Estimating Body Mass Index (BMI) from camera images with machine learning models enables rapid weight assessment when traditional methods are unavailable or impractical, such as in telehealth or emergency scenarios. Existing computer vision approaches have been limited to datasets of up to 14,500 images. In this study, we present a deep learning-based BMI estimation method trained on our WayBED dataset, a large proprietary collection of 84,963 smartphone images from 25,353 individuals. We introduce an automatic filtering method that uses posture clustering and person detection to curate the dataset by removing low-quality images, such as those with atypical postures or incomplete views. This process retained 71,322 high-quality images suitable for training. We achieve a Mean Absolute Percentage Error (MAPE) of 7.9% on our hold-out test set (WayBED data) using full-body images, the lowest value in the published literature to the best of our knowledge. Further, we achieve a MAPE of 13% on the completely unseen (during training) VisualBodyToBMI dataset, comparable with state-of-the-art approaches trained on it, demonstrating robust generalization. Lastly, we fine-tune our model on VisualBodyToBMI and achieve a MAPE of 8.56%, the lowest reported value on this dataset so far. We deploy the full pipeline, including image filtering and BMI estimation, on Android devices using the CLAID framework. We release our complete code for model training, filtering, and the CLAID package for mobile deployment as open-source contributions.

Alcohol consumption poses a significant public health challenge, presenting serious risks to individual health and contributing to over 700 daily road fatalities worldwide. Digital interventions can play a crucial role in reducing these risks. However, reliable drunk driving detection systems are vital to effectively deliver these interventions. To develop and evaluate such a system, we conducted an interventional study on a test track to collect real vehicle data from 54 participants. Our system reliably identifies non-sober driving with an area under the receiver operating characteristic curve (AUROC) of 0.84 ± 0.11 and driving above the WHO-recommended blood alcohol concentration limit of 0.05 g/dL with an AUROC of 0.80 ± 0.10. Our models rely on well-known physiological drunk driving patterns. To the best of our knowledge, we are the first to (1) rigorously evaluate the potential of (2) driver monitoring cameras and real-time vehicle data for detecting drunk driving in a (3) real vehicle.

Circadian rhythms govern biological patterns that follow a 24-hour cycle. Dysfunctions in circadian rhythms can contribute to various health problems, such as sleep disorders. Current circadian rhythm assessment methods, often invasive or subjective, limit circadian rhythm monitoring to laboratories. Hence, this study aims to investigate scalable consumer-centric wearables for circadian rhythm monitoring outside traditional laboratories. In a two-week longitudinal study conducted in real-world settings, 36 participants wore an Actigraph, a smartwatch, and a core body temperature sensor to collect activity, temperature, and heart rate data. We evaluated circadian rhythms calculated from commercial wearables by comparing them with circadian rhythm reference measures, i.e., Actigraph activities and chronotype questionnaire scores. The circadian rhythm metric acrophases, determined from commercial wearables using activity, heart rate, and temperature data, significantly correlated with the acrophase derived from Actigraph activities (r = 0.96, r = 0.87, r = 0.79; all p 0.001) and chronotype questionnaire (r = -0.66, r = -0.73, r = -0.61; all p 0.001). The acrophases obtained concurrently from consumer sensors significantly predicted the chronotype ( = 0.64; p 0.001). Our study validates commercial sensors for circadian rhythm assessment, highlighting their potential to support maintaining healthy rhythms and provide scalable and timely health monitoring in real-life scenarios.

Digital health interventions (DHIs) offer promising solutions to the rising global challenges of noncommunicable diseases by promoting behavior change, improving health outcomes, and reducing healthcare costs. However, high churn rates are a concern with DHIs, with many users disengaging before achieving desired outcomes. Churn prediction can help DHI providers identify and retain at-risk users, enhancing the efficacy of DHIs. We analyzed churn prediction models for a weight loss app using various machine learning algorithms on data from 1,283 users and 310,845 event logs. The best-performing model, a random forest model that only used daily login counts, achieved an F1 score of 0.87 on day 7 and identified an average of 93% of churned users during the week-long trial. Notably, higher-dimensional models performed better at low false positive rate thresholds. Our findings suggest that user churn can be forecasted using engagement data, aiding in timely personalized strategies and better health results.

