Diffusion tensor imaging (DTI) is a magnetic resonance imaging (MRI) method that resolves structures by three-dimensional measurement of water movement, and has shown to benefit studies of anisotropic tissues such as brain and skeletal muscle. While many studies have used DTI to non-invasively study static tissue architecture, little attempt has been made to use this technique to temporally characterize muscle during post-exercise recovery. Thus, the goal of this work was to use DTI to study the timecourse of changes in skeletal muscle following exercise. The first study was performed to test stability of DTI eigenvectors (ε1 ε2 ε3), and to determine how time-expensive parameters such as increased number of diffusion directions (NDD) or signal averages (NSA) improve vector stability. The ε2 vector was found to have more directional variability than ε1, and showed less improvement than ε1 with increased NDD or NSA. Furthermore, decreasing directional variability of ε1 was correlated with increasing NDD, not NSA (p<0.0008), while decreased variability of ε2 was correlated with increasing NSA, not NDD (p<0.0005). The variation in ε2 indicated that combining the corresponding minor eigenvalues into a measure of Radial Diffusivity is more robust than analyzing λ2 and λ3 eigenvalues separately. The second study tested the use of DTI to characterize temporal calf muscle changes following a mild in-bore dorsiflexion-eversion exercise. DTI volumes were acquired before and immediately after exercise. Anterior tibialis (ATIB), extensor digitorum longus (EDL), and peroneus longus (PER) showed significantly-elevated mean diffusivity (MD) post-exercise, while soleus (SOL) and lateral gastrocnemius (LG) did not (p<0.0001). The EDL showed greater initial MD increase and remained significantly elevated across more time points than ATIB or PER (p<0.05 to p=7.41x10−10). Significant signal increases were observed in post-exercise EDL b=0s/mm2 volumes (S0) relative to other muscles across the majority of timepoints (p<0.01 to p<0.001). The notable differences of EDL temporal MD and S0 relative to ATIB and PER may be related to the physiology of the increased Type-II fiber content in this muscle. The third and final study investigated the feasibility of a ’sliding window’ multiple-timescale temporal DTI approach, intended to acquire data with high temporal resolution and ongoing structural representation. Continuous diffusion data was acquired in the calf before and after four plantarflexion tasks, which varied by number of flexions (10 or 60), and weight load (10% or 40% of individual max). Apparent diffusion coefficient (ADC) and S0 were calculated from 3-direction subunits, while 15-direction subunits produced DTI measures such as mean diffusivity (MD). Four different post-exercise temporal patterns were observed for ADC, S0, and MD amongst the measured muscles: ’elevated-decline’, ’latent peak’, ’sub-to-peak’, and ’horizontal’. The 10-flex 10% condition elicited ’elevated-decline’ in active muscles, particularly SOLlat. Exercise of greater intensity produced ’latent peak’ and ’sub-to-peak’ patterns, with peak height related to greater workload. The 10-flex 40% trial produced a ’sub-to-peak’ pattern across all subjects only in the LG and MG, but ’latent-peak’ in these muscles 60-flex 40%. The specificity of temporal diffusion patterns according to muscle and task indicate that this technique could be beneficial to future studies of muscle function. These experiments have demonstrated the limits of DTI in the study of skeletal muscle, yet established a basis for future investigation of muscle dynamics using temporal diffusion methods. / Dissertation / Doctor of Philosophy (PhD)
Identifer | oai:union.ndltd.org:mcmaster.ca/oai:macsphere.mcmaster.ca:11375/20805 |
Date | January 2017 |
Creators | Rockel, Conrad P |
Contributors | Noseworthy, Michael D, Biomedical Engineering |
Source Sets | McMaster University |
Language | English |
Detected Language | English |
Type | Thesis |
Page generated in 0.0017 seconds