Blood and breath alcohol concentration from transdermal alcohol biosensor data: estimation and uncertainty quantification via forward and inverse filtering for a covariate-dependent, physics-informed, hidden Markov model*

Abstract

Transdermal alcohol biosensors that do not require active participation of the subject and yield near continuous measurements have the potential to significantly enhance the data collection abilities of alcohol researchers and clinicians who currently rely exclusively on breathalyzers and drinking diaries. Making these devices accessible and practical requires that transdermal alcohol concentration (TAC) be accurately and consistently transformable into the well-accepted measures of intoxication, blood/breath alcohol concentration (BAC/BrAC). A novel approach to estimating BrAC from TAC based on covariate-dependent physics-informed hidden Markov models with two emissions is developed. The hidden Markov chain serves as a forward full-body alcohol model with BrAC and TAC, the two emissions, assumed to be described by a bivariate normal which depends on the hidden Markovian states and person-level and session-level covariates via built-in regression models. An innovative extension of hidden Markov modeling is developed wherein the hidden Markov model framework is regularized by a first-principles PDE model to yield a hybrid that combines prior knowledge of the physics of transdermal ethanol transport with data-based learning. Training, or inverse filtering, is effected via the Baum-Welch algorithm and 256 sets of BrAC and TAC signals and covariate measurements collected in the laboratory. Forward filtering of TAC to obtain estimated BrAC is achieved via a new physics-informed regularized Viterbi algorithm which determines the most likely path through the hidden Markov chain using TAC alone. The Markovian states are decoded and used to yield estimates of BrAC and to quantify the uncertainty in the estimates. Numerical studies are presented and discussed. Overall good agreement between BrAC data and estimates was observed with a median relative peak error of 22% and a median relative area under the curve error of 25% on the test set. We also demonstrate that the physics-informed Viterbi algorithm eliminates non-physical artifacts in the BrAC estimates.

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

“…Our research group consisting of applied mathematicians/control engineers, psychiatrists and psychologists has developed a number of techniques for estimating BAC or BrAC from observations of biosensor measured TAC based on either first principles physics-based models [11][12][13] or data-driven machine learning techniques such as physics informed artificial neural networks 14,15 and hidden Markov models. 16 In addition, several other research groups have developed standard regression models [17][18][19] and machine learning techniques such as random forest-like extra-trees 20 to convert TAC into either BAC or BrAC.…”

Signal processing and machine learning with transdermal alcohol concentration to predict natural environment alcohol consumption.

“…The Skyn samples TAC, temperature, and motion every 20 s to facilitate highresolution analysis of alcohol consumption. Prior work using the Skyn and other devices has focused on converting measured TAC into standard drinks and estimated blood alcohol concentration (eBAC) (Dougherty et al, 2012;Fairbairn et al, 2019;Fairbairn & Kang, 2021;Karns-Wright et al, 2018;Norman et al, 2021;Oszkinat, Shao, et al, 2022). However, without reliable methods for signal processing and modeling, TAC cannot be directly translated into estimated alcohol use or eBAC because biological factors (e.g., metabolism and skin properties) and contextual factors (e.g., exercise, environmental temperature, alcohol spills) produce TAC readings that vary across individuals and settings (Barnett et al, 2014;Fairbairn & Kang, 2021;Luczak & Ramchandani, 2019;Swift, 2003;van Egmond et al, 2020).…”

Real-time recursive estimation of, and uncertainty quantification for, breath alcohol concentration via LQ tracking control-based inverse filtering of transdermal alcohol biosensor signals

“…TAC to BAC or BrAC conversion models by other researchers have included conventional linear regression models [11-13, 25, 26, 28] and, more recently, machine learning-based schemes [15]. Our group has also looked at physics informed machine learning-based schemes using hidden Markov models [38], generative adversarial neural networks [37], and long short-term memory neural networks [36].…”