The increasing number of edge devices with enhanced sensing capabilities, such as smartphones, wearables, and IoT devices equipped with sensors, holds the potential for innovative smart-edge applications in healthcare. These devices generate vast amounts of multimodal data, enabling the implementation of digital biomarkers which can be leveraged by machine learning solutions to derive insights, predict health risks, and allow personalized interventions. Training these models requires collecting data from edge devices and aggregating it in the cloud. To validate and verify those models, it is essential to utilize them in real-world scenarios and subject them to testing using data from diverse cohorts. Since some models are too computationally expensive to be run on edge devices directly, a collaborative framework between the edge and cloud becomes necessary. In this paper, we present CLAID, an open-source cross-platform middleware framework based on transparent computing compatible with Android, iOS, WearOS, Linux, macOS, and Windows. CLAID enables logical integration of devices running different operating systems into an edge-cloud system, facilitating communication and offloading between them, with bindings available in different programming languages. We provide Modules for data collection from various sensors as well as for the deployment of machine-learning models. Furthermore, we propose a novel methodology, ML-Model in the Loop for verifying deployed machine learning models, which helps to analyze problems that may occur during the migration of models from cloud to edge devices. We verify our framework in three different experiments and achieve 100% sampling coverage for data collection across different sensors as well as an equal performance of a cough detection model deployed on both Android and iOS devices. Additionally, we compare the memory and battery consumption of our framework across the two mobile operating systems.

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.

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.

Cough, a symptom associated with many prevalent respiratory diseases, can serve as a potential biomarker for diagnosis and disease progression. Consequently, the development of cough monitoring systems and, in particular, automatic cough detection algorithms have been studied since the early 2000s. Recently, there has been an increased focus on the efficiency of such algorithms, as implementation on consumer-centric devices such as smartphones would provide a scalable and affordable solution for monitoring cough with contact-free sensors. Current algorithms, however, are incapable of discerning between coughs of different individuals and, thus, cannot function reliably in situations where potentially multiple individuals have to be monitored in shared environments. Therefore, we propose a weakly supervised metric learning approach for cougher recognition based on smartphone audio recordings of coughs. Our approach involves a triplet network architecture, which employs convolutional neural networks (CNNs). The CNNs of the triplet network learn an embedding function, which maps Mel spectrograms of cough recordings to an embedding space where they are more easily distinguishable. Using audio recordings of nocturnal coughs from asthmatic patients captured with a smartphone, our approach achieved a mean accuracyof 88 % ( ± 10 % SD) on two-way identification tests with 12 enrollment samples and accuracy of 80 % and an equal error rate (EER) of 20 % on verification tests. Furthermore, our approach outperformed human raters with regard to verification tests on average by 8% in accuracy, 4% in false acceptance rate (FAR), and 12% in false rejection rate (FRR). Our code and models are publicly available.

The increasing number of mental disorders worldwide calls for novel types of prevention measures. Given the number of commuters who spend a substantial amount of time on the road, the car offers an opportune environment. This paper presents the first in-vehicle intervention study affecting mental health and well-being on public roads. We designed and implemented two in-vehicle interventions based on proven psychotherapy interventions. Whereas the first intervention uses mindfulness exercises while driving, the second intervention induces positive emotions through music. Ten ordinary and healthy commuters completed 313 of these interventions on their daily drives over two months. We collected drivers’ immediate and post-driving feedback for each intervention and conducted interviews with the drivers after the end of the study. The results show that both interventions have improved drivers’ well-being. While the participants rated the music intervention very positively, the reception of the mindfulness intervention was more ambivalent.

With the emergence of the Internet of People (IoP) and its user-centric applications, novel solutions to the many issues facing today’s societies are to be expected. These problems include unhealthy diets, with obesity and diet-related diseases reaching epidemic proportions. We argue that the proliferation of mixed reality (MR) headsets as next generation primary interfaces provides promising alternatives to contemporary digital solutions in the context of diet tracking and interventions. Concretely, we propose the use of MR headset-mounted cameras for computer vision (CV) based detection of diet-related activities and the consequential display of visual real-time interventions to support healthy food choices. We provide an integrative framework and results from a technical feasibility as well as an impact study conducted in a vending machine (VM) setting. We conclude that current neural networks already enable accurate food item detection in real-world environments. Moreover, our user study suggests that real-time interventions significantly improve beverage (reduction of sugar and energy intake) as well as food choices (reduction of saturated fat). We discuss the results, learnings, and limitations and provide an overview of further technology- and intervention-related avenues of research required by developing an MR-based user support system for healthy food choices.

Even though app store reviews provide highly valuable information on how people use mobile apps and what they expect from them, the systematic, timely analysis of an ever-growing volume of such unstructured reviews across many apps remains a challenge. We analyzed more than 300'000 review sentences belonging to 1'610 finance apps using a machine learning-based approach to investigate the impact that different aspects of finance apps have on their ratings. Additionally, we manually categorized all apps into sub-categories such as payment or trading apps to discuss our findings on an extra level of detail. This work illustrates how different aspects of mobiles apps affect their ratings, how this varies across sub-categories, and discusses the role of privacy, user interfaces, signup experiences, notifications, when the use of location services may be appropriate, and other aspects of mobile finance apps, to provide detailed insights into users' expectations and perception of finance apps.

