Smartphone provides exact renditions of adult muscle mass

Smartphones now have the capacity to accurately render an adult’s level of muscularity through their image capture capabilities in combination with device software applications, a recent study has shown.

“Skeletal muscle is a large and clinically relevant body component that has been difficult and impractical to quantify outside of specialized facilities,” the researchers said. “Advances in smartphone technology now provide the opportunity to quantify multiple body surface dimensions such as circumferences, lengths, surface areas, and volumes.”

The study included a total of 322 adults, whose appendicular lean mass (ALM) was measured by dual-energy X-ray absorptiometry (DXA), serving as the reference for muscularity. In addition, their digital anthropometric dimensions (eg, circumferences, lengths, and regional and total body surface areas and volumes) were measured using a 20-camera 3D imaging system.

The researchers developed the ALM prediction equations in selected adults using the least absolute shrinkage and selection operator (LASSO) regression. These models were then tested on the remaining participants.

Subsequently, the researchers cross-validated the accuracy of the prediction models in a second independent sample of 53 adults who underwent ALM estimation by DXA and the same digital anthropometric estimates taken with a smartphone application.

Several significant demographic and 3D digital anthropometric predictor variables were included in the LASSO models. Assessment of these models in the testing subgroup showed root mean square errors (RMSEs) of 1.56 kg in women and 1.53 kg in men, as well as R²’s of 0.74 and 0.90, respectively. [Am J Clin Nutr 2023;117:794-801]

On cross-validation of the LASSO models in the smartphone application group, the RMSEs were 1.78 kg in women and 1.50 kg in men, with R²’s of 0.79 and 0.95, respectively. No significant differences or bias was observed between the measured and predicted ALM values.

“Smartphone image capture capabilities combined with device software applications can now provide accurate renditions of the adult muscularity phenotype outside of specialized laboratory facilities,” the researchers said. 

An earlier study also evaluated the performance of a novel automated computer vision method, visual body composition (VBC), that uses 2D photographs captured by a smartphone camera to estimate the percentage total body fat (%BF). The VBC algorithm was based on a convolutional neural network (CNN).

Results of the validation study showed that VBC body fat estimates were accurate and showed no bias when compared with DXA as the reference. Additionally, VBC exceeded other bioimpedance analysis and air displacement plethysmography methods evaluated. [NPJ Digit Med 2022;5:79]

VBC also had the lowest mean absolute error and standard deviation (2.16±1.54 percent) relative to the other methods (p<0.05 for all comparisons), while the %BF showed good concordance with DXA (Lin’s concordance correlation coefficient, 0.96 for all, 0.93 in women, 0.94 in men). [NPJ Digit Med 2022;5:79]

“The wide availability of smartphones suggests that the VBC method for evaluating %BF could play an important role in quantifying adiposity levels in a wide range of settings,” the authors said.