Movement algorithms for back pain have gaps in measurement evidence
A systematic review found that accuracy dominated reporting, while repeatability and measurement error received much less attention.
Based on the published abstract. The full paper may contain additional methods, results and limitations.
The 30-second takeaway
The review examined whether machine-learning methods can assess movement in people with low back pain. The authors described validity as strong, but few studies examined reliability or measurement error. This abstract therefore offers an encouraging assessment of accuracy alongside important gaps in how consistently the tools measure movement and how much error those measurements contain.
The abstract gives no accuracy estimates, participant totals, reference standards, or detailed bias results. It is not possible to judge a particular tool's performance or suitability for clinical use from this summary.
What the tools used
In this abstract-based review, studies used movement data as inputs to machine-learning systems. Inertial sensors were the most common data source, and support vector machines were the most commonly used type of algorithm.
What accuracy leaves unresolved
Accuracy addresses how well an assessment matches its reference. Reliability concerns consistency, while measurement error concerns uncertainty in the measurement. The review's limited reporting on these latter properties constrained the authors' conclusions about clinical use.
The original publication
Machine learning accuracy for assessment of functional movement in Low back pain based on clinically applicable performance Metrics: A systematic review.
Burjawi T, El-Ansary D, Farragher J et al.
Int J Med Inform · 2026
- PubMed ID
- 41138618
- Record checked
AI-assisted research and writing. This explains one selected publication; it is not a complete review of everything known. Our approach.
One more question, understood.
Keep track of the research you’ve explored.