Application of the novel estimation method by shear wave elastography using vibrator to human skeletal muscle

Objective measurement of muscle elasticity can assist in confirming pathological states and monitoring treatment in conditions associated with altered skeletal muscle stiffness (e.g., spasticity after stroke, spinal cord injury, multiple sclerosis, myopathy, myofascial pain syndrome, neck-shoulder pain, and low back pain). Conventional clinical assessments (palpation, ramp-and-hold tests, pendulum tests, dynamometry) provide valuable information but cannot isolate mechanical properties of individual muscles from surrounding tissues. Elastic modulus (Young’s modulus) is a key parameter describing soft tissue stiffness. Strain elastography provides semiquantitative assessment by applying repetitive compression with an ultrasound probe; it lacks absolute quantification of Young’s modulus, is operator-dependent, and repeated preload can alter elasticity. Shear wave elastography enables quantitative estimation because shear wave propagation speed depends on tissue stiffness and can be measured using Doppler ultrasonography. Shear waves can be generated by external mechanical vibration or acoustic radiation force impulses (ARFI). ARFI-based systems are common but require strong ultrasound, may cause local heating near bone, and need high frame-rate imaging, making systems expensive and limiting accessibility. External mechanical vibration is a cost-effective alternative but imaging at appropriate depth and cumbersome hardware have limited adoption. Reflections and complex wave propagation can reduce velocity estimation accuracy; real-time confirmation of the shear wave wavefront (direction, reflection, refraction) could improve quantitative measurements. Color Doppler shear wave imaging is a novel wavefront imaging method using ultrasound color flow images to directly and in real time observe continuous shear waves generated by a mechanical vibrator. Adding a simple device to a conventional ultrasound system can visualize tissue stiffness of superficial tissues. However, detailed adaptation to skeletal muscle has not been investigated. This study examines the reproducibility and reliability of a novel estimation method for vibration-based shear wave elastography of human skeletal muscle.

The paper reviews musculoskeletal elastography methods: strain elastography offers semiquantitative stiffness assessment but is operator dependent and cannot yield absolute Young’s modulus. Shear wave elastography provides quantitative assessment, with shear waves generated either by external mechanical vibration or ARFI. ARFI-based systems (e.g., Supersonic, Philips, GE, Siemens, Canon, Mindray, Samsung) are widely used but are expensive, require strong ultrasound, and can lead to undesirable temperature elevation near bone; high frame rates are typically needed. External mechanical vibration-based methods (vibro-elastography) have advanced and shown applications in prostate cancer and liver fibrosis. However, widely available methods for real-time visualization of wavefronts with conventional external vibration have been lacking. Reflections from tissue boundaries can produce standing waves and reduce accuracy. Color Doppler shear wave imaging has been proposed to visualize wavefronts in real time using conventional ultrasound color flow imaging. Prior work by the authors showed feasibility and consistency of shear wave velocities measured by color Doppler shear wave imaging compared with ARFI-based methods in phantom and superficial muscles. The current work extends these approaches to assess reproducibility and reliability in skeletal muscle.

Experimental system: The novel estimation method for shear wave elastography was implemented on a Toshiba Xario SSA-660A ultrasound scanner. Continuous shear waves were generated by a short (45 mm), lightweight (30 g) linear vibration motor (electric toothbrush type) producing >400 µm displacement in the 230–300 Hz range. The selected shear wave frequency (235.8–296.6 Hz) satisfied the shear wave frequency condition (m = 1), balancing spatial resolution and actuator output. Motor displacement was measured with a MEMS accelerometer (ADXL001) at 200 kHz sampling; displacement was obtained via double integration. Harmonic distortion was negligible (second and third harmonics −44.5 dB and −79.9 dB at 276 Hz with 300 µm amplitude), indicating a near-sinusoidal output. Color flow images were recorded for 2 s, captured to a PC, and processed with Fourier analysis and a directional filter to reconstruct shear wave phase, velocity, and propagation maps. Reconstruction time was <1 s, and maps were updated every ~4 s. Participants: Fourteen healthy male college students (mean ± SD: age 21.7 ± 0.9 years; height 170.7 ± 6.6 cm; body mass 63.5 ± 11.2 kg; BMI 21.8 ± 3.3 kg/m²) volunteered. Exclusion criteria included lower-extremity contracture; history of back/lower-extremity surgery; neurological disorders; use of hormones or muscle-affecting drugs; and regular participation in competitive sports or resistance/aerobic/flexibility training. Participants maintained usual diet and avoided vigorous activity the day before and on the day of testing. Ethics approval was obtained; written informed consent was provided. Experimental design and measurement protocol: Shear wave velocities were measured in phantom gels (konjac and agar) and skeletal muscles twice per day for two days (inter-measurement interval of 1 day). For phantom measurements, two independent investigators performed measurements. Agar phantoms were prepared at 1%, 2%, 3%, 4%, and 8% concentrations by mixing powdered agar with sterilized milk (recipes specified: 1%: 4 g agar/396 ml milk; 2%: 8 g/392 ml; 3%: 12 g/388 ml; 4%: 16 g/384 ml; 8%: 32 g/368 ml). The ultrasound probe was placed directly over the measurement point; the vibrator excitation point was positioned adjacent to the probe. Probe and actuator positions were adjusted while monitoring shear wave wavefronts to ensure clear propagation. Investigators stabilized their posture (e.g., elbows on bed) to apply consistent force and performed 3–5 repeated measurements per trial. For skeletal muscle, the probe and excitation point were aligned parallel to the long axis of muscle fibers to ensure propagation along the fiber direction. The middle portion of each muscle was examined using surface landmark references consistent with standardized EMG electrode placement. Measurement sites for probe and vibrator were marked on the skin to ensure consistency across sessions; the initial B-mode image served as a reference to replicate fascia, bone, nerve, and vessel landmarks. Muscles assessed on the right side: biceps brachii, flexor carpi radialis, semitendinosus, biceps femoris, medial gastrocnemius, and tibialis anterior. For semitendinosus and biceps femoris, analyses focused on the upper half of the selected area. Statistical analysis: Normality was assessed with the Shapiro–Wilk test. Reproducibility and reliability were summarized using coefficient of variation (CV). Agar concentration–velocity relationships were assessed with Pearson correlation. Paired t-tests compared conditions (alpha 0.05). Intra-class correlation coefficients ICC(1,1) and ICC(2,1) with 95% CIs quantified intra-day, day-to-day, and inter-operator reliabilities. Reliability was interpreted using Landis and Koch guidelines.

