Search

Driver status monitoring systems are a vital component of smart cars in the future, especially in the era when an increasing amount of time is spent in the vehicle. The heart rate (HR) is one of the most important physiological signals of driver status. To infer HR of drivers, the mainstream of the existing research focused on capturing subtle heartbeat-induced vibration of the torso or leveraged photoplethysmography (PPG) that detects cardiac cycle-related blood volume changes in the microvascular. However, existing approaches rely on dedicated sensors that are expensive and cumbersome to be integrated or are vulnerable to ambient noise. Moreover, their performance on the detection of HR does not guarantee a reliable computation of the HR variability (HRV) measure, which is a more applicable metric for inferring mental and physiological status. The accurate computation HRV measure is based on the precise measurement of the beat-to-beat interval, which can only be accomplished by medical-grade devices that attach electrodes to the body. Considering these existing challenges, we proposed a facial expression-based HRV estimation solution. The rationale is to establish a link between facial expression and heartbeat since both are controlled by the autonomic nervous system. To solve this problem, we developed a tree-based probabilistic fusion neural network approach, which significantly improved HRV estimation performance compared to conventional random forest or neural network methods and the measurements from smartwatches. The proposed solution relies only on commodity camera with a lightweighted algorithm, facilitating its ubiquitous deployment in current and future vehicles. Our experiments are based on 3400 km of driving data from nine drivers collected in a naturalistic field study.

Aim To develop a non-invasive hypoglycemia detection approach using smartwatch data. Research Design and Methods We prospectively collected data from two wrist-worn wearables (Garmin vivoactive 4S; Empatica E4) and continuous glucose monitoring values in adults with diabetes mellitus on insulin treatment. Using this data, we developed a machine learning (ML) approach to detect hypoglycemia ( Results Twenty-two individuals were included in the final analysis (age 54.5±15.2y, HbA1c 6.9±0.6%, 16 male). Hypoglycemia was detected with an area under the receiver operating characteristic curve of 0.76±0.07 solely based on wearable data. Feature analysis revealed that the ML model associated increased heart rate, decreased heart rate variability, and increased tonic electrodermal activity with hypoglycemia. Conclusions Our approach may allow for non-invasive hypoglycemia detection using wearables in people with diabetes and thus complement existing methods for hypoglycemia detection and warning.

Excessive alcohol consumption causes disability and death. Digital interventions are promising means to promote behavioral change and thus prevent alcohol-related harm, especially in critical moments such as driving. This requires real-time information on a person's blood alcohol concentration (BAC). Here, we develop an in-vehicle machine learning system to predict critical BAC levels. Our system leverages driver monitoring cameras mandated in numerous countries worldwide. We evaluate our system with n = 30 participants in an interventional simulator study. Our system reliably detects driving under any alcohol influence (area under the receiver operating characteristic curve [AUROC] 0.88) and driving above the WHO recommended limit of 0.05 g/dL BAC (AUROC 0.79). Model inspection reveals reliance on pathophysiological effects associated with alcohol consumption. To our knowledge, we are the first to rigorously evaluate the use of driver monitoring cameras for detecting drunk driving. Our results highlight the potential of driver monitoring cameras and enable next-generation drunk driver interaction preventing alcohol-related harm., CHI '23: Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems, ISBN:978-1-4503-9421-5

Aim: To develop a non-invasive machine learning (ML) approach to detect hypoglycemia during real car driving based on driving (CAN) , and eye and head motion (EHM) data. Methods: We logged CAN and EHM data in 21 subjects with type 1 diabetes (18 male, 41 ± yrs, A1c 6.8 ± 0.7 % [51 ± 7 mmol/mol]) during driving in eu- (EU) and hypoglycemia (< 3.0 mmol/L, HYPO) . Participants drove in a car (Volkswagen Touran) supervised by a driving instructor on a closed test-track. Using CAN and EHM data, we built ML models to predict the probability of the driver being in HYPO. To make our approach applicable to different generations of cars, we present 3 ML models: first, a model combining CAN+EHM, representing the modern car with integrated camera. Second, a CAN model using driving data only, since modern cars are not generally equipped with EHM tracking. Third, anticipating that autonomous driving will limit the role of CAN data in the future, we tested a model solely based on EHM. Results: Mean BG in EU and HYPO was 6.3 ± 0.8 mmol/L and 2.5 ± 0.5 mmol/L (p< 0.001) , respectively. The model CAN+EHM achieved an area under the receiver operating characteristic curve of 0.88 ± 0.05, sensitivity of 0.70 ± 0.30, and specificity of 0.83 ± 0.in detecting HYPO. Further results are in Fig. 1. Conclusion: We propose ML-based approaches to non-invasively detect HYPO from driver behavior, applicable to contemporary cars and anticipating developments in automotive technology. Disclosure V.Lehmann: None. E.Fleisch: None. T.Kowatsch: Advisory Panel; Pathmate Technologies AG, Research Support; CSS Insurance, Stock/Shareholder; Pathmate Technologies AG. S.N.Lagger: None. M.Laimer: None. F.Wortmann: None. C.Stettler: None. T.Zueger: None. M.Maritsch: None. M.Notter: None. S.Schallmoser: None. C.Bérubé: None. C.Albrecht: None. M.Kraus: None. S.Feuerriegel: None. Funding Swiss National Science Foundation (SNF CRSII5_183569) , Swiss Diabetes Foundation, Diabetes Center Berne, Automobile Club Switzerland (ACS) , Federal Department of Defence, Civil Protection and Sport (DDPS) and Department of Research of the University Hospital Berne

