The urinary bladder is a hollow muscular organ located in the pelvic floor, whose primary function is to store and release urine. The bladder is capable of altering its size and shape based on the volume of urine produced by the kidneys. Alterations in bladder size occur when stretch receptors are activated triggering the urge to urinate. When bladder dysfunction takes place, the inability to control urination results in a condition known as urinary incontinence (UI). UI affects 38.5% of U.S. men aged 60 years and above (Cao et al., 2022) and more than 60% of community dwelling adult women (Patel et al., 2022). In their study of the effect of UI on quality of life (QoL) in adults, Veronese, et al. reported that UI was associated with worse QoL (Veronese et al., 2022). Similarly, Cheng et al. explored the relationship between UI, depression, and anxiety and determined patients with UI had significantly higher levels of depression and anxiety compared to those without UI (Cheng et al., 2020). Beyond the impacts on individuals QoL, especially in older adults, UI (together with overactive bladder syndrome) was estimated to incur an economic burden of $65.9 billion on the US healthcare system in 2007 (Coyne et al., 2014).

With such a significant impact on people and the healthcare system, technologies to monitor the bladder are an active area of research. These technologies aim to support generating accurate data regarding the bladder to inform personalized treatment of those with bladder dysfunction. Continuous monitoring of bladder volume and pressure has the potential to help patients self-manage their conditions (by providing external cues regarding bladder volume for planning/execution of voiding), provide detailed information about bladder activity for clinicians, and evaluate changes in activity over time or in response to treatment. In their review of bladder monitoring systems, Semproni et al. identified imaging (e.g., ultrasound, electrical impedance tomography, and magnetic resonance imaging), localized tissue bioimpedance, and near-infrared spectroscopy as modalities that have been investigated for clinical and wearable bladder monitoring (Semproni et al., 2022). Bioimpedance sensing was identified as a promising modality for continuous monitoring (Semproni et al., 2022). This modality quantifies the passive, frequency-dependent electrical properties of a biological tissue, which are dependent on tissue type, structure/geometry, and fluid status (Dean et al., 2008; Grimnes and Martinsen, 2015). This association with fluid (Fenech and Jaffrin, 2004; Jaffrin and Morel, 2008) motivates the investigation of bioimpedance-based techniques for bladder volume monitoring. The non-invasive and automated evaluation of bladder volume has the potential to support diagnosis of bladder dysfunction by complementing or replacing bladder diaries as a clinical instrument. Bladder diaries are used to collect patient history on fluid intake, voiding patterns, and urine leakage events to assist clinicians in the diagnosis and treatment of UI (Gynecology & the American Urogynecologic Society, 2015). While providing critical information to a clinician, this instrument has limitations including compliance (e.g., failure to complete) and additional patient burden (Mehta et al., 2023). These issues could be resolved through the development of an accurate, reliable, non-invasive, and unobtrusive device for bladder volume monitoring.

To design such a device, bioimpedance technology is an attractive solution due to its low-cost (in comparison to other imaging modalities), safety (excitation currents are non-ionizing and below perception thresholds), and high potential for integration into wearable systems. Examples of recent reports of wearable bioimpedance systems focused on knee health (Teague et al., 2020; Critcher et al., 2023), the cardiovascular system (Wang et al., 2020; Qiu et al., 2022), and even bladder monitoring (Reichmuth et al., 2020) highlight this potential. Noyori et al. reported in their review of bioimpedance for bladder monitoring that this modality has been investigated for this application as early as 1975 (Noyori et al., 2022). Previous studies on this topic have reported abdomen impedance decreases due to increases in bladder volume and, conversely, abdomen impedance increases due to decreases (voiding) of the bladder (Abbey and Close, 1983; Shin et al., 2017; Gaubert et al., 2020; Reichmuth et al., 2020). These alterations are attributed to the fluid increase or decrease in the abdominal region. For these studies, the body position was controlled such that fluid changes in the abdomen could be attributed to changes in bladder volume. However, body position and changes in body position can induce fluid redistribution, which are not bladder volume related. Previous studies have reported changes in segmental resistance (which is the real component of measured impedance, discussed in further detail in later sections) resulting from transitions from sit to stand, sit to supine, supine to sit, etc., (BERG et al., 1993; Zhu et al., 1998; Fenech and Jaffrin, 2004; Yadollahi et al., 2015). From these works, a common trend is that segmental leg resistance changes with body positions as a result of the change in the hydrostatic gravitation gradient. More specifically, lying in a supine or head-down tilt (HDT) position increases the segmental leg resistance due to the movement of fluid out of this body segment and into the torso. Although this fluid shift has been previously documented, there has been limited investigation of the effect it could have on bladder monitoring using abdomen tissue bioimpedance. Only Shin et al. detailed how tissue resistance was altered during continuous bladder volume monitoring as a result of body positioning, reporting differences in abdomen resistance between sitting, standing, and supine positions collected over a short 10-min period (Shin et al., 2017). They attributed these changes to alterations in the shape of the abdomen (Shin et al., 2017). Their research did not explore how fluid shifts within the body, which may occur as a result of longer periods in different body positions, may impact the abdomen resistance and subsequent interpretation of bladder volume.

