pith. sign in

arxiv: 2604.15381 · v1 · submitted 2026-04-16 · 🪐 quant-ph

Hydration Monitoring Using Urinary Biomarkers: A Hybrid Classical Quantum Predictive Modeling Framework

Saul Gonzalez-Bermejo , Tommaso Albrigi , Borja Vazquez-Morado , Urko Regueiro-Ramos , Daniel Casado-Fauli , Sergi Consul-Pacareu , Laia Alentorn , Jordi Ferre
show 2 more authors
Valentino Asole Parfait Atchade-Adelomou
This is my paper

Pith reviewed 2026-05-10 11:52 UTC · model grok-4.3

classification 🪐 quant-ph
keywords hydration monitoringurinary biomarkersquantum machine learningvariational quantum circuitshybrid classical-quantum modelsregression predictiondigital health
0
0 comments X

The pith

Hybrid quantum-classical models predict hydration status from urinary biomarkers collected via smart toilets.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper sets out to show that variational quantum circuits and a new modular hybrid architecture can perform regression to estimate hydration levels from passive urinary measurements such as specific gravity, conductivity, and volume. A reader would care because continuous, non-invasive monitoring of this key health indicator could support earlier intervention in dehydration-related conditions without requiring active user effort. The work also examines how near-term quantum hardware performs on real physiological data compared with standard classical regressors.

Core claim

The authors introduce the Quantum Sequential Model, a modular framework that assembles flexible hybrid quantum-classical pipelines, and demonstrate through direct comparison that symmetry-constrained quantum regressors and the QSM architecture can be applied to the regression task on urinary biomarker data, yielding insights into both the feasibility of quantum machine learning for digital health and the practical constraints of current quantum devices.

What carries the argument

The modular Quantum Sequential Model (QSM), which assembles variational quantum circuits into configurable hybrid pipelines to regress hydration status from biomarker inputs.

If this is right

  • Classical regression baselines can be directly benchmarked against quantum and hybrid variants on the same continuous hydration prediction task.
  • Near-term quantum hardware can be evaluated for its ability to process physiological time-series data in health-monitoring pipelines.
  • Passive biomarker collection enables ongoing regression without requiring users to provide active samples.
  • The modular design allows incremental addition of quantum layers to existing classical health models.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the hybrid approach scales, similar quantum-classical pipelines could be tested on other continuous biomarkers such as electrolyte balance or metabolic markers.
  • Hardware limitations noted here point to the value of testing error-mitigation techniques before expecting quantum advantage in sensor-driven health tasks.
  • The framework could be extended to multi-output regression that jointly predicts hydration together with related renal-function indicators.

Load-bearing premise

The urinary biomarkers gathered by the smart toilet system carry enough information about hydration status to permit useful regression predictions from either classical or quantum models.

What would settle it

A held-out test set on which all tested models, classical and quantum alike, produce predictions no better than a constant mean baseline would show that the biomarker data lack sufficient signal for the claimed regression task.

Figures

Figures reproduced from arXiv: 2604.15381 by Borja Vazquez-Morado, Daniel Casado-Fauli, Jordi Ferre, Laia Alentorn, Parfait Atchade-Adelomou, Saul Gonzalez-Bermejo, Sergi Consul-Pacareu, Tommaso Albrigi, Urko Regueiro-Ramos, Valentino Asole.

Figure 1
Figure 1. Figure 1: Visual model performance comparison on the Conductivity dataset. [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Visual model performance comparison on the USG dataset. Top: true [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
read the original abstract