Modern vehicles typically are equipped with assistance systems to support drivers in staying vigilant. To assess the driver state, such systems usually split characteristic vehicle signals into smaller segments which are subsequently fed into algorithms to identify irregularities in driver behavior. In this paper, we compare four different approaches for vehicle signal segmentation to predict driver impairment on a dataset from a drunk driving study (n=31). First, we evaluate two static approaches which segment vehicle signals based on fixed time and distance lengths. Intuitively, such approaches are straightforward to implement and provide segments with a specific frequency. Next, we analyze two dynamic approaches that segment vehicle signals based on pre-defined thresholds and well-defined maneuvers. Although more sophisticated to define, the more specific characteristics of driving situations can potentially improve a driver state prediction model. Finally, we train machine learning models for drunk driving detection on vehicle signals segmented by these four approaches. The maneuver-based approach detects impaired driving with a balanced accuracy of 68.73%, thereby outperforming time-based (67.20%), distance-based (65.66%), and threshold-based (61.53%) approaches in comparable settings. Therefore, our findings indicate that incorporating the driving context benefits the prediction of driver states.

Background: Due to the high rate of asymptomatic patients, the impact of postprandial hypoglycemia after bariatric surgery (PBH) on daily activities such as driving remains controversial. We sought to assess driving performance in patients with PBH. Methods: In a random single-blind crossover study, 10 active drivers with PBH (2 males, age 38±15years, BMI 27.2±4.6kg/m2, gastric bypass 5±2years ago) ingested 75g glucose (GLU) to induce hypoglycemia or aspartame (ASP). A simulator was driven during 10min before and at 20 (D1), 80 (D2), 125 (D3) and 140min (D4) after GLU/ASP, reflective of expected blood glucose (BG) increase (D1), decrease (D2), and hypoglycemia (D3 and D4). Seven driving features (traffic violations and high accelerations) were integrated in a Bayesian hierarchical regression model to assess altered driving performance in GLU vs. ASP. Results: Mean peak and nadir BG after GLU were 182±24 and 47±14mg/dl, respectively. BG was stable after ASP (85±4mg/dl). Driving performance differed between GLU and ASP at times of hypoglycemia (Figure 1), particularly D4, whereas no differences were found for D1 and D2. The posterior probability for impaired driving after GLU was 16% in D1, 82% for D2, 81% for D3 and 98% for D4. Conclusion: Our findings suggest impaired driving performance during PBH. Disclosure A. Tripyla: None. V. Lehmann: None. D. Herzig: None. J. Meier: None. N. Banholzer: None. M. Maritsch: None. S. Feuerriegel: None. F. Wortmann: None. L. Bally: None. Funding Swiss National Science Foundation (PCEGP3_186978, CRSII5_183569, PCEGP1_186932); Dexcom, Inc. (OUS-2020-014)

