On variability and detecting unreliable measurements in animal cystometry.

Introduction: Animal cystometry, a process of infusing fluid into the urinary bladder to evoke reflex contractions, is a common way to study the effects of pathology, injury, or experimental therapy on lower urinary tract (LUT) dynamics. By monitoring fluid movement during the cystometric micturition cycle, one can compute important quantities that indicate the health and function of the LUT, such as bladder capacity and voiding efficiency. Unfortunately, volume measurements in these difficult studies are often unpreventably corrupted by noise, leading to uncertainty when estimating key cystometric parameters. Methods: This work proposes a criterion, based on measurable quantities, that flags micturition cycles in cystometry studies that are likely to contain large measurement errors, potentially allowing experimenters to remove them from analysis to obtain a more accurate summary of LUT dynamics. Results: We describe the criterion, validate it against experimental data, and use computer simulations to demonstrate its utility.