ML-UrineQuant - ML-UrineQuant A machine learning program for identifying and quantifying mouse urine on absorbent paper - Background and impact

Aug 5, 2026

Upload by Rowan

Void spot assay (VSA) supports functional bladder voiding assessment in mice, but quantifying urine spot size and spatial distribution from filter-paper images is hindered by interlaboratory variability in image contrast, resolution, and non-void artifacts. ML-UrineQuant addresses this by using a Mask R-CNN–based machine learning algorithm trained for object recognition to detect and quantitate urine spots across a broad range of spot sizes. The model maintains high accuracy under different illumination and contrast conditions. It enables individual laboratories to fine-tune performance to their specific image characteristics.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Introduction

Void spot assay (VSA)

Challenges in image analysis

Limitations of existing software

Image-contrast requirements and variability

The Jackson Laboratory The Mouseion at the JAXlibrary | Faculty Research 2025 | Faculty & Staff Research | | --- | --- | 3-1-2025 ML-UrineQuant: A machine learning program for identifying and quantifying mouse urine on absorbent paper. Warren G Hill Bryce MacIver Gary Churchill Mariana G DeOliveira Mark L Zeidel See next page for additional authors Follow this and additional works at: [https://mouseion.jax.org/stfb2025](https://mouseion.jax.org/stfb2025) Authors Warren G Hill, Bryce MacIver, Gary Churchill, Mariana G DeOliveira, Mark L Zeidel, and Marcelo Cicconet Received: 30 October 2024 | Revised: 20 January 2025 | Accepted: 30 January 2025 DOI: 10.14814/phy2.70243 METHODS ARTICLE ML-UrineQuant: A machine learning program for identifying and quantifying mouse urine on absorbent paper Warren G. Hill1 | Bryce MacIver1 | Gary A. Churchill2 | Mariana G. DeOliveira3 | Mark L. Zeidel1 | Marcelo Cicconet4 1Laboratory of Voiding Dysfunction, Division of Nephrology, Department of Medicine, Beth Israel Deaconess Medical Center & Harvard Medical School, Boston, Massachusetts, USA 2The Jackson Laboratory, Bar Harbor, Maine, USA 3Laboratory of Pharmacology, Sao Francisco University, Sao Paulo, Brazil 4Image and Data Analysis Core, Department of Cell Biology, Harvard Medical School, Boston, Massachusetts, USA Correspondence Warren G. Hill, Laboratory of Voiding Dysfunction, Division of Nephrology, Department of Medicine, Beth Israel Deaconess Medical Center & Harvard Medical School, RN349B, 99 Brookline Ave, Boston, MA 02215, USA. [Email:](Email: whill@bidmc.harvard.edu)[ whill@bidmc.harvard.edu](Email: whill@bidmc.harvard.edu) Present address Marcelo Cicconet, Deliberate Solutions Inc., New York, New York, USA Funding information Jackson Laboratory (JAX), Grant/ Award Number: P30AG038070; HHS | NIH | National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), Grant/Award Number: P20DK097818 Abstract The void spot assay has gained popularity as a way of assessing functional bladder voiding parameters in mice, but analyzing the size and distribution of urine spot patterns on filter paper with software remains problematic due to interlaboratory differences in image contrast and resolution quality and non-void artifacts. We have developed a machine learning algorithm based on Region-based Convolutional Neural Networks (Mask-RCNN) that was trained in object recognition to detect and quantitate urine spots across a broad range of sizes—MLUrineQuant. The model proved extremely accurate at identifying urine spots ina wide variety of illumination and contrast settings. The overwhelming advantage it offers over current algorithms will be to allow individual labs to fine-tune the model on their specific images regardless of the image characteristics. This should be a valuable tool for anyone performing lower urinary tract research using mouse models. KEYWORDS artificial intelligence, mice, micturition, python, urology, void spot on paper, voiding dysfunction, VSA, VSOP This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2025 The Author(s). Physiological Reports published by Wiley Periodicals LLC on behalf of The Physiological Society and the American Physiological Society. Physiological Reports. 2025;13:e70243 . [https://doi.org/10.14814/phy2.70243](https://doi.org/10.14814/phy2.70243) [wileyonlinelibrary.com/journal/phy2](wileyonlinelibrary.com/journal/phy2) 1 of 10 2 of 10 HILL et al. 1 | INTRODUCTION In lower urinary tract research using mice, a common technique for assessing bladder and urethral function is the void spot assay (VSA) . This simple noninvasive assay involves placing mice in a cage with some form of absorbent paper on the cage floor, and after a period of time, recovering the paper to as

FAQs

What problem does ML-UrineQuant address in the void spot assay workflow?

It targets the difficulty of accurately analyzing urine spot size and distribution on absorbent paper when image contrast, resolution quality, and non-void artifacts vary between laboratories.

How does ML-UrineQuant detect and quantify urine spots?

It uses a machine learning model based on Region-based Convolutional Neural Networks (Mask-RCNN), trained for object recognition to detect and quantitate urine spots across multiple sizes.

Why can current image analysis software be unreliable across different labs?

Because many tools rely on thresholding that requires specific image features—especially strong pixel intensity contrast between urine and paper—which can change with mouse physiology, paper type, and UV illumination quality and wavelength.

Aug 5, 2026

Upload by Rowan