To investigate the feasibility of a noncontrast MRI framework combining velocity-selective arterial spin labeling (VSASL) and artificial neural network (ANN)-based gradient-echo sampling of spin-echo (GESSE) analysis for quantifying exercise-induced changes in calf muscle perfusion and oxygen metabolism. Ten healthy volunteers (mean age: 31.1 +/- 5.45 years) underwent MRI scans before and after standardized eccentric calf muscle exercise. Muscle blood flow (MBF) was assessed using VSASL with 3D GRASE readout, while muscle oxygen extraction fraction (MOEF) was estimated from GESSE data using an ANN-based method. Skeletal muscle oxygen consumption (SMVO(2)) was calculated based on Fick's principle. All maps were registered to T2-weighted images, and linear mixed-effects models were applied to evaluate the effects of time and muscle group. Following exercise, a significant increase in MBF was observed (Delta = 6.91 mL/100 g/min, p < 0.001), primarily in the gastrocnemius and soleus muscles. MOEF also increased significantly (Delta = 0.014, p = 0.029), with no significant differences across muscles. SMVO(2) showed a significant overall increase (Delta = 25.56 mumol O(2)/100 g/min, p < 0.001), with a significant interaction between time and muscle (p = 0.028). The greatest SMVO(2) increases were observed in the gastrocnemius (68.76% +/- 36.1%) and soleus (51.94% +/- 44.0%). This noncontrast approach using VSASL and ANN-based GESSE demonstrates the feasibility of spatially resolved quantification of calf muscle hemodynamics and oxygen metabolism. This method may serve as a valuable tool for assessing microvascular and metabolic function in both research and clinical practice.
Keywords
