Towards Enhanced Bladder Volume Measurement with Context-Aware Bio-Impedance Sensor Fusion

Urine volume is a vital parameter of health and its continuous monitoring is necessary for managing illness in various patient populations. State-of-the-art methods include invasive catheters or sporadic measures with ultrasound. Non-invasive urine volume measurement in vivo is challenging due to the confounding effects of muscle activity. This work proposes a multi-sensor wearable system based on non-invasive tetrapolar bio-impedance (BI) measurements along with the measurement of muscle activity by including electromyography (EMG) and novel low power charge variation (QVAR) sensing integrated into inertial measurements unit, placed on the lower abdomen. A comparison of the EMG and QVAR is conducted and revealed that the QVAR sensor had a correlation coefficient of 0.93 with the EMG and consumed a sixteenth of the power of the EMG, making it more suitable for long-term, multisensor and continuous monitoring. Also, electrode placements on the lower abdomen were investigated to understand the effect of different muscle activations during bladder voiding experiments. Measurements of muscle clenching, coughing and localised movements were performed at the forearm and the lower abdomen with all three sensors. Investigations on bladder voiding experiments showed that an unexpected BI decrease at the beginning of the bladder voiding coincided with an increase in the electromyography signal over the abdominis rectus muscles and could be the cause of the unexpected BI change at the start. Therefore, a combination of low-power sensors of BI and QVAR can pave the path towards a robust sensing method to measure bladder volume changes.