Hydration status is a key physiological indicator associated with cellular homeostasis, renal function, and overall health. Recent advances in smart sensing environments enable passive monitoring of urinary biomarkers that can provide continuous insight into hydration dynamics. In this work, we investigate predictive modeling approaches for hydration monitoring using biomarker data collected through the Predict Health Toilet (PHT) system. The problem is formulated as a regression task using urinary indicators such as urine specific gravity, conductivity, and volume. We evaluate classical machine learning models and quantum machine learning architectures based on variational quantum circuits. In particular, we introduce a modular Quantum Sequential Model (QSM) designed to construct flexible hybrid quantum classical predictive pipelines. Experimental results compare classical regression models, symmetry-constrained quantum regressors, and QSM architectures. The results provide insights into the potential role of quantum machine learning in digital health monitoring systems and highlight the opportunities and current limitations of near-term quantum computing for physiological data analysis.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The manuscript formulates hydration monitoring as a regression task using urinary biomarkers (specific gravity, conductivity, volume) collected via the Predict Health Toilet system. It evaluates classical machine learning regressors against variational quantum circuit models and introduces a modular Quantum Sequential Model (QSM) for constructing hybrid quantum-classical pipelines. Experimental comparisons are claimed to yield insights into the role of quantum machine learning in digital health monitoring and the limitations of near-term quantum hardware for physiological data analysis.

Significance. If the unreported experimental comparisons were to demonstrate that the QSM or symmetry-constrained quantum regressors achieve meaningfully better generalization or extract physiological signal beyond what classical baselines capture, the work could illustrate a concrete use case for variational quantum algorithms in passive health sensing. The modular QSM design, if rigorously specified, would be a constructive contribution toward reusable hybrid pipelines. At present the absence of any performance numbers prevents evaluation of whether these potential benefits are realized.

major comments (3)
  1. Abstract: the statement that 'experimental results compare classical regression models, symmetry-constrained quantum regressors, and QSM architectures' and 'provide insights' is unsupported because no R², RMSE, correlation coefficients, or baseline metrics are supplied; without these the central claim that the biomarkers contain usable signal for regression cannot be assessed.
  2. Methods/Results (QSM definition): the Quantum Sequential Model is introduced as a 'modular' hybrid architecture, yet no circuit diagram, variational ansatz, or explicit composition rule with classical layers is given, making it impossible to determine whether QSM differs substantively from standard variational quantum regressors or merely renames them.
  3. Results section: the comparison of classical, symmetry-constrained quantum, and QSM models is described at high level only; the lack of dataset size, train/test protocol, hyperparameter details, or statistical significance tests renders any conclusion about 'opportunities and current limitations of near-term quantum computing' untestable and non-reproducible.
minor comments (2)
  1. The abstract would be strengthened by a single sentence summarizing the best quantitative result (e.g., 'QSM achieved R² = X on held-out data versus Y for the classical baseline').
  2. Notation for the variational parameters and symmetry constraints should be defined explicitly the first time they appear rather than left implicit.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the careful and constructive review of our manuscript. The comments highlight important areas for improving clarity, reproducibility, and support for our claims. We address each major comment below and commit to revisions that strengthen the paper without altering its core contributions.

read point-by-point responses

  1. Referee: Abstract: the statement that 'experimental results compare classical regression models, symmetry-constrained quantum regressors, and QSM architectures' and 'provide insights' is unsupported because no R², RMSE, correlation coefficients, or baseline metrics are supplied; without these the central claim that the biomarkers contain usable signal for regression cannot be assessed.

    Authors: We agree that the abstract requires concrete quantitative support to substantiate its claims. The current version outlines the experimental framework and comparisons but does not embed the specific performance metrics. In the revised manuscript, we will report the R², RMSE, and correlation coefficients for the classical regression models, symmetry-constrained quantum regressors, and QSM architectures, along with baseline comparisons. This will directly address the assessability of the biomarkers' signal and the insights provided. revision: yes

  2. Referee: Methods/Results (QSM definition): the Quantum Sequential Model is introduced as a 'modular' hybrid architecture, yet no circuit diagram, variational ansatz, or explicit composition rule with classical layers is given, making it impossible to determine whether QSM differs substantively from standard variational quantum regressors or merely renames them.