Phantom gels:

• Konjac jelly: Mean shear wave velocities were 2.21 ± 0.03 m/s (first measurement; CV 1.38%) and 2.21 ± 0.01 m/s (second; CV 0.38%) on day 1; 2.21 ± 0.02 m/s (first; CV 0.83%) and 2.28 ± 0.02 m/s (second; CV 0.78%) on day 2. Summary CVs: intra-day 0.2–0.6%; day-to-day 0.6–0.7%; inter-operator 0.5–0.7%.
• Agar gels (1–8%): Shear wave velocity increased with agar concentration; strong positive correlation r = 0.99–0.994 (Figure 2 shows r = 0.994; regression y = 0.3485x + 2.8894). Reliability metrics (Table 1): • Intra-day: ICC 0.98 (95% CI 0.80–1.00), CV 1.8–4.8%. • Day-to-day: ICC 0.96 (95% CI 0.65–1.00), CV 0.6–7.7%. • Inter-operator: ICC 0.98 (95% CI 0.65–1.00), CV 1.6–4.0%. Skeletal muscle (right side; biceps brachii, flexor carpi radialis, semitendinosus, biceps femoris, medial gastrocnemius, tibialis anterior): Overall intra-day CV < 9.8% with ICC > 0.92; day-to-day CV < 9.6% with ICC > 0.90. Per-muscle reliability (Figure 3):
• Biceps brachii: Intra-day ICC 0.93 (0.81–0.98), CV 0.3–4.5%; Day-to-day ICC 0.95 (0.85–0.98), CV 0.1–3.8%.
• Flexor carpi radialis: Intra-day ICC 0.92 (0.79–0.98), CV 0.1–4.8%; Day-to-day ICC 0.92 (0.79–0.98), CV 0.2–4.8%.
• Semitendinosus: Intra-day ICC 0.96 (0.88–0.99), CV 0.5–7.4%; Day-to-day ICC 0.95 (0.86–0.98), CV 0.2–7.8%.
• Biceps femoris: Intra-day ICC 0.92 (0.78–0.97), CV 0.6–9.8%; Day-to-day ICC 0.90 (0.72–0.97), CV 0.2–9.6%.
• Medial gastrocnemius: Intra-day ICC 0.99 (0.97–1.00), CV 0.3–4.1%; Day-to-day ICC 0.99 (0.97–1.00), CV 0.7–4.5%.
• Tibialis anterior: Intra-day ICC 0.98 (0.95–1.00), CV 0.5–9.0%; Day-to-day ICC 0.98 (0.93–0.99), CV 0.7–7.5%. Comparison to prior work: Estimated biceps brachii shear wave velocities at 276 Hz (~3.02–5.12 m/s) were consistent with values inferred from Ballyns et al. using external vibration (~4 m/s), supporting validity of the method.

The study addressed the need for an accessible, reliable quantitative method to assess skeletal muscle stiffness by proposing a vibration-based shear wave elastography method with color Doppler wavefront imaging. By directly visualizing shear wave wavefronts, operators could optimize probe orientation and excitation placement to ensure uniform propagation along muscle fibers, enhancing measurement reliability. The method demonstrated excellent reproducibility in phantoms (very low CVs and high ICCs) and good-to-excellent intra-day and day-to-day reliability across major superficial muscles in healthy subjects. The strong correlation between agar concentration and measured shear wave velocity supported sensitivity of the approach to stiffness changes. Measured shear wave velocities and reliability metrics were comparable to or better than those reported with established ARFI-based systems, while relying on simpler, potentially lower-cost equipment. These findings suggest the method can effectively quantify muscle elasticity, enabling broader clinical and research applications where device cost and complexity are barriers.

Using vibration-based shear wave elastography with color Doppler wavefront imaging, the study confirmed adequate reproducibility and reliability for quantifying elasticity in phantoms and commonly assessed superficial skeletal muscles. The proposed method holds promise as a practical tool for evaluating muscle stiffness in clinical and research settings. Future work should validate performance across diverse populations (different ages, sexes, ethnicities) and in patients with various pathologies, and extend measurements to additional muscle regions and deeper tissues.

• No concurrent validation against other quantitative methods was performed within this study (though prior work by the authors reported consistency with ARFI-based measurements in phantoms and superficial muscles).
• Participants were asymptomatic young male volunteers, and measurements were performed in static (resting) muscles, limiting generalizability to broader patient populations and clinical conditions.
• Only the middle portion of each muscle was assessed; regional variations (e.g., muscle–tendon junctions) were not evaluated.
• Only major superficial muscles were measured; feasibility for deeper tissues may require adjustments (e.g., actuator frequency tuning to account for shear wave absorption) and further investigation.