This protocol uses 3D MRI, Doppler ultrasonography, and other advanced MR imaging techniques such as diffusion tensor MRI and tagging MRI to address several important issues. These include: (1) the compressibility of the human tongue and its common, yet untested, reference as a muscular hydrostat; (2) task-induced interactions between lingual musculature and vasculature and region-specific vascular demands; (3) changes in lingual fiber orientation, length, and strain distribution as a function of contraction tasks; and (4) effect of normal aging and disease processes on lingual myoarchitecture. By quantitatively addressing these issues, this protocol will contribute to a better understanding of the functional biomechanical as well as myoarchitectural intricacies of the in vivo human tongue. Our MRI data continue to show an average of about 8% (max = 12%) increase in tongue volume during maximal oropharyngeal voluntary isometric contraction. Our major accomplishment during 2003 was the validation of the MRI volume measurements. Validation studies were completed on three ex vivo models (two human and one calf). In these studies, measurements were compared between the volume measurements made from traced/rendered tongue MR images and the physical tongue volume across the factors of tracer (two trained biomedical engineering students), MR scanner (GE vs. Philips), scan type (current photon density protocol vs. higher resolution parameters), and imaging plane (sagittal, axial, and coronal). The physical volumes were determined from the excised tongues (excised after MR scanning of the whole head) using the water displacement method. Results showed no significant difference between the traced and the physical volumes as a function of scanner, scan type, or tracer. Tracer difference was significant for the coronal and axial imaging planes, but not for the sagittal plane (which is the plane for our current in vivo studies). Tracer training and segmentation practice were found to be crucial. With training and practice, mean volume measurement error (i.e., difference from the physical mean) ranged from 0.22% (+/- 2.53) to 1.13% (+/- 2.93) for the sagittal plane. In diffusion tensor MRI, we have encountered susceptibility artifacts. To overcome this problem, we are in the process of developing a new fast, sensitivity encoding pulse sequence (known as SENSE) for in vivo imaging. Using tagging MRI, however, we have been able to collect in vivo 2D and 3D data on task-induced changes in tissue strain maps. These data will supplement our current in vivo fast spin echo proton density MRI studies with dynamic information on regional strain distribution in lingual tissue and are likely to provide biomechanical insights into the mechanisms for tongue volume changes. Evidence from our blood flow studies continues to reveal a linear relationship between changes in intramuscular pressures resulting from varying degrees of muscle contraction and patterns of hyperemia. Further, different reperfusion patterns in functional tongue segments suggest regional specificity in intramuscular vascular demands. This appears to correspond with recent histological confirmation of region-specific fiber types in the intrinsic tongue muscles. In sum, this protocol will continue to evaluate the muscular hydrostat concept, elucidate the biomechanical intricacies of the in vivo human tongue, and gain insights into the biomechanical elements and limitations of rehabilitative lingual exercises in patients with weakened or compromised lingual function as well as preventive tongue strengthening exercises in adults undergoing normal aging.
