Smartphone video-based estimates knee extension moments during chair rise relate to MRI measures of muscle function

This study demonstrates that smartphone video-based estimates of knee extension moments using OpenCap correlate significantly with MRI measures of quadriceps muscle volume and microstructure, offering a scalable and accessible alternative to costly imaging and dynamometry for assessing muscle function in clinical and large-scale settings.

The Gist

The Big Picture: Finding a "Smartphone" for Muscle Health

Imagine you want to know how strong a car engine is.

• The Gold Standard: You could take the engine out, put it on a massive, expensive test bench in a lab, and measure exactly how much horsepower it produces. This is accurate, but it's slow, costs a fortune, and you can't do it in a regular mechanic's shop.
• The Old Way: You could just time how fast the car accelerates from 0 to 60. It's easy to do, but it doesn't tell you why the car is slow. Maybe the engine is weak, or maybe the driver just didn't press the gas pedal hard enough, or maybe the driver is leaning back in the seat to make the car go faster.
• The New Idea: What if you could point your smartphone at the car, record a video, and use a smart app to calculate the engine's power just by watching how the car moves?

This paper is about doing exactly that for human muscles.

The researchers wanted to see if they could use a simple smartphone video to measure how strong our leg muscles (specifically the quadriceps) really are, and if that video data matches up with the "gold standard" medical tests.

---

The Three Tools in the Study

To test their idea, the researchers used three different ways to measure muscle strength in 19 healthy people (ranging from 30 to 78 years old):

1. The "MRI" (The X-Ray Vision):

   • What it is: A giant, expensive machine that takes super-detailed pictures of the muscles inside the leg.
   • What it measures: It looks at two things:
     • Size: How much muscle meat is there? (Like the size of the engine block).
     • Quality: How healthy is the muscle tissue? Are the fibers tight and strong, or loose and fatty? (Like the quality of the fuel and the spark plugs).
   • The Problem: It's too expensive and slow to use on everyone in a doctor's office.

2. The "Dynamometer" (The Lab Bench):

   • What it is: A heavy, specialized machine where you sit and push against a padded bar as hard as you can.
   • What it measures: The exact amount of force your muscle can generate.
   • The Problem: Like the MRI, it's bulky, expensive, and requires a trained technician.

3. The "Smartphone Video" (The New Hero):

   • What it is: Two people record you standing up from a chair five times as fast as you can, using just two regular smartphones.
   • What it measures: Using a free software called OpenCap, the app analyzes the video to calculate the force your knee is generating while you stand up.
   • The Goal: To see if this cheap, easy video method can tell us as much about muscle health as the expensive MRI and dynamometer.

---

The "Five Times Sit-to-Stand" Test (The Old Way)

Usually, doctors just time how long it takes you to stand up and sit down five times.

• The Flaw: The paper found that time is a tricky metric.
• The Analogy: Imagine two runners. Runner A has a weak engine but leans forward and swings their arms wildly to compensate. Runner B has a strong engine but stands up slowly and carefully. If you only look at the stopwatch, you might think they are the same, or you might miss the fact that Runner A is actually struggling with their engine.
• The Result: In this study, the time it took to stand up had no connection to the actual muscle size or quality seen on the MRI. It was too easy to "cheat" by changing your body position.

---

The Big Discovery

The researchers compared the data from all three methods. Here is what they found:

1. The Smartphone Video Worked: The force calculated by the smartphone video (the knee extension moment) matched up very well with the MRI pictures.

   • If the MRI showed big, healthy muscles, the smartphone video showed high force.
   • If the MRI showed smaller or lower-quality muscles, the video showed lower force.
   • The Analogy: The smartphone video was like a "smart detective" that could figure out the engine's power just by watching the car move, even without opening the hood.

2. The "Composite Score" was Key: The MRI measures both Size (Volume) and Quality (Microstructure).

   • The smartphone video and the heavy lab machine (dynamometer) were best at predicting a score that combined both size and quality.
   • This suggests that to truly understand muscle strength, you need to look at dynamic movement (moving fast), not just static holding (pushing against a wall).

3. The "Time" Test Failed: As mentioned, simply timing the test didn't tell the doctors anything about the muscle's health. It was too easy for people to use tricks (like leaning their torso forward) to make the task look easier than it was.

---

Why This Matters

The Problem: As we age, our muscles get weaker and smaller (a condition called sarcopenia). This leads to falls and loss of independence. But right now, we don't have a cheap, easy way to catch this early. The MRI is too expensive, and the stopwatch test isn't sensitive enough.

The Solution: This study suggests that OpenCap (the smartphone video tool) is a game-changer.

• It's Accessible: You can do it in a doctor's office, a gym, or even at home with two phones.
• It's Fast: It takes about 5 minutes.
• It's Smart: It doesn't just time you; it calculates the actual force your muscles are producing, catching early signs of weakness before you even start falling or slowing down.

The Bottom Line

Think of your muscles like a car engine.

• The Stopwatch just tells you how long the trip took, but it doesn't tell you if the engine is failing.
• The MRI is a full engine teardown, which is great but impractical for everyone.
• The Smartphone Video is like a new, super-smart diagnostic tool that listens to the engine and watches the car move, giving you a highly accurate report on the engine's health in just a few minutes.

This paper proves that this "smartphone diagnostic" is accurate enough to replace the expensive, hard-to-get tests for many people, potentially helping doctors catch muscle weakness much earlier and keep people moving for longer.

