A study combined ultrasound and EMG to track biceps fatigue. We explain the findings, key limits and relevance for endurance athletes over 35.
A study published in BMC Sports Science, Medicine and Rehabilitation on October 23, 2026, combined ultrasound and surface electromyography to examine localized biceps fatigue. The researchers compared advanced resistance trainees with novices during a dumbbell-curl test. Their question was whether the measurements could capture differences between groups and distinguish stages of fatigue.
The study report describes two related lines of investigation. One compared repetitions and physiological responses between participants with different training backgrounds. The other tested a system for classifying fatigue stages from ultrasound video and electrical recordings.
For endurance athletes over 35, the interesting question is how those laboratory findings connect to training decisions. That connection starts with separating what the researchers measured from what an athlete might hope to learn.
The researchers reported that advanced trainees completed an average of 16.3 repetitions. Novices averaged 13.3 repetitions under the same fixed-load protocol. The difference was statistically significant (p < 0.001).
The advanced group also showed a smaller increase in biceps thickness between baseline and exhaustion. The average increase was 12.69% in advanced trainees compared with 18.14% in novices. This difference was also statistically significant (p = 0.003).
Those findings describe distinct outcomes: how many curls participants completed and how muscle thickness changed during the task. Keeping those outcomes separate matters. A repetition count is a performance result for that exercise, while a thickness measurement describes a physical change observed during it.
The electrical recordings added another view. Surface electromyography, abbreviated sEMG, captured changes in muscle electrical activity. Its root mean square amplitude increased as fatigue developed. Median frequency followed an inverted-U pattern, rising and then falling rather than moving steadily in one direction.
The researchers also measured arterial peak systolic velocity, a blood-flow velocity measure. It increased with fatigue. However, the groups did not differ significantly in their values at exhaustion (p = 0.672).
Taken together, the results show why the authors were interested in combining measurements. Muscle thickness and blood-flow velocity describe different features from electrical activity. The researchers interpreted these signals as complementary associations during localized fatigue, rather than interchangeable versions of a single fatigue score.
The study included 40 healthy men divided into advanced resistance-training and novice groups. Each group contained 20 participants. Everyone performed right-arm curls with a fixed 10-kg dumbbell until voluntary exhaustion.
Measurements were collected at baseline, moderate fatigue and exhaustion. The researchers tracked biceps thickness and arterial peak systolic velocity alongside sEMG amplitude and median frequency. This allowed them to compare several features of the same exercise task as fatigue progressed.
The classification analysis used ultrasound video and sEMG recordings from 35 participants whose recordings were free of artifacts. Its task was to distinguish the three fatigue stages. The authors reported 86.23% accuracy using participant-level, five-fold internal cross-validation.
That accuracy figure belongs to the classification analysis. It is separate from the comparison showing that advanced trainees completed more repetitions. One result concerns identifying stages within recorded data. The other concerns differences between the training-background groups.
The fixed load is the central limitation of the group comparison. Although everyone lifted the same 10-kg dumbbell, that weight represented different relative intensities for different participants. The authors therefore caution against treating the group differences as effects of training experience alone. The experiment did not place everyone under an equivalent challenge relative to individual capacity.
The participant sample and exercise also set boundaries. This was a localized biceps-curl experiment in healthy men. It was not an endurance-performance study or a trial of whole-body training in older athletes. Nor did it test whether adding progressive resistance training improves endurance performance.
The thickness change was measured during the exercise bout. It should not be interpreted as long-term muscle growth or evidence of better recovery afterward. Those are different outcomes from the acute measurements reported here.
Finally, the authors describe the classifier results as exploratory. Internal cross-validation is not external validation in a new population or training setting. They identify individualized loading and larger, more diverse samples as necessary next steps alongside external validation. The paper does not establish a validated field tool or a routine monitoring protocol for athletes and coaches.
These boundaries define the study's contribution: an early assessment of combined fatigue measurements under a specific experimental protocol.
For a veteran runner or cyclist, there are two separate questions here: whether strength training supports performance and whether a measurement system can guide training decisions. Those questions require different evidence. A useful result for one does not automatically answer the other.
Reendure's focus is practical training and recovery guidance for athletes over 35. The useful translation in this case is to match each decision to the evidence that actually addresses it. A method for observing fatigue and a program for improving performance should not be treated as the same intervention.
A separate 2026 systematic review and meta-analysis of 17 controlled trials examined heavy strength training in cyclists. It reported small-to-moderate improvements in time-trial or time-to-exhaustion performance. Changes in VO₂max and several other aerobic markers did not reach statistical significance.
That review is the more relevant evidence when discussing strength training for cycling performance. It assessed performance outcomes in cyclists rather than changes during a biceps-curl task. Its findings also show why the outcome matters: improvements in certain performance tests did not mean every measured aerobic marker changed significantly.
A separate 2026 review focused on resistance training for cyclists identified masters as one of the populations needing further research. The age-specific application therefore deserves its own attention. General cycling findings should not be presented as an established ideal program for every older athlete.
For endurance readers, the practical distinction is straightforward. Use endurance-specific research to assess the case for strength work. Use measurement research to understand what a proposed monitoring method is designed to assess. Neither needs to stand in for the other.
This paper does not provide a reason to copy the curl-to-exhaustion test into an endurance program. The task was a way to study localized fatigue. Its role in the experiment was to generate measurements, not to serve as a workout recommendation.
Likewise, the reported classifier accuracy is not a personal target. It does not tell an athlete how fatigued they should become or how much training to complete. Those decisions require evidence and context beyond the classification result.
A practical reading of the research keeps three questions separate:
That separation helps prevent an impressive measurement result from becoming an unsupported training rule. It also keeps attention on the outcome that matters for the decision, whether that is cycling performance or the usefulness of a proposed assessment.
For related measurement research, see our coverage of ultrasound assessment of calf stiffness. For the separate question of session planning, our guide to adjusting workouts around stress and fatigue addresses the training decision itself.
Where decisions involve pain or personal health concerns, qualified clinical care is the appropriate route. A research-stage fatigue score should not be used to interpret symptoms or make treatment decisions.
The study adds a detailed view of localized biceps fatigue. Advanced trainees completed more curls and showed a smaller average increase in muscle thickness under the researchers' protocol. Combining ultrasound with sEMG also produced encouraging fatigue-stage classification results within the study dataset.
For endurance athletes over 35, the immediate value is understanding the measurement approach rather than changing training around it. The broader case for strength training should rest on endurance-specific evidence, with attention to the population and outcomes studied.
The modest takeaway is useful enough: fatigue has several measurable features, and combining signals may help researchers describe them. This study offers an early assessment of that approach, not a new rule for how hard to train or when to stop.
Follow ReEndure for practical insights on endurance training, recovery, nutrition and healthy aging. Stay connected for new articles, research led guidance and ideas to help you perform better for longer.
Read practical ideas on endurance training, recovery, nutrition and healthy aging to keep progressing for years to come.
Read the Blog