This provides the motivation for this work to evaluate the following question: are changes in abdominal resistance (a component of the overall tissue impedance) of human participants in response to short-term (4 h) 6° HDT body positioning similar to reported changes of participants in short-term supine (0°) body positioning from the literature? We hypothesize that the headword fluid shift induced by the HDT position will increase the abdomen resistance of this region and mask the resistance decrease caused by a bladder volume increase over the observation period. The use of the HDT position to induce fluid shifts, even though it is not a typical body position during daily living, is to facilitate comparison with available studies with participants in a supine (0°) position. The following sections outline the methods for collecting the abdomen tissue resistance from study participants, the methods for analysis, analysis results, and the discussion of these results in the context of previous bladder volume studies and the implications for wearable systems.

To evaluate the study question, the abdomen resistance from 10 to 100 kHz was collected from 10 healthy, young adults before and after their placement in a 6° HDT body position for 4 h. While tissue impedance (Z) has both resistance (R) and reactance (X) components, tissue resistance is most often associated with tissue fluids (Fenech and Jaffrin, 2004; Jaffrin and Morel, 2008). For this reason, this study collected and reported only the abdomen resistance and not abdomen reactance. It is important to note that these data were collected during the execution of the research study first reported by Freeborn et al. (2023) investigating segmental (leg, arm, and torso) fluid shifts during HDT body positioning. The presentation and analysis of the abdomen resistance and bladder volumes here represent the first reporting of these data.

The multi-frequency (10 kHz–100 kHz) abdomen resistance data collected from all 10 study participants at 0 min (black lines) and 240 min (red) during the HDT protocol are given in Figure 2. The 1% uncertainty intervals, selected based on reports by Fu and Freeborn regarding deviations of this magnitude with this instrument in high-residual impedance configurations (Fu and Freeborn, 2019), are shown as shaded regions for all measurements. The baseline resistance of participants ranges from 38.3 Ω to 127.4 Ω (mean: 75.6 ± 28 Ω) at 10 kHz and 24.4 Ω to 115.5 Ω (mean: 65.0 ± 28.2 Ω) at 100 kHz. From a visual inspection of participant resistances, seven of nine cases show that the 240 min values are greater in value than those collected at 0 min.

The 10 kHz and 100 kHz resistances of the study participants are summarized using boxplots in Figures 3A, B, respectively. A Wilcoxon signed-rank test was conducted to determine the effect of time in the HDT position on the abdomen resistance. All statistical tests were completed using R version 4.3.1 and RStudio (Version 2023.06.2 + 561). The median 10 kHz resistance increase (3.49 Ω) between measurements at 0 min and 240 min was not statistically significant (p = 0.16). Furthermore, the median 100 kHz resistance increase (5.22 Ω) between measurements at 0 min and 240 min was not statistically significant (p = 0.13).

The volume of urine voided by the study participants after 240 min in the HDT position is given in Figure 4A. Measured volumes range from 200 mL to 900 mL with an average of 506 mL. With participants voiding their bladder immediately before starting the HDT protocol, this suggests each had an increase in bladder volume over the 240 min period. Additionally, participants’ self-reported urge to urinate at 0 and 240 min is given in Figure 4B, with median values of 1 and 5, respectively. A Wilcoxon signed-rank test was conducted to evaluate the statistical significance of this observed time effect. The increase in the median urge to urinate between measurements at 0 and 240 min was statistically significant (p = 0.014). These self-reports further support the increase in urine volume in the bladder over the protocol period.

The average (and median) increase in abdomen resistance was 2.0 Ω (3.49 Ω) and 2.3 Ω (5.22 Ω) at 10 kHz and 100 kHz, respectively, after 4 h in the 6° HDT position, but these differences were not statistically significant (p < 0.05) based on the Wilcoxon signed-ranked tests. Over this time period, the bladder volume of these participants increased by an average of 506 mL (assuming they had an empty bladder after voiding prior to the protocol), which was expected to decrease the abdomen resistance in this region. This expectation is supported by previous studies reporting abdomen resistance from healthy adults in supine body positions.

• Abbey and Close reported statistically significant decreases with urinary bladder filling and increases with voiding from their study with 40 healthy adult participants. From visual review of data from individual participants in their study, changes up to approximately 3 Ω for the 75 kHz impedance were noted for bladder filling and voiding of volumes from 100 mL to 900 mL (Abbey and Close, 1983).

