutive clinical studies (M-mode, 2D, spectral and color Doppler) were recorded simultaneously by sVHS and MPEG-1 (1.36 megabits/sec) Each study pair was reviewed and reported using a standardized computer system, containing over 500 possible diagnostic responses across the entire range of echo findings, with a >2 week interval between viewing the corresponding MPEG and sVHS recordings. In 50 study pairs, discrepancies were analyzed by two consensus readers who reviewed both the MPEG and sVHS recordings blinded to the origin of the inital reader's results. Results: On average, there were 23 diagnostic responses per study, with an overall exact concordance rate between MPEG and sVHS of approximately 84%. Of 207 discrepancies in the subset of 50 paired readings, consensus review agreed with MPEG in 106 cases and with sVHS in 101 cases (p=0.79). Of 13 discrepancies involving regional LV wall motion, consensus review agreed with MPEG in 7 cases and sVHS in 6 cases (p=1.0). Conclusion: MPEG-1 video compression yields overall diagnostic echo image quality equivalent to sVHS. Given the narrow bandwidth requirements (high compression ratio) and widespread availability of MPEG encoders and decoders, this represents an important compression strategy for digital echocardiography.
AB - Image compression is important to digital echocardiography, given network and storage constraints, but may adversely affect image quality. We investigated the diagnostic quality of the MPEG-1 (Motion Picture Experts Group) digital video compression standard relative to superVHS video (sVHS). Methods: 60 consecutive clinical studies (M-mode, 2D, spectral and color Doppler) were recorded simultaneously by sVHS and MPEG-1 (1.36 megabits/sec) Each study pair was reviewed and reported using a standardized computer system, containing over 500 possible diagnostic responses across the entire range of echo findings, with a >2 week interval between viewing the corresponding MPEG and sVHS recordings. In 50 study pairs, discrepancies were analyzed by two consensus readers who reviewed both the MPEG and sVHS recordings blinded to the origin of the inital reader's results. Results: On average, there were 23 diagnostic responses per study, with an overall exact concordance rate between MPEG and sVHS of approximately 84%. Of 207 discrepancies in the subset of 50 paired readings, consensus review agreed with MPEG in 106 cases and with sVHS in 101 cases (p=0.79). Of 13 discrepancies involving regional LV wall motion, consensus review agreed with MPEG in 7 cases and sVHS in 6 cases (p=1.0). Conclusion: MPEG-1 video compression yields overall diagnostic echo image quality equivalent to sVHS. Given the narrow bandwidth requirements (high compression ratio) and widespread availability of MPEG encoders and decoders, this represents an important compression strategy for digital echocardiography.
UR - http://www.scopus.com/inward/record.url?scp=2342574665&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=2342574665&partnerID=8YFLogxK
M3 - Article
AN - SCOPUS:2342574665
SN - 0894-7317
VL - 10
JO - Journal of the American Society of Echocardiography
JF - Journal of the American Society of Echocardiography
IS - 4
ER -