This paper provides an extended abstract. The full paper is currently under review for Electronic Markets journal.

The personal transport sector constitutes an important target of energy conservation and emission reduction programs. In this context, eco-feedback technologies that provide information on the driving behavior have shown to be an effective means to stimulate changes in driving in favor of both, reduced costs and environmental impact. This study extends the literature on eco-feedback technologies as it demonstrates that a smartphone application can improve fuel efficiency even under conditions where monetary incentives are not given, i.e. where the drivers do not pay for fuel. The field test, which took place with 50 corporate car drivers, demonstrates an improvement in the overall fuel efficiency by 3.23%. The theoretical contribution underlines the assumption that context-related feedback can favorably influence behavior even without direct financial benefits for the agent. Given the large share of corporate cars, findings are also of high practical importance and motivate future research on eco-driving feedback technologies.

When it comes to vehicle accidents, people are stressed out and overstrained, even if it is just a car body damage and no one is hurt. They often lack adequate and immediate assistance and may worry about the lengthy and paper-based loss report to their insurance carrier. At the same time, it is crucial for insurance companies to receive early and detailed case circumstances in order to decrease costs and assist customers with value-added services. Against this background, we propose the usage of mobile phones in order to assist people in the aftermath of an accident. We present a concept for mobile claims assistance along with a proto-typical implementation that features an asynchronous communication between mobile phones and claims management enterprise systems based on mobile Web Services. Finally, we discuss the user perspective on mobile insurance applications and present data we collected using a combination of focus groups and user surveys.

Location Based Services (LBS) promise interesting business opportunities. Today, most LBS are either implemented in hardware devices, or downloaded and installed by mobile phone users as software applications. Both approaches lead to scattered markets and hinder standardization. This paper suggests an alternative approach, which is to enhance mobile web-browsers with location information and implement LBS at the server-side. We define the design space for the location enhanced mobile web and present an implementation of a location enhanced web browser for the currently predominant mobile phone operating system, i.e. Symbian S60.

Reducing their energy consumption has become an important objective for many people. Consumption transparency and timely feedback are essential to support those who want to adjust their behavior in order to conserve energy. In this work, we propose an interactive system that provides instantaneous feedback concerning the energy usage on household and device level. For that, we used and extended the capabilities of a smart electricity meter, built a web-based API to enable interoperability with other applications, and developed a mobile phone interface that allows users to monitor, control, and measure the consumption of single appliances. Our system illustrates a way how usage barriers can be lowered and how high user involvement can be created. By providing users the electricity feedback needed -- in real-time and on device level -- the system allows for identifying the biggest energy guzzlers and helps users decrease their energy consumption.

This paper presents a concept that enables consumers to access and share product recommendations using their mobile phone. Based on a review of current product recommendation mechanisms it devises a concept called APriori. APriori leverages the potential of auto-ID-enabled mobile phones (barcode/RFID) to receive and submit product ratings. Since mobile users cannot be expected to have the patience and time to compose text-based reviews on mobile phones, we introduce a new rating concept that allows users to generate new rating criteria. The concept is tailored to the limited attention and input options of mobile users in real-world environment. This work describes the architecture, implementation, and evaluation of APriori. For an evaluation we have taken the approach of interviewing 26 users in the frames of a formative user study, with the goal to further improve the system for an application in the real world. In addition, the paper discusses open issues regarding community-based product recommendations on mobile phones and proposes solutions.

This contribution is concerned with the business value of Radio Frequency Identification (RFID) technology in retail. We present a case study of an RFID project at Galeria Kaufhof, a subsidiary of Metro Group and one of the largest department store chains in Europe. The project encompasses a variety of RFID applications at the intersection of store logistics and customer service. The contribution that our study makes to the literature is threefold. First, we describe an innovative large-scale trial that goes beyond what was done in earlier projects in several respects. The most fundamental difference from previous trials is the full integration of RFID event data with point-of-sale (POS) and master data, which for the first time offers the retailer the opportunity to directly observe and analyse physical in-store processes. Second, the heterogeneity of RFID applications implemented by Kaufhof allows us to theorise about the effects that RFID may have on business processes from an IT value perspective. We develop a conceptual model to explain the different cause-and-effect chains between RFID investments and their impact on firm performance, the role of complementary and contextual factors, and the difficulty of assessing these impacts using objective performance measures. Third, we compare the case to a prior trial conducted by Kaufhof about 5 years earlier. The differences between the lessons that the company learned in the two projects illustrate the impact of technological advances and standardisation efforts in recent years on managerial perceptions of RFID business value, which allows for the derivation of a number of useful implications for practice.