Combination of Skin Sympathetic Nerve Activity and Urine Biomarkers in Improving
Diagnostic Accuracy for Urge Urinary Incontinence
Topic of the Paper
The paper focuses on developing a more accurate and objective diagnostic approach for
urge urinary incontinence (UUI), a particularly bothersome subtype of overactive bladder
(OAB). UUI is characterized by involuntary leakage of urine accompanied by a sudden,
compelling urge to urinate. Because UUI can significantly affect a patient’s quality of life,
improving its diagnosis is critical. This study proposes an integrated method that combines
measurements of skin sympathetic nerve activity (SKNA)—an indicator of autonomic nervous
system function—with urinary biomarkers. By using both physiological and biochemical signals,
the authors aim to enhance the diagnostic accuracy and reliability of detecting UUI.
Previous Problems and Challenges
Traditionally, UUI has been diagnosed based on subjective tools such as patient-reported
symptom questionnaires, including the Overactive Bladder Symptom Score (OABSS) and the
International Prostate Symptom Score (IPSS). These assessments are limited by variability in
patient responses and lack of physiological grounding. Moreover, although previous studies have
explored the use of urinary biomarkers to detect lower urinary tract symptoms, their clinical
application remains limited due to insufficient sensitivity and specificity. Another challenge is
the inconsistent use of urine creatinine normalization in biomarker analysis, which affects the
comparability and reliability of results. Furthermore, earlier works lacked rigorous validation
strategies, such as cross-validation or hold-out test sets, raising concerns about the
generalizability of those findings in clinical settings.
Proposed Solutions by This Study
To overcome these challenges, this study introduces a novel diagnostic algorithm that
integrates autonomic nervous system parameters and urinary biomarkers. The researchers used a
noninvasive neuECG method to quantify SKNA, offering a real-time physiological indicator of
sympathetic activity. At the same time, they measured nine urinary cytokines and chemokines,
both in their raw forms and after normalization with urinary creatinine, to account for
concentration variability. These data sets were then analyzed using machine learning algorithms,
including support vector machines (SVM) and logistic regression with L1 regularization, which
allowed the identification of the most significant predictive variables while reducing model
complexity. A six-fold stratified cross-validation was employed to ensure robustness and to
simulate performance in real-world applications.
Experimental Results and Discussion
The experimental analysis revealed several important findings regarding the diagnostic
value of combining skin sympathetic nerve activity (SKNA) and urinary biomarkers in patients
with urge urinary incontinence (UUI). First, UUI participants exhibited significantly elevated
SKNA during both baseline and recovery phases when compared to non-UUI controls,
suggesting heightened sympathetic nervous system activity. This aligns with previous studies
that associate autonomic dysfunction with overactive bladder symptoms, supporting SKNA as a
valid physiological marker for UUI.
In addition to SKNA, the analysis of urinary biomarkers demonstrated that certain proinflammatory cytokines—especially MCP-1, MIP-1β, and IP-10—were notably higher in the
UUI group when normalized by urinary creatinine. These biomarkers are known to be involved
in chemotactic signaling and inflammation, which may reflect underlying bladder tissue irritation
or immune activation in UUI patients. Importantly, the normalized values provided more
consistent and predictive results than raw values, reinforcing the necessity of creatinine
correction to reduce intra-individual variation due to hydration or concentration differences in
urine samples.
When evaluating the diagnostic performance of different combinations, the most
effective model was the one that integrated both SKNA and calibrated biomarkers using a
support vector machine (SVM) algorithm. This model achieved an area under the curve (AUC)
of 0.80, with a sensitivity of 72.2% and specificity of 83.3%, representing a strong balance
between true positive and true negative rates. In contrast, models using only SKNA or only raw
biomarkers yielded significantly lower performance metrics, confirming that a multidimensional
approach is essential for accurate UUI diagnosis. The findings were further validated through
six-fold stratified cross-validation, adding robustness and minimizing the risk of overfitting.
Beyond diagnostic accuracy, the study also explored which features contributed most to
the model’s predictive power. Baseline SKNA, along with calibrated MCP-1, MIP-1β, and IP10, consistently had higher model weights across different cross-validation folds, indicating their
central role in distinguishing UUI from non-UUI cases. This reinforces the idea that both
autonomic imbalance and localized inflammation may serve as complementary indicators of UUI
pathology.
Overall, these results highlight the clinical promise of combining physiological and
biochemical assessments for diagnosing UUI. The approach offers a noninvasive, scalable, and
more objective alternative to conventional symptom-based tools. With further validation in
larger, more diverse populations, this model could be refined into a practical diagnostic aid for
frontline clinicians and contribute to more personalized treatment strategies for patients suffering
from overactive bladder syndromes.
Personal Reflection: Strengths and Areas for Improvement
One of the key strengths of this study is its innovative integration of physiological and
biochemical markers to improve the diagnosis of urge urinary incontinence (UUI). By combining
skin sympathetic nerve activity (SKNA) with creatinine-normalized urinary biomarkers, the
authors proposed a multidimensional diagnostic approach that moves beyond traditional reliance
on subjective questionnaires. This noninvasive strategy is both practical and clinically relevant,
particularly given the need for objective tools in the assessment of lower urinary tract symptoms.
The application of advanced machine learning techniques, such as support vector machines with
L1 regularization, and the use of stratified cross-validation further strengthen the credibility and
robustness of the study’s findings.
However, there are also several areas where the study could be improved. The relatively
small sample size and the predominance of female participants may limit the generalizability of
the results. Future studies would benefit from including a larger and more diverse population,
particularly with better gender balance. Additionally, certain potential confounding factors, such
as participants' diet, physical activity, and hormonal status, were not fully controlled during the
urine collection process, which could have influenced biomarker levels and SKNA readings.
While the diagnostic model demonstrated strong statistical performance, its feasibility and costeffectiveness in everyday clinical settings remain uncertain and warrant further investigation.
Overall, despite these limitations, the study offers valuable contributions and opens new avenues
for more accurate and individualized diagnosis of UUI.
In addition to its clinical significance, this study has broader implications for future
research and interdisciplinary applications. For example, the methodology of combining
physiological signals with molecular biomarkers could be extended to the diagnosis of other
disorders involving autonomic dysfunction, such as irritable bowel syndrome, fibromyalgia, or
even certain mental health conditions like anxiety. The concept of using skin sympathetic nerve
activity as a real-time indicator of bodily stress responses could also inform the development of
wearable health-monitoring devices, which may provide continuous assessments of a patient's
autonomic status. Moreover, the rigorous use of machine learning models in analyzing
biomedical data reflects a growing trend in precision medicine, where algorithms are
increasingly used to guide diagnostics and treatment planning. This study serves as a good
example of how engineering, medicine, and data science can work together to address real-world
healthcare challenges.
Resources:
Chen, YC., Chen, HW., Liu, TY. et al. Combination of skin sympathetic nerve activity and urine
biomarkers in improving diagnostic accuracy for urge urinary incontinence. Sci Rep 15, 14117
(2025). https://doi.org/10.1038/s41598-025-98889-x