Rapid and accurate left ventricular chamber quantification using a novel CMR segmentation algorithm: A clinical validation study

PURPOSE: To evaluate the clinical performance of a novel automated left ventricle (LV) segmentation algorithm (LV-METRIC) that involves no geometric assumptions. MATERIALS AND METHODS: LV-METRIC and manual tracing (MT) were used independently to quantify LV volumes and LVEF (ejection fraction) for 151 consecutive patients who underwent cine-CMR (steady-state free precession). Phase contrast imaging was used to independently measure stroke volume. RESULTS: LV-METRIC was successful in all cases. Mean LVEF was within 1 point of MT (Delta 0.6 +/- 2.3%, P < 0.05), with smaller differences among patients with (0.5 +/- 2.5%) versus those without (0.9 +/- 2.3%; P = 0.01) advanced systolic dysfunction (LVEF <or= 35% by MT). LV volumes by LV-METRIC were slightly smaller than MT during end-diastole (3.9 +/- 6.8 mL, P < 0.001) and end-systole (1.4 +/- 5.5 mL, P < 0.01). Mean processing time was 22 +/- 13 seconds for LV-METRIC and 4:59 +/- 1:56 minutes for MT (P < 0.001). Processing time correlated with LV blood volume by MT (r = 0.43) and LV-METRIC (r = 0.55), but slope was 10-fold steeper for MT (0.02 vs. 0.001), indicating greater proportionate time increases in relation to chamber dilation. Compared to stroke volume by phase contrast, LV-METRIC yielded smaller differences (0.3 +/- 18.3 mL) than MT (2.5 +/- 17.2 mL; P < 0.001). CONCLUSION: Among a broad series of consecutive patients undergoing CMR, automated LVEF by LV-METRIC was within 1 point of MT with processing time reduced 14-fold. Stroke volume by LV-METRIC yielded improved agreement with an independent standard of phase contrast imaging.