PURPOSE: To quantify the cervical muscle volume variation by means of three-dimensional reconstruction from MRI images. MATERIALS AND METHODS: Sixteen subjects were scanned using a Philips MRI scanner, including 11 men and 5 women, aged from 23 to 33 years, weighting between 49-80 kg. The deformation of a parametric specific object method was used to develop three-dimensional muscle models from contours on a small number of MRI images. Six subjects were reconstructed by two observers for evaluating the reliability by means of intraclass correlation coefficient (ICC). The results were also compared with in vivo measurement on a single specimen from a reference literature. The difference in left and right muscles volumes was assessed with a paired Wilcoxon signed rank test. RESULTS: The results showed good reliability by means of ICC study and were consistent with the in vivo specimen measurements. The left and right paired muscle volumes showed no significant difference. Interindividual variance was large that could reach 364 cm(3) , but the ratio of a given muscle volume to the total volume was less variable, always lower than 13%. The maximum cross sectional areas of cervical muscles varied greatly between individuals and the maximum values were mostly found at the C6-C7 level. CONCLUSION: This study provides initial results which could be used as reference data for clinical evaluation and biomechanical model development.
PURPOSE: To quantify the cervical muscle volume variation by means of three-dimensional reconstruction from MRI images. MATERIALS AND METHODS: Sixteen subjects were scanned using a Philips MRI scanner, including 11 men and 5 women, aged from 23 to 33 years, weighting between 49-80 kg. The deformation of a parametric specific object method was used to develop three-dimensional muscle models from contours on a small number of MRI images. Six subjects were reconstructed by two observers for evaluating the reliability by means of intraclass correlation coefficient (ICC). The results were also compared with in vivo measurement on a single specimen from a reference literature. The difference in left and right muscles volumes was assessed with a paired Wilcoxon signed rank test. RESULTS: The results showed good reliability by means of ICC study and were consistent with the in vivo specimen measurements. The left and right paired muscle volumes showed no significant difference. Interindividual variance was large that could reach 364 cm(3) , but the ratio of a given muscle volume to the total volume was less variable, always lower than 13%. The maximum cross sectional areas of cervical muscles varied greatly between individuals and the maximum values were mostly found at the C6-C7 level. CONCLUSION: This study provides initial results which could be used as reference data for clinical evaluation and biomechanical model development.