Aim: To non-invasively detect hypoglycemia in individuals with type 1 diabetes (T1D) based on gaze behavior while driving. Methods: Controlled hypoglycemia was induced in 19 individuals (12 males, age 32 ± 7.1 yrs) with T1D (HbA1c 7.1 ± 0.6% [54 ± 6 mmol/mol]) using an adapted hypoglycemic clamp protocol. Gaze and blood glucose (BG) data were gathered while driving in a simulator during three 18 min sessions: session 1 (BG 90-144 mg/dL), session 2 (BG declining from 72 to 45 mg/dL), and session 3 (BG 36-45 mg/dL). A gradient-boosting machine learning (ML) model was built for hypoglycemia (BG Results: Mean venous BG was 105.4 ± 11.4 mg/dL during session 1, declined from 61.4 ± 6.1 mg/dL to 47.2 ± 8.5 mg/dL during session 2, and was 42.7 ± 4.1 mg/dL during session 3, respectively. Gaze analysis provided 29,968 data samples (1,577.5 ± 52 per subject, 10,041 euglycemia, 19,927 hypoglycemia). Overall, ML achieved an area under the receiver-operating-characteristics curve of 0.83 ± 0.09 for hypoglycemia detection with leave-one-subject-out cross-validation. Conclusion: ML-based gaze analysis shows high accuracy in non-invasive hypoglycemia detection while driving. Our approach offers promising potential in various settings where cameras are available. Disclosure V. Lehmann: None. E. Fleisch: Research Support; Self; CSS Health Insurance, Switzerland, Stock/Shareholder; Self; Pathmate-Technologies AG, Switzerland. F. Wortmann: None. C. Stettler: None. M. Maritsch: None. T. Zueger: None. A. Marxer: None. C. Berube: None. M. Kraus: None. C. Albrecht: None. S. Feuerriegel: None. T. Kowatsch: Advisory Panel; Self; Pathmate Technologies AG, Switzerland and Germany, Stock/Shareholder; Self; Pathmate Technologies AG, Switzerland and Germany. Funding Swiss National Science Foundation (SNF CRSII5_183569); Insel Group; Diabetes Center Berne

CHI EA '20: Extended Abstracts of the 2020 CHI Conference on Human Factors in Computing Systems, ISBN:978-1-4503-6819-3

Background and Objective: Researchers use wearable sensing data and machine learning (ML) models to predict various health and behavioral outcomes. However, sensor data from commercial wearables are prone to noise, missing, or artifacts. Even with the recent interest in deploying commercial wearables for long-term studies, there does not exist a standardized way to process the raw sensor data and researchers often use highly specific functions to preprocess, clean, normalize, and compute features. This leads to a lack of uniformity and reproducibility across different studies, making it difficult to compare results. To overcome these issues, we present FLIRT: A Feature Generation Toolkit for Wearable Data; it is an open-source Python package that focuses on processing physiological data specifically from commercial wearables with all its challenges from data cleaning to feature extraction. Methods: FLIRT leverages a variety of state-of-the-art algorithms (e.g., particle filters, ML-based artifact detection) to ensure a robust preprocessing of physiological data from wearables. In a subsequent step, FLIRT utilizes a sliding-window approach and calculates a feature vector of more than 100 dimensions – a basis for a wide variety of ML algorithms. Results: We evaluated FLIRT on the publicly available WESAD dataset, which focuses on stress detection with an Empatica E4 wearable. Preprocessing the data with FLIRT ensures that unintended noise and artifacts are appropriately filtered. In the classification task, FLIRT outperforms the preprocessing baseline of the original WESAD paper. Conclusion: FLIRT provides functionalities beyond existing packages that can address unmet needs in physiological data processing and feature generation: (a) integrated handling of common wearable file formats (e.g., Empatica E4 archives), (b) robust preprocessing, and (c) standardized feature generation that ensures reproducibility of results. Nevertheless, while FLIRT comes with a default configuration to accommodate most situations, it offers a highly configurable interface for all of its implemented algorithms to account for specific needs., Computer Methods and Programs in Biomedicine, 212, ISSN:0169-2607, ISSN:1872-7565

The reactions of the human body to physical exercise, psychophysiological stress and heart diseases are reflected in heart rate variability (HRV). Thus, continuous monitoring of HRV can contribute to determining and predicting issues in well-being and mental health. HRV can be measured in everyday life by consumer wearable devices such as smartwatches which are easily accessible and affordable. However, they are arguably accurate due to the stability of the sensor. We hypothesize a systematic error which is related to the wearer movement. Our evidence builds upon explanatory and predictive modeling: we find a statistically significant correlation between error in HRV measurements and the wearer movement. We show that this error can be minimized by bringing into context additional available sensor information, such as accelerometer data. This work demonstrates our research-in-progress on how neural learning can minimize the error of such smartwatch HRV measurements., Comment: Adjunct Proceedings of the 2019 ACM International Joint Conference on Pervasive and Ubiquitous Computing and the 2019 International Symposium on Wearable Computers