    Authors: The QSM is presented in the methods as a modular framework for sequential hybrid quantum-classical pipelines, with its structure described textually. We acknowledge that the absence of a diagram and explicit specifications limits evaluation of its novelty. In the revision, we will add a circuit diagram, specify the variational ansatz (including any symmetry constraints), and provide explicit composition rules for integrating classical layers. This will clarify its distinction from standard variational quantum regressors and support its reusability as a constructive contribution. revision: yes

  3. Referee: Results section: the comparison of classical, symmetry-constrained quantum, and QSM models is described at high level only; the lack of dataset size, train/test protocol, hyperparameter details, or statistical significance tests renders any conclusion about 'opportunities and current limitations of near-term quantum computing' untestable and non-reproducible.

    Authors: We recognize that the results section is currently high-level and lacks the details needed for full reproducibility and evaluation of the conclusions. The revised manuscript will include the dataset size from the Predict Health Toilet system, the train/test protocol (including any cross-validation), hyperparameter details and optimization methods, and statistical significance tests (e.g., for comparing model performances). These additions will make the reported opportunities and limitations of near-term quantum computing testable and reproducible. revision: yes

Circularity Check

0 steps flagged

No circularity: standard empirical model comparison on biomarker data

full rationale

The paper formulates the task as regression on urinary biomarkers (specific gravity, conductivity, volume) collected via the PHT system and directly compares classical regressors against variational quantum circuits and a modular QSM architecture. No equations, parameter fits, or derivations are described that reduce to their own inputs by construction; there are no self-definitional steps, fitted inputs renamed as predictions, or load-bearing self-citations that justify uniqueness or ansatzes. The central claim rests on experimental results from observed data rather than any tautological reduction, making the work self-contained as a straightforward empirical evaluation.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 1 invented entities

The abstract relies on standard domain assumptions about biomarker reliability and typical machine-learning modeling choices; the only novel element is the named QSM architecture whose details are not provided.

free parameters (1)
  • Variational parameters in quantum circuits
    Standard trainable parameters in variational quantum circuits; number and initialization unspecified.
axioms (1)
  • domain assumption Urinary biomarkers (specific gravity, conductivity, volume) are reliable and sufficient indicators for regression-based hydration prediction.
    This premise underpins the entire regression task formulation.
invented entities (1)
  • Quantum Sequential Model (QSM) no independent evidence
    purpose: Modular architecture for constructing flexible hybrid quantum-classical predictive pipelines.
    Newly named component introduced to organize the hybrid models.

pith-pipeline@v0.9.0 · 5510 in / 1423 out tokens · 56187 ms · 2026-05-10T11:52:07.806042+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

20 extracted references · 20 canonical work pages · 2 internal anchors

  1. [1]

    Water, hydration, and health.Nutrition Reviews, 68(8):439–458, 2010

    Barry M Popkin, Kristen E D’Anci, and Irwin H Rosenberg. Water, hydration, and health.Nutrition Reviews, 68(8):439–458, 2010

    work page 2010

  2. [2]

    Hydration assessment of athletes.Sports Medicine, 44(1):97–104, 2014

    Samuel N Cheuvront and Robert W Kenefick. Hydration assessment of athletes.Sports Medicine, 44(1):97–104, 2014

    work page 2014

  3. [3]

    Urine specific gravity and hydration assessment

    Lawrence E Armstrong. Urine specific gravity and hydration assessment. International Journal of Sport Nutrition and Exercise Metabolism, 17(4):350–360, 2007

    work page 2007

  4. [4]

    A smart toilet system for integrated health monitoring.Nature Biomedical Engineering, 4:624–635, 2020

    Seung-Min Park et al. A smart toilet system for integrated health monitoring.Nature Biomedical Engineering, 4:624–635, 2020

    work page 2020

  5. [5]

    A guide to deep learning in healthcare.Nature Medicine, 25:24–29, 2019

    Andre Esteva et al. A guide to deep learning in healthcare.Nature Medicine, 25:24–29, 2019

    work page 2019

  6. [6]

