Accurate assessment of resting metabolic rate (RMR) is necessary for calorie-based recommendations in diet and exercise training interventions. BodyMetrix™ is an ultrasound-based device that provides an estimate of RMR based on body composition, but has not been proven valid or reliable. Therefore, we evaluated the agreement between Katch-McArdle prediction equation used by BodyMetrix™, with indirect calorimetry, Harris-Benedict, WHO, and Sabounchi prediction equations of RMR. In total, 32 men and 22 women were measured for body composition via BodyMetrix™ and RMR via indirect calorimetry. All prediction equations demonstrated significantly lower RMR values (p < 0.001) relative to indirect calorimetry. Katch-McArdle equation strongly correlated with other prediction equations (p < 0.001), and had a moderate (r = 0.658, p < 0.001) correlation with indirect calorimetry. There was a tendency toward underestimation for obese individuals. Therefore, we suggest that estimates from BodyMetrix™ may be used as a relative, rather than an absolute measure of RMR.