A summary of these comparison studies and the results from this work are presented in Table 2 (where N.S. denotes not statistically significant). While using impedance technology has been reported in other studies, with a 2022 review provided by Noyori et al. (2022), the comparison studies referenced here had similar electrode configurations and body positions as this work. Overall, the data reported in the three comparison studies support that abdomen resistance (in the band from 1 kHz to 100 kHz measured in a tetrapolar configuration) decreases approximately 1 Ω–3 Ω as bladder volume increases in healthy adult participants (in supine body positions). An important assumption underlying our summary is that when values are reported as impedance without clear reference to either resistance or impedance magnitude, the reported values do represent resistance or are a close approximation of it. However, this highlights a limitation of this comparison and the need for future studies to clearly report impedance datasets as resistance/reactance, impedance/phase, or other specific combinations to improve direct comparisons of reported data across studies. From the presented summary though, we assume that resistance decreases up to 3 Ω would be observed in the HDT study participants if abdomen resistance was dominated by bladder volume increases. However, it is significant that this was not observed even with the observed increase in bladder volume (quantified by the post-study bladder voiding) of the study participants.

The lack of significant changes in the abdomen resistance, even with a bladder volume increase, could be attributed to a fluid decrease in the abdomen (but not the bladder) from the fluid shift toward the neck/chest induced by the gravitational gradient in this body position. As reported by Freeborn et al. in their investigation of segmental resistance changes in this body position, this group of participants had increases of approximately 11%–12% for the 10 kHz segmental leg resistance and 9% for the 100 kHz segmental resistance (Freeborn et al., 2023). For reference, the segmental leg resistances of the study participants were collected using the electrode configurations shown in Figure 1. The resistance increases support that the fluid was leaving the legs and moving into the torso. While the distribution of fluid in the torso was not evaluated in that study, the increase in abdomen resistance noted here for the majority of participants supports that fluids from the legs did not settle in the abdomen. If that occurred, it would be expected to decrease the resistance of this region between the placed measurement electrodes.

Therefore, the fluid shift caused by the study body position appears to be the dominant feature in the collected abdomen resistance of the participants studied, offsetting the changes in abdomen resistance resulting from the bladder filling. The significance of this is the impact that fluid shifts in the body could have on the interpretation of the abdomen resistance for bladder volume and activity monitoring. As an increase in abdomen resistance is typically interpreted as a decrease in bladder volume, if body position (and length of time in body position) is not factored into analysis/interpretation, incorrect volume estimates could be generated from a bioimpedance-based system.

Limitations of the current study that impact the conclusions that can be drawn from this observational data include the following:

• The electrodes were placed using relative spacing between pairs (I+/V+ and I−/V−) for each person with the intent of measuring relatively similar abdomen regions relative to body size. This approach may alter the contribution of the urinary bladder to the abdomen impedance based on the induced current distribution in the body. The two participants with resistance decreases in Figure 3 could be a result of this, with their abdomen resistance having higher sensitivity to bladder volume based on electrode spacing and bladder location within their body. However, it is not possible to evaluate this aspect using only the abdomen resistance available in this study and requires further future investigation.

While bioimpedance-based systems continue to have the potential to support bladder monitoring applications, it is clear that further focus is needed to advance knowledge relating to resistance data (and also reactance data, which have yet to be explored) to bladder volumes/activity and to develop robust wearable solutions for free-living monitoring. Future recommendations for such systems include functionality to 1) identify data artifacts from motion or contraction, 2) correct or remove data artifacts to limit misinterpretation, 3) identify body position and length of time in a body position to estimate additional fluid shifts, 4) differentiate between fluid shifts attributed to bladder volume changes and changes due to body position, and 5) compensate for electrode movement (both within and between days) to increase the usability of this system. Beyond this technical set of features, it will be important to design systems to meet the unique needs of the groups these systems aim to support. With the significant impact that UI has on older adult populations, it will be important to evaluate the specific needs of this population and how to design these systems to meet them and be easily usable.

Achieving these functionalities may require the use of additional sensing modalities to identify both tissue impedance and context of collected measurements. Future studies should explore sensing modalities that complement bioimpedance to capture movement (e.g., inertial measurement units) and muscle activity (e.g., surface electromyography). The additional sensing modalities are expected to increase the accuracy of fluid estimates by helping identify when motion-related artifacts are introduced to the resistance data, which can be used to either remove or correct degraded data. These future efforts are expected to help advance solutions to provide comprehensive bladder monitoring to support the personalized and continuous monitoring of people with bladder dysfunction.

There were no statistically significant differences in the abdomen resistance of the study participants after 4 h in the 6° HDT position even though an average bladder volume increase of 506 mL was observed. The expected resistance decreases from increases in bladder volume are believed to have been masked by the fluid shifts induced by the gravitational gradient in this position. This highlights that body position and fluid redistribution from it can significantly impact the interpretation of abdomen resistance for bladder monitoring applications. Therefore, future bioimpedance-based wearable systems for bladder volume/activity monitoring will require methods to differentiate bladder fluid changes from body position-based fluid redistribution for accurate and robust reporting.