Sports medicine project · Dubai · 2025 to now
Find the weak spot before it finds you.
FlexTrack shows athletes and coaches where on the body injury risk is building, using the signals sports medicine already trusts: past injuries, training load, sleep and soreness. The check-in prototype works today. The smart patch is next.
Started as a response to the Barça Innovation Hub brief in the Hult Future Leaders Challenge 2025.
The problem
Injuries rarely come from nowhere
Most muscle injuries are non-contact. They build up out of things a team can see coming: a player returning from an old strain, a week where training suddenly doubles, a run of short nights, a calf that has been tight for days.
Elite clubs pay sports scientists to watch for these patterns. School, academy and club athletes usually have no one doing it. The warning signs are there, but they are scattered across a coach's memory, a physio's notes and an athlete's own body.
FlexTrack puts those signals in one place and maps them onto the body, so the question changes from "is this athlete tired?" to "which part of this athlete needs attention today?"
Prototype v1 · Check-in
See it work
This is the FlexTrack check-in running live. It opens on a Sample athlete a basketball player in a heavy week who strained a hamstring last month. Change anything and the body map updates. Tap a body part to see why it is flagged and what to do.
Effort uses the 0 to 10 session-RPE scale. Load = minutes × effort, in arbitrary units (AU). "Usual" means your average week over the last four.
Selected
All regions, highest first
How to read the number: it is a rough multiple of risk compared with a rested athlete with no injury history, built from published findings and a few labelled design assumptions (see Evidence). Multiplying several risk factors overstates the combined effect, so anything above 10 is shown as 10+. It is a screening and awareness tool, not a diagnosis. Sharp pain, swelling, a "pop", numbness or pain at night means stop and see a doctor. This page stores nothing; everything runs in your browser.
How the score works
Four signals, one body map
Each body region starts at 1.0. FlexTrack multiplies in the factors that research links to injury, then turns the result into a plain instruction. Symptoms decide whether to see someone; numbers alone never do.
| Signal | What FlexTrack does | What the research says | Confidence |
|---|---|---|---|
| Injury history ×2.7 before · ×4.8 recent | Raises risk in the exact region that was injured. | A 2020 meta-analysis found any previous hamstring strain carried a risk ratio of 2.7, and a recent one 4.8. FlexTrack borrows these weights for every region until pilot data allows region-specific ones. [3] | Strong |
| Sleep ×1.7 under 8 h | Applies a body-wide multiplier when average sleep drops below 8 hours. | Adolescent athletes sleeping under 8 hours were 1.7 times as likely to have been injured. [4] | Moderate |
| Training load spike ×1.4 above 1.3 | Compares this week's load (minutes × effort) with the usual week and flags jumps above 30%. | Session-RPE is a validated way to measure load across sports and ages. [5] Whether the acute-to-chronic ratio predicts injury is disputed, so FlexTrack shows it as a coaching flag, not a verdict. [6] | Contested |
| Soreness ×1.12 per point | Uses the athlete's 0 to 10 rating for each region. A 7 or higher always triggers "get it checked". | Design assumption. Soreness is a symptom, so it drives the referral rule more than the score. | Assumption |
| Sport hot spots ×1.15 to ×1.3 | Gives a small boost to regions that commonly get injured in each sport (ankles in basketball, shoulders in volleyball, lower back in gymnastics). | Design assumption, to be replaced by pilot data. | Assumption |
| Machine learning Phase 3 | Not used yet. The current score is transparent on purpose. | Published injury-prediction models range from barely better than chance (AUC 0.52) to strong (AUC 0.87), and small datasets are the main weakness. [7] FlexTrack will train a model only once pilot data exists, and will publish how well it does. | Planned |
Phase 4 · Concept
The FlexTrack patch
The check-in relies on what athletes report. The patch adds what their muscles report. It sticks over the region the map flags (a hamstring, a calf) and measures local fatigue during the session.
Surface EMG: muscle fatigue
Proven in wearablesAs a muscle tires, the frequency of its electrical signal shifts lower. Research groups have already built EMG patches that track this in real time during exercise. [8]
Motion sensor (IMU): load and asymmetry
Proven in wearablesCounts jumps, sprints and landings, and spots when one leg starts doing more work than the other.
Sweat rate and sodium
On the marketSingle-use sweat patches already measure these for hydration. [9] FlexTrack would use them as context, not as an injury signal.
Skin temperature
Context onlyHelps interpret the other signals in heat, which matters for athletes training in the Gulf.
Sweat lactate
Research stageLeft out for now. Sweat lactate does not yet track blood lactate reliably, and reviews call for more validation. [10]
Where FlexTrack fits
Others track the whole body. FlexTrack tracks the part.
