The demo returns a max-principal-strain estimate for each of seventeen brain regions — mostly white-matter tracts, plus the brainstem and a whole-brain summary. That table can look intimidating at first glance — a column of acronyms and numbers, all in the same units. Here’s how to read it without becoming a neuroanatomist.
Start with the colour, not the number
The dot beside every region is on the same colour scale as the 3D brain. Green is low strain, yellow is moderate, orange-to-red is elevated. If the table is mostly green, the impact was within everyday exposure ranges. A red dot anywhere is a prompt to look closer — not a diagnosis, but a flag.
What the regions are doing
The regions trace* tracks aren’t anatomical fluff. Each one is a bundle of long-distance connection fibres that carry signals between distant parts of the brain. They were chosen because:
- They concentrate in axonal injury more than the brain’s grey matter does, so they tell you about diffuse injury risk.
- They are well-defined in white-matter atlases, so the predictions are reproducible across studies.
- Several of them — notably the corticospinal tract (CST) and the superior longitudinal fasciculus (SLF) — recur as injury hotspots in the published rugby and American-football literature.
If your highest strain consistently shows up in CST and SLF, you’re seeing what most of the existing cohort studies see. That’s not reassuring on its own, but it does mean the model is responding to your impact the same way it responds to the dataset behind it.
A single high number is not a story; a pattern is
A useful instinct: look for repetition rather than for any single outlier.
- A single moderate-strain impact in a match is unremarkable for elite contact sport.
- The same region being moderately strained across many impacts in a session, week, or season is the interesting pattern. That’s what the longer-term exposure summaries are for.
The forthcoming weekly / monthly summary view of trace* will surface this directly: the same impact data, aggregated over time, broken down by region.
What strain does not tell you
- It does not tell you that injury occurred. Strain is a mechanical loading estimate. Whether it produced microscopic damage in a given athlete depends on tissue tolerance, hydration, prior history, age, and a long list of other factors.
- It does not replace symptom screening. If a player reports headache, balance loss, or memory disruption after an impact — regardless of what the numbers say — the clinical pathway should run.
- It does not predict CTE risk. No tool currently can. Strain helps you understand mechanical exposure; CTE risk is the cumulative product of exposure, biology, and time, and the science is still actively being mapped.
The right way to use the numbers is the same as the right way to use any monitoring tool: notice patterns, ask better questions, talk to the medic.
Sources & further reading
- Chan, E. Y. K., Yu, X., Qin, C. & Ghajari, M. (2025) — Balancing efficiency and accuracy. Engineering Applications of Artificial Intelligence 162, 112489. — Source of the per-region MPS90 predictions and their test-set accuracy.
- Donat, C. K. et al. (2021) — From biomechanics to pathology: predicting axonal injury from patterns of strain after traumatic brain injury. Brain 144 (1). — Why white-matter tracts, not grey matter, are the regions to watch.
- Hajiaghamemar, M. et al. (2021) — Multi-scale white matter tract embedded brain finite element model. Biomechanics and Modeling in Mechanobiology 20.