Is Ultrasound a Reliable Way to Measure Muscle Growth?
Ultrasound muscle thickness is not a reliable indicator of resistance training-induced muscle growth. A study published in Medicine & Science in Sports & Exercise found that ultrasound measurements not only underestimated MRI-measured muscle growth by more than half, but showed no meaningful correlation with actual changes in muscle size — meaning the two methods were essentially tracking different things entirely.
For lifters and researchers who have been using ultrasound as a practical stand-in for MRI, this finding is significant. It raises a straightforward question: if common measurement tools can't accurately capture what's happening inside your muscle, how should you actually track resistance training progress?
Key Finding
After 15 weeks of structured lower-body resistance training, changes in ultrasound-measured muscle thickness (ΔMTUS) significantly underestimated changes in MRI-measured cross-sectional area and muscle volume for the quadriceps as a whole, and for two of its four constituent muscles specifically. More critically, there were no statistically significant correlations between ultrasound and MRI measurements — for the overall quadriceps or any individual muscle. The two methods were not just imprecise relative to each other; they were unrelated.
Study Details
What was studied: The researchers compared ultrasound muscle thickness measurements against MRI-based measurements of muscle cross-sectional area and volume in the quadriceps femoris, following a standardised resistance training program.
Population: 39 healthy young males completed 15 weeks of lower-body resistance training, performing knee extension, knee flexion, and leg press three sessions per week.
Methodology: Before and after the training period, participants underwent musculoskeletal scans of the dominant thigh using both ultrasound and MRI. Images were manually segmented to quantify changes in muscle thickness (ultrasound) and changes in maximum anatomical cross-sectional area and volume (MRI) for the quadriceps as a whole and each of its four constituent muscles: vastus lateralis, vastus medialis, vastus intermedius, and rectus femoris.
Results
The numbers here are worth sitting with:
- For the overall quadriceps, vastus lateralis, and vastus medialis, ultrasound underestimated muscle growth by more than half compared to MRI criterion measures (p values between 0.001 and 0.012).
- The average individual-level error from ultrasound was typically larger than the actual change detected by MRI — meaning the noise in the ultrasound measurement exceeded the signal it was trying to capture.
- For vastus intermedius and rectus femoris, ultrasound did not significantly differ from MRI measures, suggesting the method may perform better for some muscles than others.
- Critically, no significant correlations were found between ultrasound and MRI measurements for any muscle, in either absolute or relative terms (r values ranging from -0.197 to 0.422, all non-significant after adjustment).
The absence of correlation is the most important result. A poor correlation would mean ultrasound is imprecise. No correlation means it is measuring something different — or measuring the same thing so poorly that the information is not usable at the individual level.
Limitations
Before drawing sweeping conclusions, a few honest caveats:
Male-only sample. The study included only healthy young males, so it is not yet clear whether these findings generalise to women or older populations, who may respond differently to resistance training and have different tissue compositions.
Single measurement site per muscle. Ultrasound thickness was measured at specific anatomical landmarks. Muscle hypertrophy is not uniform along a muscle's length, and different measurement sites might yield different results — though this is also an argument against ultrasound's practicality rather than in its favour.
Trained vs. untrained populations. All participants were described as healthy young males completing a structured program. Whether the same disconnect between ultrasound and MRI holds in more experienced lifters, or in populations with greater baseline muscle mass, is not directly addressed by this data.
What This Means for Your Training
This study is not a reason to panic about how you have been assessing progress. Most lifters are not walking into ultrasound clinics between training blocks. But the findings do carry practical weight, especially for anyone who has been treating ultrasound-based body composition assessments — whether from a sports science lab, a physio, or a fitness-adjacent clinic — as a precise readout of muscle growth.
The core takeaway: Ultrasound muscle thickness, as currently used in research and applied settings, should not be treated as an accurate proxy for actual muscle hypertrophy. The measurement error at the individual level is too large to be actionable.
So what does hold up as a reliable way to track resistance training progress?
- Performance metrics over time. Load, reps, and volume per session are direct outputs of your training. If your squat is progressing across a 12-week program, your quadriceps are almost certainly adapting — you do not need an imaging device to confirm it. Tracking your training consistently, session by session, gives you a longitudinal record that no single-point measurement can match.
- Structural MRI remains the gold standard for measuring muscle hypertrophy in research settings, but it is expensive, inaccessible for most, and not something you need for day-to-day programming decisions.
- DEXA scans offer a more accessible alternative for body composition tracking, though they measure lean mass regionally rather than individual muscle volume directly.
- Circumference measurements are low-tech and imprecise, but they are cheap and consistent — and consistency of method matters more than method precision over long timeframes.
The deeper point this study reinforces is one that applies to all progress tracking: the tool you use matters less than whether you are using it consistently and interpreting it honestly. A training log that captures every session — loads, sets, reps, perceived effort — gives you far more usable signal than a single ultrasound scan taken at two time points.
Kenso is built around exactly this principle. Rather than chasing a single measurement of progress, the app tracks your training history session by session, applying a rule-based progression engine to surface when and how to advance your program. The Kenso AI Coach, powered by Claude and with full access to your training history, can help you interpret trends across weeks and months — the kind of longitudinal picture that a one-off ultrasound scan simply cannot provide.
For researchers and coaches, this study is a pointed reminder to scrutinise the validity of measurement tools before building conclusions on them. For individual lifters, it is a useful reminder that the most reliable measure of muscle adaptation is probably already in your training log — if you are keeping one.
FAQ
Is ultrasound accurate for measuring muscle growth from resistance training?
Based on this study, no. Ultrasound muscle thickness measurements significantly underestimated MRI-measured muscle growth and showed no meaningful correlation with actual changes in muscle size or volume at the individual level. The researchers recommend exercising substantial caution when using or interpreting ultrasound in the context of resistance training adaptation.
What is the gold standard for measuring muscle hypertrophy?
MRI — specifically changes in maximum anatomical cross-sectional area and muscle volume — is considered the criterion measure for quantifying resistance training-induced muscle growth. It is expensive and not widely accessible outside research settings, but it provides accurate, three-dimensional data that ultrasound cannot replicate.
Why did ultrasound underestimate muscle growth so significantly?
The study does not fully explain the mechanism, but muscle thickness measured at a single point along a muscle does not capture the full three-dimensional change in muscle volume. Hypertrophy is not uniform across a muscle's length, and ultrasound measures a narrow cross-section rather than total volume — which likely explains at least part of the discrepancy.
What should lifters use to track resistance training progress instead?
For practical purposes, consistently logged performance data — load, sets, and reps over time — is the most accessible and actionable measure of training progress. Apps like Kenso that track session-by-session performance give you a longitudinal record that reflects real adaptation far more reliably than a single measurement tool used at two time points.
Do these findings apply to all muscles equally?
Not entirely. In this study, ultrasound performed better for the vastus intermedius and rectus femoris than for the vastus lateralis and vastus medialis. However, the absence of significant correlations across all muscles suggests the limitations are broad enough that ultrasound should not be selectively trusted for specific muscles without further validation.
Citation
Balshaw TG, Funnell MP, McDermott EJ, Maden-Wilkinson TM, James LJ, et al. Ultrasound Muscle Thickness Is a Poor Index of Criterion Magnetic Resonance Imaging Measures of Resistance Training-Induced Muscle Growth. Medicine & Science in Sports & Exercise. 2025. DOI: 10.1249/MSS.0000000000004078
Track what actually moves — your training. Download Kenso on iOS and build a session-by-session record of your progression at kensoforge.com.