This original study evaluates whether a non-invasive wearable surface electromyography (sEMG) configuration can capture pelvic floor muscle (PFM) activity more successfully from the outer perineum compared to the inner perineum, and whether body position influences signal acquisition success.
PFM training requires repeated daily engagement for clinical effectiveness; however, existing EMG-based biofeedback systems [1-2] often rely on intravaginal probes or adhesive/gel-based electrodes, which are difficult to use multiple times a day. Wearable systems using conductive fabric electrodes offer a more practical approach for repeated use within daily routines.

This preliminary supervised clinical feasibility study used a within-subject repeated-measures design, where each participant was evaluated across all electrode locations and positions. Data from 14 women (aged 32.5+-12 years) were included after exclusion of three datasets with acquisition issues. The cohort included both healthy volunteers and individuals with symptoms of pelvic floor weakness, identified using a standard questionnaire. Each participant contributed recordings from two electrode locations (outer perineum and inner perineum) across three positions (sitting, standing, supine), resulting in 84 recordings.
A wearable shorts-based system incorporating dry conductive fabric electrodes (3cm diameter) was used without gels. Electrodes were integrated into a stretchable garment, available in multiple sizes to ensure consistent anatomical placement and good skin contact. Two channels recorded from the outer perineum (posterior peri-anal region) and inner perineum (anterior perineal region below the labia majora) [3]. While placement was fixed within the garment, effective inter-electrode distance varied with position, reflecting realistic wearable conditions.
Signals were acquired using a custom-built dual-channel EMG system, evaluated against a BIOPAC MP36 system at the beginning of the study to confirm signal acquisition reliability.
Participants were supervised by a trained physiotherapist, with correct PFM contraction verified using external perineal palpation. The protocol included three 5-second sustained contractions (~10s, 30s, 50s), rest intervals, followed by three rapid contractions. Positions were tested in rotated order to minimize sequence bias, and participants were positionally stabilized before recording.
Signals were acquired at 2000 Hz and processed using: 4th-order Butterworth band-pass filter (20-300 Hz), 50 Hz notch filter, median filtering and detrending. Segmentation was performed based on contraction windows defined by timing instructions. 
Considering variability in sEMG amplitude, analysis focused on signal usability classification. A recording was classified as successful if contraction segments demonstrated clear, interpretable activation with positive SNR and correspondence with instructed contraction timing, without significant flat, saturated, or noise-driven patterns.

The wearable conductive-fabric electrode configuration demonstrated feasibility for capturing interpretable PFM activity under supervised conditions. Recordings from the outer perineum were ~1.8 times more likely to meet predefined success criteria compared to the inner perineum. Sitting provided the most stable acquisition conditions, with recordings being ~2.2 times more likely to meet predefined success criteria compared to standing, and ~2.8 times compared to supine.

Lower signal usability at the inner perineum may be influenced by anatomical factors such as increased skin folding and relative motion, affecting electrode-skin contact stability. Positional differences suggest that mechanical factors inherent to wearable systems influence signal acquisition, with sitting providing more stable contact and standing introducing greater movement-related disturbances.
Importantly, the study demonstrates that conductive fabric-based wearable electrodes can capture interpretable pelvic floor activation signals in a supervised setting, supporting their potential for repeated-use biofeedback applications consistent with prior use of surface EMG in pelvic floor training contexts [2]. These findings provide practical insight into site selection and positional considerations for wearable pelvic floor EMG.

These findings support the outer perineum as a promising site for further development of non-invasive wearable pelvic floor biofeedback systems, and highlight the importance of body position in optimizing signal acquisition using such systems.

Wennergren, H., Larsson, L.E. and Sandstedt, P., 1989. Surface electromyography of pelvic floor muscles in healthy children: a methodological study. Scandinavian Journal of Caring Sciences, 3(2), pp.63-69.
Hill, A. and Alappattu, M., 2017. Quality-of-life outcomes following surface electromyography biofeedback as an adjunct to pelvic floor muscle training for urinary incontinence: A case report. The Journal of Women's & Pelvic Health Physical Therapy, 41(2), pp.73-82.
Frawley, H., Shelly, B., Morin, M., Bernard, S., Bø, K., Digesu, G.A., Dickinson, T., Goonewardene, S., McClurg, D., Rahnama'i, M.S. and Schizas, A., 2021. An International Continence Society (ICS) report on the terminology for pelvic floor muscle assessment. Neurourology and urodynamics, 40(5), pp.1217-1260.

Continence 19S (2026) 102567
DOI: 10.1016/j.cont.2026.102567