    An introduc- tion to quantum machine learning.Contemporary Physics, 56(2):172– 185, 2015

    Maria Schuld, Ilya Sinayskiy, and Francesco Petruccione. An introduc- tion to quantum machine learning.Contemporary Physics, 56(2):172– 185, 2015

    work page 2015

  7. [7]

    Quantum machine learning.Nature, 549:195–202, 2017

    Jacob Biamonte et al. Quantum machine learning.Nature, 549:195–202, 2017

    work page 2017

  8. [8]

    Springer, 2006

    Christopher M Bishop.Pattern Recognition and Machine Learning. Springer, 2006

    work page 2006

  9. [9]

    Springer, 2009

    Trevor Hastie, Robert Tibshirani, and Jerome Friedman.The Elements of Statistical Learning. Springer, 2009

    work page 2009

  10. [10]

    Springer, 2016

    Ian T Jolliffe and Jorge Cadima.Principal Component Analysis. Springer, 2016

    work page 2016

  11. [11]

    Quantum computing in the nisq era and beyond.Quantum, 2:79, 2018

    John Preskill. Quantum computing in the nisq era and beyond.Quantum, 2:79, 2018

    work page 2018

  12. [12]

    Unique: A general-purpose platform for benchmarking classical and quantum machine learning algorithms.TechRxiv, 2025

    Daniel Casado-Faul ´ı, Sa´ul Gonz ´alez-Bermejo, Marc Rovira, Alba Mei, Laia Coronas Sala, Sergi C `onsul-Pacareu, Esteve Vilella, Jordi Alb ´o- Canals, Parfait Atchade-Adelomou, Xavier Vilasis-Cardona, and Elisabet Golobardes Ribe. Unique: A general-purpose platform for benchmarking classical and quantum machine learning algorithms.TechRxiv, 2025

    work page 2025

  13. [13]

    Adam: A Method for Stochastic Optimization

    Diederik P. Kingma and Jimmy Ba. Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980, 2017

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  14. [14]

    PennyLane: Automatic differentiation of hybrid quantum-classical computations

    Ville Bergholm, Josh Izaac, Maria Schuld, Christian Gogolin, Carsten Blank, Keri McKiernan, Johannes Jakob Meyer, Nathan Killoran, et al. Pennylane: Automatic differentiation of hybrid quantum-classical com- putations.arXiv preprint arXiv:1811.04968, 2022

    work page internal anchor Pith review arXiv 2022

  15. [15]

    Adri ´an P ´erez-Salinas, Alba Cervera-Lierta, Elies Gil-Fuster, and Jos ´e I. Latorre. Data re-uploading for a universal quantum classifier.Quantum, 4:226, 2020

    work page 2020

  16. [16]

    Fourier se- ries weight in quantum machine learning.arXiv preprint arXiv:2302.00105, 2023

    Parfait Atchade-Adelomou and Kent Larson. Fourier series weight in quantum machine learning.arXiv preprint arXiv:2302.00105, 2023

    work page arXiv 2023

  17. [17]

    Hall.Lie Groups, Lie Algebras, and Representations: An Elementary Introduction, volume 222 ofGraduate Texts in Mathematics

    Brian C. Hall.Lie Groups, Lie Algebras, and Representations: An Elementary Introduction, volume 222 ofGraduate Texts in Mathematics. Springer, 2 edition, 2015

    work page 2015

  18. [18]

    Xgboost: A scalable tree boosting system

    Tianqi Chen and Carlos Guestrin. Xgboost: A scalable tree boosting system. InProceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’16, page 785–794. ACM, August 2016

    work page 2016

  19. [19]

    https://xgboost.readthedocs.io/ en/latest/python/index.html

    Xgboost python package documentation. https://xgboost.readthedocs.io/ en/latest/python/index.html. [Accessed 04-03-2026]

    work page 2026

  20. [20]

    Bergstra, D

    J. Bergstra, D. Yamins, and D. D. Cox. Making a science of model search, 2012

    work page 2012