Technical Summary

1. Problem Statement

Preserving muscle function is critical for maintaining independence in aging populations, yet clinical assessment of muscle force-generating capacity is hindered by a lack of accessible, sensitive tools.

• Gold Standards: Magnetic Resonance Imaging (MRI) provides high-fidelity measures of muscle volume (quantity) and microstructure/quality (e.g., radial diffusivity), while dynamometry measures peak joint moments. However, both are costly, time-consuming, and require specialized equipment, making them impractical for routine clinical screening or large-scale studies.
• Current Clinical Practice: Clinicians rely on low-fidelity timed functional tests, such as the Five Times Sit-to-Stand (5xSTS) test. While practical, these tests measure only the time to completion. They fail to account for compensatory movement strategies (e.g., increased trunk lean) and are often insensitive to early declines in muscle function that occur before performance time is significantly affected.
• The Gap: There is a need for a scalable, accessible, and high-fidelity assessment tool that can detect early deficits in muscle quantity and quality without requiring MRI or dynamometry.

2. Methodology

The study evaluated whether OpenCap, an open-source platform using smartphone videos to quantify musculoskeletal dynamics, could serve as a proxy for MRI-derived muscle metrics.

• Participants: 19 healthy adults (ages 30–78; 63.2% female).
• Data Collection Protocol:
  1. MRI: Participants underwent 3T MRI scans to measure quadriceps muscle volume and radial diffusivity (a diffusion tensor imaging metric related to fiber size and type II fiber proportion). A composite MRI score was created by standardizing and summing volume and radial diffusivity to represent integrated muscle quantity and quality.
  2. Dynamometry: Isometric and isokinetic (concentric at 120°/s) knee extension moments were measured using a HUMAC NORM dynamometer.
  3. Functional Test: Participants performed a maximal-effort 5xSTS test.
     • Timing: A stopwatch recorded total completion time.
     • OpenCap: Two smartphones recorded the test. Markerless motion capture was used to generate a 3D musculoskeletal model (OpenSim) to estimate peak knee extension moments, trunk orientation, and angular velocity.
• Statistical Analysis: Linear regression and Pearson correlations were used to relate functional outcomes (OpenCap moments, dynamometry, 5xSTS time, kinematics) to MRI metrics. False discovery rates were controlled using the Benjamini-Hochberg procedure.

3. Key Contributions

• Validation of Smartphone Kinetics: The study demonstrates that joint moments derived from smartphone videos (OpenCap) correlate strongly with gold-standard MRI measures of muscle microstructure and volume.
• Superiority over Timed Tests: It provides evidence that kinetic measures (force/moments) are more sensitive to muscle quality than spatiotemporal measures (time) or kinematic compensations (trunk angle).
• Composite Muscle Health Metric: The study introduces and validates a composite score (Volume + Radial Diffusivity) that better reflects dynamic force-generating capacity than volume alone.
• Scalability: It proposes a 5-minute, low-cost clinical workflow that replaces or augments traditional timed tests with biomechanically rich data.

4. Key Results

• OpenCap vs. MRI:
  • OpenCap-derived peak knee extension moments showed significant correlations with quadriceps volume ($r=0.63, p=0.014$) and radial diffusivity ($r=0.61, p=0.016$).
  • The strongest association was with the composite MRI score ($r=0.77, p=0.002$).
• Dynamometry vs. MRI:
  • Isokinetic dynamometry correlated with volume ($r=0.66$) and the composite score ($r=0.73$), but not significantly with radial diffusivity after correction.
  • Isometric dynamometry correlated only with volume ($r=0.75$) and showed no association with radial diffusivity or the composite score.
• Timed Test & Kinematics vs. MRI:
  • 5xSTS Time: No significant correlation with volume, radial diffusivity, or the composite score ($p > 0.72$).
  • Kinematics: Trunk orientation and angular velocity did not correlate with any MRI metrics.
• Interpretation: Dynamic assessments (OpenCap and isokinetic dynamometry) capture both muscle quantity and quality, whereas static assessments (isometric dynamometry) and timed tests primarily reflect quantity or are confounded by compensation strategies.

5. Significance and Implications

• Clinical Utility: OpenCap offers a scalable alternative to MRI and dynamometry. It can be deployed in clinics or remote settings using standard smartphones and free software, enabling the assessment of muscle force-generating capacity in populations where imaging is impractical.
• Early Detection: Because the 5xSTS time failed to correlate with MRI metrics, the study suggests that timed tests may miss early-stage sarcopenia or muscle decline. OpenCap's ability to measure the moment (force) rather than just the time allows for the detection of deficits before they manifest as slower performance times.
• Muscle Quality: The correlation between OpenCap moments and radial diffusivity suggests that smartphone-based kinetics can indirectly assess muscle microstructure (fiber type and size), which is crucial for understanding dynamic power and fall risk.
• Future Directions: While validated in healthy adults, the authors note the need for validation in clinical populations (e.g., those with diagnosed sarcopenia or neuromuscular disorders) and longitudinal studies to track progression.

In conclusion, the paper establishes that smartphone video-based kinetic analysis is a sensitive, accessible, and high-fidelity method for assessing muscle function, outperforming traditional timed functional tests and offering a practical bridge between gold-standard imaging and routine clinical care.