Shape factor versus truncated cone-based quantification of quadriceps and hamstring muscle volumes-A choice between accuracy and precision

This study aimed to determine the average location of maximal anatomical cross-sectional area of the quadriceps and hamstrings and to investigate the agreement of different muscle volume estimation methods. Magnetic resonance imaging datasets were acquired from 39 soccer players. Muscle volumes were calculated using slice-by-slice segmentation and compared with the shape factor and truncated cone-based estimates. Descriptive data were expressed as means ± standard deviations, and Bland-Altman plots were used for agreement analyses. The average location of maximal anatomical cross-sectional area was at 61 ± 10%, 64 ± 10%, 29 ± 3%, and 56 ± 6% for the vastus lateralis, rectus femoris, vastus medialis, and vastus intermedius, respectively. For the hamstrings, the maximal anatomical cross-sectional area was at 45 ± 3%, 48 ± 9%, 58 ± 7%, and 38 ± 8% for the biceps femoris short head, biceps femoris long head, semitendinosus, and semimembranosus, respectively. Relative biases ranged from 2% to 9% for the shape factor and from 6% to 14% for the truncated cone method. The ranges of agreement were -15% to 29% and -1% to 19%, respectively. The shape factor method showed better accuracy, while the truncated cone method displayed better precision.