Progress Photos vs Measurements: What Tracks Change
A 2023 study in The American Journal of Clinical Nutrition compared photo-based 3D optical body scans to DXA, the imaging method researchers use as a reference. The question mattered: if you're tracking your own body over weeks, should you trust a photo-based scan, a tape measure, or a scale? The answer, it turns out, depends less on which tool you use and more on whether you're looking at a trend or a single number.
What the researchers actually compared
The study pooled data from 133 people across six separate intervention studies, with follow-up periods averaging 13 weeks (ranging from 3 to 23 weeks). Each person had both a 3D optical scan and a DXA scan at baseline and again at follow-up, so researchers could check whether the two methods agreed on how much change had actually happened.
Agreement was strong for the big-picture numbers. In women, the optical scan tracked DXA's fat mass change with an R2 of 0.86, fat-free mass at 0.73, and appendicular lean mass at 0.70. In men, those numbers were 0.75, 0.75, and 0.52. An R2 near 1.0 means the two methods moved together almost perfectly; these are solidly in the 'this is a real signal' range, not noise.
Why tracking a trend beats trusting one photo
Here's the detail that matters most for anyone checking their own progress: the error was smaller when comparing change over time than when comparing a single scan to a single DXA reading. The researchers noted that RMSEs for one-off baseline or follow-up comparisons were higher, meaning worse, than RMSEs for the change comparisons themselves.
In plain terms, a single scan can be off in ways that are specific to that day: hydration, lighting, posture, how you stood. But when you compare two scans taken weeks apart, those individual quirks tend to cancel out, and what's left is closer to the real underlying change. This is the core argument for tracking over time rather than reacting to any one photo or measurement.
Where the agreement gets shakier
Not every number held up equally well. Regional measurements and small changes were noisier than total-body ones, and visceral fat change agreement was notably weaker, with an R2 around 0.23 to 0.25. That's a reminder that broad totals, like overall fat mass or lean mass, are more trustworthy to watch than a single localized number that hasn't moved much.
It's also worth naming the limits of the study itself. This was a retrospective pooling of six different intervention studies with varying protocols, comparing one specific optical scanner to DXA, which is itself an estimate rather than a perfect ground truth. The researchers also cited a separate comparison (Tinsley et al.) where a different photo-based scanning approach performed considerably worse against a gold-standard method, with concordance as low as 0.22 to 0.40 for fat mass change. The lesson: results depend heavily on the specific device and algorithm, so this shouldn't be read as a blanket endorsement of every photo-based tool on the market.
What this means for your own scans
This is the logic behind how Trainr presents your numbers. A single scan gives you a snapshot: waist, chest, arms, an estimated body-fat percentage. That snapshot is useful, but the research above suggests it's the second, third, and fourth scan that tell you something more reliable, because comparing change over weeks smooths out the day-to-day noise that any single measurement carries.
In practice, that means watching your waist or body-fat trend line move over a month matters more than obsessing over whether this week's number is a fraction off. Your first scan with Trainr is free, so there's no cost to starting the trend line. Everything after that is about watching direction, not chasing a single perfect reading. As always, body-composition estimates from any consumer scan, including Trainr's, are estimates, not diagnostic measurements.
Scan yourself — the first one is freeOne photo tells you where you stand; a string of them, taken the same way over weeks, tells you where you're actually going.
Trainr is a wellness tool, not a medical device. Measurements and body-composition figures are estimates for self-tracking; for medical questions, talk to a professional.