A Deep Learning-Based Platform for Workers' Stress Detection Using Minimally Intrusive Multisensory Devices

The advent of Industry 4.0 necessitates substantial interaction between humans and machines, presenting new challenges when it comes to evaluating the stress levels of workers who operate in increasingly intricate work environments. Undoubtedly, work-related stress exerts a significant influence on individuals' overall stress levels, leading to enduring health issues and adverse impacts on their quality of life. Although psychological questionnaires have traditionally been employed to assess stress, they lack the capability to monitor stress levels in real-time or on an ongoing basis, thus making it arduous to identify the causes and demanding aspects of work. To surmount this limitation, an effective solution lies in the analysis of physiological signals that can be continuously measured through wearable or ambient sensors. Previous studies in this field have mainly focused on stress assessment through intrusive wearable systems susceptible to noise and artifacts that degrade performance. One of our recently published papers presented a wearable and ambient hardware-software platform that is minimally intrusive, able to detect human stress without hindering normal work activities, and slightly susceptible to artifacts due to movements. A limitation of this system is its not very high performance in terms of the accuracy of detecting multiple stress levels; therefore, in this work, the focus was on improving the software performance of the platform, using a deep learning approach. To this purpose, three neural networks were implemented, and the best performance was achieved by the 1D-convolutional neural network with an accuracy of 95.38% for the identification of two levels of stress, which is a significant improvement over those obtained previously.

Office Workers' Views About the Uses, Concerns, and Acceptance of Hand Hygiene Data Collected From Smart Sanitizers: Exploratory Qualitative Interview Study

BACKGROUND

COVID-19 and the prospect of future pandemics have emphasized the need to reduce disease transmission in workplaces. Despite the well-established link between good hand hygiene (HH) and employee health, HH in nonclinical workplaces has received little attention. Smart sanitizers have been deployed in clinical settings to motivate and enforce HH. This study is part of a large project that explores the potential of smart sanitizers in office settings.

OBJECTIVE

Our previous study found that for office workers to accept the deployment of smart sanitizers, they would need to find the data generated as useful and actionable. The objectives of this study were to identify (1) the potential uses and actions that could be taken from HH data collected by smart sanitizers (2) the concerns of office workers for the identified uses and actions and (3) the circumstances in which office workers accept HH monitoring.

METHODS

An interview study was conducted with 18 office workers from various professions. Interview questions were developed using a framework from personal informatics. Transcripts were analyzed thematically.

RESULTS

A wide range of uses of smart sanitizer data was identified including managing hygiene resources and workflows, finding operating sanitizers, communicating the (high) standard of organizational hygiene, promoting and enforcing organizational hygiene policies, improving workers' own hygiene practices, executing more effective interventions, and identifying the causes of outbreaks. However, hygiene is mostly considered as a private matter, and it is also possible that no action would be taken. Office workers were also concerned about bullying, coercion, and use of hygiene data for unintended purposes. They were also worried that the data could be inaccurate or incomplete, leading to misrepresentation of hygiene practices. Office workers suggested that they would be more likely to accept monitoring in situations where hygiene is considered important, when there are clear benefits to data collection, if their privacy is respected, if they have some control over how their data are collected, and if the ways in which the data will be used are clearly communicated.

CONCLUSIONS

Smart sanitizers could have a valuable role in improving hygiene practices in offices and reducing disease transmission. Many actionable uses for data collected from smart systems were identified. However, office workers consider HH as a personal matter, and acceptance of smart systems is likely to be dynamic and will depend on the broad situation. Except when there are disease outbreaks, smart systems may need to be restricted to uses that do not require the sharing of personal data. Should organizations wish to implement smart sanitizers in offices, it would be advisable to consult widely with staff and develop systems that are customizable and personalizable.

Fiber Bragg Grating Sensor Networks Enhance the In Situ Real-Time Monitoring Capabilities of MLI Thermal Blankets for Space Applications

The utilization of Fiber Bragg Grating (FBG) sensors in innovative optical sensor networks has displayed remarkable potential in providing precise and dependable thermal measurements in hostile environments on Earth. Multi-Layer Insulation (MLI) blankets serve as critical components of spacecraft and are employed to regulate the temperature of sensitive components by reflecting or absorbing thermal radiation. To enable accurate and continuous monitoring of temperature along the length of the insulative barrier without compromising its flexibility and low weight, FBG sensors can be embedded within the thermal blanket, thereby enabling distributed temperature sensing. This capability can aid in optimizing the thermal regulation of the spacecraft and ensuring the reliable and safe operation of vital components. Furthermore, FBG sensors offer sev eral advantages over traditional temperature sensors, including high sensitivity, immunity to electromagnetic interference, and the ability to operate in harsh environments. These properties make FBG sensors an excellent option for thermal blankets in space applications, where precise temperature regulation is crucial for mission success. Nevertheless, the calibration of temperature sensors in vacuum conditions poses a significant challenge due to the lack of an appropriate calibration reference. Therefore, this paper aimed to investigate innovative solutions for calibrating temperature sensors in vacuum conditions. The proposed solutions have the potential to enhance the accuracy and reliability of temperature measurements in space applications, which can enable engineers to develop more resilient and dependable spacecraft systems.

