The purpose of this study was to examine the reliability and validity of the "panoramic" brightness mode ultrasonography (US) method to detect training-induced changes in muscle cross-sectional area (CSA) by comparison with results obtained using magnetic resonance imaging (MRI). Out of 27 young male volunteers, 20 subjects were assigned to training group and seven to non-training control group. Muscle CSAs of vastus lateralis were analyzed by MRI and US before and after 21 weeks of either heavy resistance training or control period. Measured by both the US and MRI, the resistance training induced significant increases (~13-14%, P < 0.001) in muscle CSA, whereas no changes were observed in control group. A high repeatability was found between the two consequent US measurements (intraclass correlation coefficient, ICC of 0.997) with standard error of measurement (SEM) of 0.38 cm(2) and smallest detectable difference of 1.1 cm(2). Validity of the US method against MRI in assessing CSA of VL produced ICC of 0.905 and SEM of 0.87 cm(2) with high limits of agreement analyzed by Bland and Altman method. However, the MRI produced systematically (10 +/- 4%, P < 0.01) larger CSA values than the US method. The US showed high agreement against MRI in detecting changes in muscle CSA (ICC of 0.929, SEM of 0.94 cm(2)). The results of this study showed that the panoramic US method provides repeatable measures of a muscle CSA although MRI produced larger absolute CSA values. Moreover, this US method detects training-induced changes in muscle CSA with a comparable degree of precision to MRI.