Good tools already exist. Each answers a different question.
| Tool | Measures | Answers | Gap FlexTrack fills |
|---|---|---|---|
| WHOOP, Oura | Heart rate variability, sleep, overall strain (wrist or finger) | Is my body recovered today? | Not which muscle or joint is at risk |
| Catapult and GPS vests | Distance, speed, accelerations | How much work did the team do? | Built for elite budgets; no injury history or symptoms |
| Gatorade Gx patch | Sweat rate and sodium loss | How much should I drink? | Hydration only |
| FlexTrack | Injury history, load, sleep, soreness by region; patch adds local muscle fatigue | Which part of me needs attention, and who should I see? | Region-level risk for school and club athletes |
From pitch to prototype
What changed when I checked the science
The first version of FlexTrack was a competition pitch. Turning it into something real meant testing every claim against the research and changing what did not hold up.
| In the 2025 pitch | Now | Why |
|---|---|---|
| "Nano-sensors" measure micro-strain, hydration, lactate and blood flow | Surface EMG, motion, temperature and sweat sensors | These are proven in wearables today. Sweat lactate is still unreliable. [8][10] |
| "AI-powered" from day one | Transparent score now; machine learning after the pilot | A model is only as good as its data, and small datasets are the main weakness in published injury models. [7] |
| Patch first | Check-in first, patch later | The strongest predictor (past injury) needs no hardware. Athletes can use the check-in this season. |
| Headline injury figures without sources | Every number on this page is cited | Injury cost and incidence figures now come from UEFA's injury study and Howden's injury index. [1][2] |
| Built around one elite club | Built for school, academy and club athletes | Elite clubs already have sports scientists. Younger athletes usually have no one watching these signals. |
Roadmap
Next steps
1 · Check-in prototype
October 2026Body map and transparent risk score, running on this page.
2 · Team pilot
November 2026 to February 2027Six-week pilots with school and club teams in Dubai across basketball, volleyball, gymnastics and dance.
3 · Learning model
Spring 2027Replace assumptions with weights learned from pilot data, and publish accuracy honestly.
4 · Patch prototype
Summer 2027A working EMG and motion patch built from off-the-shelf sensor boards, tested on the hamstring and calf.
Tested with athletes
What testers say
FlexTrack's check-in has been trialled with student athletes in India and Egypt.
Mumbai, India
Our students used FlexTrack across their training sessions and found it quick, simple and genuinely useful. It turns scattered signals into a clear picture of where an athlete needs care. I recommend it highly.
Cairo, Egypt
Our athletes checked in after every session, and the body map flagged the same areas our physiotherapist was already watching. That gave our coaching staff real confidence in it.
Pilot programme
Bring FlexTrack to your team
FlexTrack is looking for school, academy and club teams to run a free six-week pilot.
Your team gets
- Free access to the FlexTrack check-in for every athlete
- A weekly body-map summary for the coach
- An end-of-pilot report on load, sleep and flagged regions
We ask for
- A one-minute check-in from each athlete after training
- The coach or physio to log any injury that causes missed sessions
Our privacy promise
- No athlete names in the dataset; each athlete gets a code
- Athletes under 18 join only with a parent's consent and the coach's agreement
- Anyone can withdraw at any time, and their data is deleted
Impact log
Founder
Laila Zaghloul
IB Diploma student in Dubai, planning to study biology on a pre-med track.
I have danced for twelve years and done gymnastics for ten, and I play basketball and volleyball. I want to become a doctor because I like helping people. FlexTrack is where those two parts of my life meet: an athlete's body gives warning signs, and medicine is about learning to read them early.
Sources
- Ekstrand J et al. Hamstring injury rates have increased during recent seasons: the UEFA Elite Club Injury Study 2001/02 to 2021/22. BJSM, 2022. Summary
- Howden. Men's European Football Injury Index 2023/24 and 2024/25. 2023/24 · 2024/25
- Green B et al. Recalibrating the risk of hamstring strain injury: a 2020 systematic review and meta-analysis. BJSM 54(18), 2020. PubMed
- Milewski MD et al. Chronic lack of sleep is associated with increased sports injuries in adolescent athletes. J Pediatr Orthop, 2014. Summary
- Haddad M et al. Session-RPE method for training load monitoring: validity, ecological usefulness, and influencing factors. Front Neurosci, 2017. Article
- Impellizzeri FM et al. Training load and its role in injury prevention: methodological issues. PMC
- Van Eetvelde H et al. Machine learning methods in sport injury prediction and prevention: a systematic review. J Exp Orthop, 2021. PMC; Leckey C et al. Machine learning approaches to injury risk prediction in sport: a scoping review. BJSM, 2025. Article
- Liu SH et al. An EMG patch for the real-time monitoring of muscle-fatigue conditions during exercise. Sensors, 2019. PMC
- IEEE Spectrum. How your smartphone can see you sweat (Gx Sweat Patch). Article
- Yang G et al. Wearable device for continuous sweat lactate monitoring in sports: a narrative review. Front Physiol, 2024. PMC
- Hult International Business School. Hult Future Leaders Challenge 2025, with Barça Innovation Hub. hult.edu