A Development of an IoT-Based Connected University System: Progress Report

In this paper, a report on the development of an Internet of Things (IoT)-based connected university system is presented. There have been multiple smart solutions developed at the university over recent years. However, the user base of these systems is limited. The IoT-based connected university system allows for integration of multiple subsystems without the need to implement all of them in the same environment, thus enabling end-users to access multiple solutions through a single common interface. The implementation is based on microservice architecture, with the focus mainly on system robustness, scalability, and universality. In the system design, four subsystems are currently implemented, i.e., the subsystem for indoor navigation, the subsystem for parking assistants, the subsystem for smart classrooms or offices, and the subsystem for news aggregation from university life. The principles of all implemented subsystems, as well as the implementation of the system as a web interface and a mobile application, are presented in the paper. Moreover, the implementation of the indoor navigation subsystem that uses signals from Bluetooth beacons is described in detail. The paper also presents results proving the concept of the Bluetooth-based indoor navigation, taking into account different placements of nodes. The tests were performed in a real-world environment to evaluate the feasibility of the navigation module that utilizes deterministic fingerprinting algorithms to estimate the positions of users' devices.

Concept Review of a Cloud-Based Smart Battery Management System for Lithium-Ion Batteries: Feasibility, Logistics, and Functionality

Energy storage plays an important role in the adoption of renewable energy to help solve climate change problems. Lithium-ion batteries (LIBs) are an excellent solution for energy storage due to their properties. In order to ensure the safety and efficient operation of LIB systems, battery management systems (BMSs) are required. The current design and functionality of BMSs suffer from a few critical drawbacks including low computational capability and limited data storage. Recently, there has been some effort in researching and developing smart BMSs utilizing the cloud platform. A cloud-based BMS would be able to solve the problems of computational capability and data storage in the current BMSs. It would also lead to more accurate and reliable battery algorithms and allow the development of other complex BMS functions. This study reviews the concept and design of cloud-based smart BMSs and provides some perspectives on their functionality and usability as well as their benefits for future battery applications. The potential division between the local and cloud functions of smart BMSs is also discussed. Cloud-based smart BMSs are expected to improve the reliability and overall performance of LIB systems, contributing to the mass adoption of renewable energy.

Programmable Self-Regulation with Wrinkled Hydrogels and Plasmonic Nanoparticle Lattices

This paper describes a self-regulating system that combines wrinkle-patterned hydrogels with plasmonic nanoparticle (NP) lattices. In the feedback loop, the wrinkle patterns flatten in response to moisture, which then allows light to reach the NP lattice on the bottom layer. Upon light absorption, the NP lattice produces a photothermal effect that dries the hydrogel, and the system then returns to the initial wrinkled configuration. The timescale of this regulatory cycle can be programmed by tuning the degree of photothermal heating by NP size and substrate material. Time-dependent finite element analysis reveals the thermal and mechanical mechanisms of wrinkle formation. This self-regulating system couples morphological, optical, and thermo-mechanical properties of different materials components and offers promising design principles for future smart systems.

Recent Development of Multifunctional Sensors Based on Low-Dimensional Materials

With the demand for accurately recognizing human actions and environmental situations, multifunctional sensors are essential elements for smart applications in various emerging technologies, such as smart robots, human-machine interface, and wearable electronics. Low-dimensional materials provide fertile soil for multifunction-integrated devices. This review focuses on the multifunctional sensors for mechanical stimulus and environmental information, such as strain, pressure, light, temperature, and gas, which are fabricated from low-dimensional materials. The material characteristics, device architecture, transmission mechanisms, and sensing functions are comprehensively and systematically introduced. Besides multiple sensing functions, the integrated potential ability of supplying energy and expressing and storing information are also demonstrated. Some new process technologies and emerging research areas are highlighted. It is presented that optimization of device structures, appropriate material selection for synergy effect, and application of piezotronics and piezo-phototronics are effective approaches for constructing and improving the performance of multifunctional sensors. Finally, the current challenges and direction of future development are proposed.

