Automated segmentation of routine cine-CMR for volumetric assessment of presence and severity of left ventricular diasto
- PDF / 653,442 Bytes
- 2 Pages / 610 x 792 pts Page_size
- 27 Downloads / 155 Views
BioMed Central
Open Access
Poster presentation
Automated segmentation of routine cine-CMR for volumetric assessment of presence and severity of left ventricular diastolic dysfunction Dorinna D Mendoza*, Noel CF Codella, Sonia Sethi, Keigo Kawaji, Shant Manoushagian, Martin R Prince, Richard B Devereux, James K Min, Matthew D Cham, Yi Wang and Jonathan W Weinsaft Address: Weill Cornell Medical College, New York, NY, USA * Corresponding author
from 13th Annual SCMR Scientific Sessions Phoenix, AZ, USA. 21-24 January 2010 Published: 21 January 2010 Journal of Cardiovascular Magnetic Resonance 2010, 12(Suppl 1):P86
doi:10.1186/1532-429X-12-S1-P86
Abstracts of the 13th Annual SCMR Scientific Sessions - 2010
Meeting abstracts - A single PDF containing all abstracts in this Supplement is available here. http://www.biomedcentral.com/content/files/pdf/1532-429X-11-S1-infoThis abstract is available from: http://jcmr-online.com/content/12/S1/P86 © 2010 Mendoza et al; licensee BioMed Central Ltd.
Objective To compare automated volumetric segmentation of routine cine-CMR to echocardiography (echo) for assessment of LV diastolic function.
Introduction DD alters LV relaxation and compliance, thereby changing timing profiles of LV filling. Cine-CMR provides volumetric data concerning LV filling but is not widely used to assess DD as manual planimetry of all images across all temporal phases requires tracing over one hundred images per exam. Thus, whereas cine-CMR is a reference for systolic function, additional testing using echo is typically necessary to identify DD. We tested whether automated segmentation of routine cine-CMR could be used to stratify presence and severity of DD.
Methods The population consisted of consecutive post-myocardial infarction patients who prospectively underwent CMR (1.5 T, SSFP) and echo within 1 day. CMR and echo were independently interpreted. Echo was the reference for DD (graded using American Society of Echocardiography criteria). Cine-CMR assessed LV filling using our group's previously validated automated segmentation algorithm (LVMETRIC) - volumetric segmentation of all LV short-axis slices across all temporal phases generated a volumetric-
filling profile. Three parameters were tested: Peak Filling Rate (PFR) - maximal Ävolume/Äphase. Time to Peak Filling Rate (TPFR) - interval between end-systole and peak filling. Diastolic Volume Recovery (DVR) - proportion of diastole required to recover a given percent (80%) stroke volume.
Results 113 patients were studied (57 ± 13 years, 43%HTN, LVEF = 54 ± 13%). 64% had echo-evidenced DD (25% grade 1, 63% grade 2, 12% grade 3). LV-METRIC successfully generated LV filling curves (190 ± 36 images/exam) in 99% patients (1 = arrhythmia). Patients with DD had longer DVR (73 ± 13 vs. 66 ± 16%, p = 0.01), longer TPFR (239 ± 164 vs. 177 ± 123 msec, p = 0.03), and a trend towards lower PFR (234 ± 83 vs. 265 ± 77 ml/sec, p = 0.07) vs. those without DD by echo. TPFR by cine-CMR correlated with echo deceleration time by echo (r = 0.25, p = 0.
Data Loading...
