Application of Artificial Intelligence to Automate the Reconstruction of Muscle Cross-Sectional Area Obtained by Ultrasound.

PURPOSE: Manual reconstruction (MR) of the vastus lateralis (VL) muscle cross sectional area (CSA) from sequential ultrasound (US) images is accessible, reproducible and has concurrent validity with magnetic resonance imaging. However, this technique requires numerous controls and procedures during image acquisition and reconstruction, making it laborious and time-consuming. The aim of this study was to determine the concurrent validity of VL CSA assessments between MR and computer vision-based automatic reconstruction (AR) of CSA from sequential images of the VL obtained by US. METHODS: The images from each sequence were manually rotated to align the fascia between images and thus visualize the VL CSA. For the AR, an artificial neural network model was utilized to segment areas of interest in the image, such as skin, fascia, deep aponeurosis, and femur. This segmentation was crucial to impose necessary constraints for the main assembly phase. At this stage, an image registration application, combined with differential evolution, was employed to achieve appropriate adjustments between the images. Next, the VL CSA obtained from the MR (n = 488) and AR (n = 488) techniques were used to determine their concurrent validity. RESULTS: Our findings demonstrated a low coefficient of variation (CV) (1.51%) for AR compared to MR. The Bland-Altman plot showed low bias and close limits of agreement (+1.18 cm2, -1.19 cm2), containing more than 95% of the data points. CONCLUSIONS: The AR technique is valid compared to MR when measuring VL CSA in a heterogeneous sample.

Effects of Resistance Training Overload Progression Protocols on Strength and Muscle Mass.

The aim of this study was to compare the effects of progressive overload in resistance training on muscle strength and cross-sectional area (CSA) by specifically comparing the impact of increasing load (LOADprog) versus an increase in repetitions (REPSprog). We used a within-subject experimental design in which 39 previously untrained young persons (20 men and 19 women) had their legs randomized to LOADprog and REPSprog. Outcomes were assessed before and after 10 weeks of training. Muscle strength was assessed using the one repetition maximum (1RM) test on the leg extension exercise, and the CSA of the vastus lateralis was assessed by ultrasonography. Both protocols increased 1RM values from pre (LOADprog: 52.90±16.32 kg; REPSprog: 51.67±15.84 kg) to post (LOADprog: 69.05±18.55 kg, REPSprog: 66.82±17.95 kg), with no difference between them (P+>+0.05). Similarly, both protocols also increased in CSA values from pre (LOADprog: 21.34±4.71 cm²; REPSprog: 21.08±4.62 cm²) to post (LOADprog: 23.53±5.41 cm², REPSprog: 23.39±5.19 cm²), with no difference between them (P+>+0.05). In conclusion, our findings indicate that the progression of overload through load or repetitions can be used to promote gains in strength and muscle hypertrophy in young men and women in the early stages of training.

Effects of resistance training with controlled versus self-selected repetition duration on muscle mass and strength in untrained men.

The aim of this study was to compare the effect of self-selected repetition duration (SELF), with and without volume load (VL) equalized with controlled repetition duration (CON) on muscle strength and hypertrophy in untrained males. We used a within-subjects design in which 20 volunteers (age: 24.7 ± 2.9 years) had one leg randomly assigned to CON (i.e., 2 s concentric, 2 s eccentric) and the other to SELF or to self-selected repetition duration with equalized volume load (SELF-EV). One repetition maximum (1-RM) and muscle cross-sectional area (CSA) were measured at baseline (Pre) and after (Post) resistance training (RT; 2×/wk for 8 weeks). For the main study variables (1-RM and muscle CSA), a mixed-model analysis was performed, assuming repetition duration (SELF, SELF-EV and CON), and time (Pre and Post) as fixed factors and the subjects as random factor for each dependent variable (1-RM and CSA). All RT protocols showed significant increases in values of 1-RM from Pre (CON: 73.7 ± 17.6 kg; SELF: 75.9 ± 17.7 kg; and SELF-EV: 72.6 ± 16.9 kg) to Post (CON: 83.4 ± 19.9 kg, effect size (ES): 0.47; SELF: 84 ± 19.1 kg, ES: 0.43; and SELF-EV: 83.2 ± 19.9 kg, ES: 0.57, P < 0.0001). Muscle CSA values increased for all protocols from Pre (CON: 12.09 ± 3.14 cm2; SELF: 11.91 ± 3.71 cm2; and SELF-EV: 11.93 ± 2.32 cm2) to Post (CON: 13.03 ± 3.25 cm2, ES: 0.29; SELF: 13.2 ± 4.16 cm2, ES: 0.32; and SELF-EV: 13.2 ± 2.35 cm2, ES: 0.53, P < 0.0001). No significant differences between protocols were found for both 1-RM and CSA (P > 0.05). Performing RT with SELF, regardless of VL, was equally effective in inducing increases in muscle strength and hypertrophy compared to CON in untrained men.