IntroductionSpaceflight results in marked muscle atrophy and a corresponding loss in muscle strength. Methods to evaluate spaceflight-induced changes to muscle size and strength are difficult to achieve within the tight confines of spacecrafts. Therefore, convenient methods to non-invasively monitor muscle that are predictive of actual muscle function are critical for future missions to ensure the health and safety of crew members. PurposeEvaluate the ability of non-invasive in vivo electrical impedance myography (EIM) combined with statistical models to predict muscle size and strength in rodent models of micro- and partial gravity. MethodsMale and female Fisher rats (N = 120), half gonadectomized, were divided into three different weightbearing (WB) conditions of 40 animals each, including 0%WB (0 % of weight-bearing, simulated microgravity), 40%WB (40 % of weight-bearing, simulated Martian gravity), and 100%WB (100 % of weight-bearing, full weight-bearing controls). Rats remained in designated interventions for 28 days. Afterward, rats underwent a series of musculoskeletal strength assessments and measurement of EIM. Rats were then euthanized and gastrocnemius tissues collected. Machine-learning (ML) algorithms were applied to full spectrum EIM data to predict various muscle parameters. Results40%WB and 0%WB rats had lower grip strength and plantar flexion compared to 100%WB. Correspondingly, 40%WB and 0%WB also had reduced gastrocnemius mass compared to 100%WB. EIM resistance values demonstrated a dose-dependency response, with greater resistance values associated with reduced gravitational load. ML-enhanced EIM yielded strong predictions of muscle plantar flexion force and muscle mass, with root mean squared errors of 18.4 % of 22.0 %, and R2 values of 0.87 an 0.88, respectively. ConclusionML-enhanced EIM may be a helpful tool to non-invasively monitor muscle changes predictive of muscle force production and mass during exposure to reduced gravity.
Read full abstract