Abstract 6113: A New Noninvasive Clinical Method for Combined Quantification of Cardiac and Vascular Mechanical Function - Pixel Based Multimodality Tissue Tracking (MMTT)
Background: We introduce and validate pixel based multimodality tissue tracking (MMTT) as a new noninvasive method for quantification of atrial, cardiac and vascular tissue deformation from magnetic resonance (MR), multidetector computer tomography (MDCT) and ultrasound recordings.
Method: 25 patients and 10 healthy volunteers were studied with the latter undergoing untagged and tagged MRI (1.5 T), echocardiography and carotid artery short-axis ultrasound examinations, while patient MDCT (64-slice) and untagged and tagged MR recordings were selected from our database of patients with ischemic heart disease. Longitudinal (LV, RV and atriums) and circumferential strain (LV, RV, aorta and carotid artery) were quantified using automated frame-to-frame tracking of pixel patterns visible in tagged MR (tags), untagged MR and MDCT (border contours), and ultrasound (speckle) cine recordings (Fig 1⇓). LV tagged MR HARP analysis was used for validation.
Results: There were good correlations and agreements for pooled LV strains between the reference method and MMTT analyses of tagged MR (r=0.83, P<0.001 and −0.9±3.8%), non-tagged MR (r=0.77, P<0.001 and 0.8±4.6%), MDCT (r=0.64, P<0.001 and 0.4±5.5%) and echocardiography (r=0.32, P<0.001 and 0.7±4.7%). RV, atrial and aortic strains by MMTT also demonstrated good correlation and agreement between imaging modalities (Fig⇓).
Conclusion: This study demonstrates that MMTT can quantify cardiac and vascular deformation from non-tagged and tagged MRI, MDCT and ultrasound recordings, representing a new universal method for comprehensive and integrated clinical assessment of cardiovascular function and mechanical interaction.