Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

The Effects of Personalized Feedback About ALDH2*2, Alcohol Use, and Associated Health Risks on Drinking Intention and Consumption: The Role of Self-Efficacy and Perceived Threat.

Alcohol, clinical & experimental research·2026
Same author

Reply to Ko, Null within-twin estimates on education and dementia: cautions for within-family contrasts.

European journal of epidemiology·2025
Same author

Is educational attainment protective against developing dementia? A twin study of genetic and environmental contributions.

European journal of epidemiology·2025
Same author

Genetic and environmental influences on alcohol consumption in middle to late life.

Psychology of addictive behaviors : journal of the Society of Psychologists in Addictive Behaviors·2025
Same author

REAL-TIME RECURSIVE ESTIMATION OF, AND UNCERTAINTY QUANTIFICATION FOR, BREATH ALCOHOL CONCENTRATION VIA LQ TRACKING CONTROL-BASED INVERSE FILTERING OF TRANSDERMAL ALCOHOL BIOSENSOR SIGNALS.

Applied mathematics for modern challenges·2024
Same author

A population model-based linear-quadratic Gaussian compensator for the control of intravenously infused alcohol studies and withdrawal symptom prophylaxis using transdermal sensing.

Optimal control applications & methods·2024

Related Experiment Video

Updated: Jul 12, 2025

Fast and Accurate Exhaled Breath Ammonia Measurement
06:27

Fast and Accurate Exhaled Breath Ammonia Measurement

Published on: June 11, 2014

13.5K

An Abstract Parabolic System-Based Physics-Informed Long Short-Term Memory Network for Estimating Breath Alcohol

Clemens Oszkinat1, Susan E Luczak2, I Gary Rosen1

  • 1Department of Mathematics, University of Southern California, Los Angeles, 90089, CA, USA.

Neural Computing & Applications
|October 24, 2023
PubMed
Summary
This summary is machine-generated.

A new physics-informed LSTM model accurately estimates breath alcohol concentration from transdermal alcohol biosensor data. This advanced method significantly reduces errors compared to previous models, offering a more reliable alternative for alcohol monitoring.

Keywords:
Biosensor dataLSTMPhysics-Informed Machine LearningTransdermal alcohol concentration

More Related Videos

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

3.8K
Real-time Breath Analysis by Using Secondary Nanoelectrospray Ionization Coupled to High Resolution Mass Spectrometry
08:23

Real-time Breath Analysis by Using Secondary Nanoelectrospray Ionization Coupled to High Resolution Mass Spectrometry

Published on: March 9, 2018

9.0K

Related Experiment Videos

Last Updated: Jul 12, 2025

Fast and Accurate Exhaled Breath Ammonia Measurement
06:27

Fast and Accurate Exhaled Breath Ammonia Measurement

Published on: June 11, 2014

13.5K
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

3.8K
Real-time Breath Analysis by Using Secondary Nanoelectrospray Ionization Coupled to High Resolution Mass Spectrometry
08:23

Real-time Breath Analysis by Using Secondary Nanoelectrospray Ionization Coupled to High Resolution Mass Spectrometry

Published on: March 9, 2018

9.0K

Area of Science:

  • Biomedical Engineering
  • Machine Learning
  • Physiology

Background:

  • Transdermal alcohol concentration (TAC) offers a promising, less invasive alternative to traditional breathalyzers or self-reported drinking diaries for estimating alcohol levels.
  • Accurate estimation of breath alcohol concentration (BAC) is crucial for various applications, including public health and legal compliance.
  • Existing methods for BAC estimation from TAC data face challenges with data variability and accuracy.

Purpose of the Study:

  • To develop and validate a novel physics-informed Long Short-Term Memory (LSTM) network for estimating BAC from TAC data.
  • To integrate a physics-based model of ethanol diffusion into the LSTM architecture to improve estimation accuracy and physical interpretability.
  • To quantify the uncertainty of BAC estimations and investigate the influence of various covariates on the model's performance.

Main Methods:

  • A physics-informed LSTM network was developed, incorporating a first-principles physics-based population model for ethanol diffusion through skin as a regularization term.
  • The model utilized a dataset of controlled laboratory-collected breath and TAC data from 40 participants, including various personal, physiological, and environmental covariates.
  • Uncertainty quantification was achieved using Monte Carlo dropout, and the importance of covariates was assessed using Shapley values.

Main Results:

  • The physics-informed LSTM model demonstrated successful application to both training and testing datasets, showing improved generalization ability for new drinking episodes.
  • The integration of physics-based information enhanced model performance, and uncertainty quantification provided credible bands that accurately captured the true BAC signal.
  • The proposed model achieved significant reductions in relative error (58% and 72%) and relative peak error (33% and 76%) compared to previous machine learning models.

Conclusions:

  • Physics-informed machine learning, specifically the developed LSTM network, provides a robust and accurate method for estimating BAC from TAC data.
  • The model's ability to incorporate physiological principles and quantify uncertainty makes it a valuable tool for reliable alcohol monitoring.
  • This approach significantly outperforms existing machine learning models, paving the way for improved transdermal alcohol biosensor applications.