A prototype smartwatch for monitoring dynamic, compound and plyometric exercises in cancer prehabilitation: a...

Cancer prehabilitation programmes increasingly rely on home-based exercise interventions, yet objective monitoring of exercise compliance remains challenging. Current commercial wearables focus primarily on aerobic activities and lack capability for monitoring dynamic, compound, and plyometric exercises essential for cancer prehabilitation.

To develop and validate a prototype smartwatch capable of monitoring specific prehabilitation exercises with objective compliance tracking.

We developed a prototype using the M5StickC PLUS2 development board with integrated 6-axis inertial measurement unit. Seven standardised cancer prehabilitation exercises were monitored using a novel Rep Error Height (REH) scoring algorithm. The study comprised two phases: algorithm development with 6 healthcare workers providing reference waveforms, and validation with 20 healthy volunteers performing exercises under observation and at home for 4 days. Usability was assessed using the adapted mHealth App Usability Questionnaire (MAUQ).

Algorithm performance demonstrated high correlation (88.9%-98.8%) between visual and algorithmic repetition counting across all exercises. Median compliance rates exceeded 90% overall, though specific exercises showed lower compliance on certain days. Usability assessment revealed mean overall scores exceeding 5.0, with ease of use rated most favourably. However, only 35.7% of participants rated device usefulness above 5.0, primarily due to lack of real-time feedback mechanisms.

The prototype has the potential to address the gaps in current wearable technology for cancer prehabilitation monitoring. Whilst demonstrating high accuracy for exercise detection, algorithm refinement is needed to accommodate diverse movement patterns and variations in exercise execution.