Wearable Sensors and Systems for Wound Healing-Related pH and Temperature Detection

Wound healing is a complex tissue regeneration process involving many changes in multiple physiological parameters. The pH and temperature of a wound site have long been recognized as important biomarkers for assessing wound healing status. For effective wound management, wound dressings integrated with wearable sensors and systems used for continuous monitoring of pH and temperature have received much attention in recent years. Herein, recent advances in the development of wearable pH and temperature sensors and systems based on different sensing mechanisms for wound status monitoring and treatment are comprehensively summarized. Challenges in the areas of sensing performance, infection identification threshold, large-area 3-dimensional detection, and long-term reliable monitoring in current wearable sensors/systems and emerging solutions are emphasized, providing critical insights into the development of wearable sensors and systems for wound healing monitoring and management.

Application of IoT in Healthcare: Keys to Implementation of the Sustainable Development Goals

We live in complex times in the health, social, political, and energy spheres, and we must be aware of and implement new trends in intelligent social health systems powered by the Internet of Things (IoT). Sustainable development, energy efficiency, and public health are interrelated parameters that can transform a system or an environment for the benefit of people and the planet. The integration of sensors and smart devices should promote energy efficiency and ensure that sustainable development goals are met. This work is carried out according to a mixed approach, with a literature review and an analysis of the impact of the Sustainable Development Goals on the applications of the Internet of Things and smart systems. In the analysis of results, the following questions are answered about these systems and applications: (a) Are IoT applications key to the improvement of people’s health and the environment? (b) Are there research and case studies implemented in cities or territories that demonstrate the effectiveness of IoT applications and their benefits to public health? (c) What sustainable development indicators and objectives can be assessed in the applications and projects analyzed?

An Intelligent Human-Unmanned Aerial Vehicle Interaction Approach in Real Time Based on Machine Learning Using Wearable Gloves

The interactions between humans and unmanned aerial vehicles (UAVs), whose applications are increasing in the civilian field rather than for military purposes, are a popular future research area. Human–UAV interactions are a challenging problem because UAVs move in a three-dimensional space. In this paper, we present an intelligent human–UAV interaction approach in real time based on machine learning using wearable gloves. The proposed approach offers scientific contributions such as a multi-mode command structure, machine-learning-based recognition, task scheduling algorithms, real-time usage, robust and effective use, and high accuracy rates. For this purpose, two wearable smart gloves working in real time were designed. The signal data obtained from the gloves were processed with machine-learning-based methods and classified multi-mode commands were included in the human–UAV interaction process via the interface according to the task scheduling algorithm to facilitate sequential and fast operation. The performance of the proposed approach was verified on a data set created using 25 different hand gestures from 20 different people. In a test using the proposed approach on 49,000 datapoints, process time performance of a few milliseconds was achieved with approximately 98 percent accuracy.

Deep Learning-Based Industry 4.0 and Internet of Things towards Effective Energy Management for Smart Buildings

Worldwide, energy consumption and saving represent the main challenges for all sectors, most importantly in industrial and domestic sectors. The internet of things (IoT) is a new technology that establishes the core of Industry 4.0. The IoT enables the sharing of signals between devices and machines via the internet. Besides, the IoT system enables the utilization of artificial intelligence (AI) techniques to manage and control the signals between different machines based on intelligence decisions. The paper’s innovation is to introduce a deep learning and IoT based approach to control the operation of air conditioners in order to reduce energy consumption. To achieve such an ambitious target, we have proposed a deep learning-based people detection system utilizing the YOLOv3 algorithm to count the number of persons in a specific area. Accordingly, the operation of the air conditioners could be optimally managed in a smart building. Furthermore, the number of persons and the status of the air conditioners are published via the internet to the dashboard of the IoT platform. The proposed system enhances decision making about energy consumption. To affirm the efficacy and effectiveness of the proposed approach, intensive test scenarios are simulated in a specific smart building considering the existence of air conditioners. The simulation results emphasize that the proposed deep learning-based recognition algorithm can accurately detect the number of persons in the specified area, thanks to its ability to model highly non-linear relationships in data. The detection status can also be successfully published on the dashboard of the IoT platform. Another vital application of the proposed promising approach is in the remote management of diverse controllable devices.

Reliable Industry 4.0 Based on Machine Learning and IoT for Analyzing, Monitoring, and Securing Smart Meters

The modern control infrastructure that manages and monitors the communication between the smart machines represents the most effective way to increase the efficiency of the industrial environment, such as smart grids. The cyber-physical systems utilize the embedded software and internet to connect and control the smart machines that are addressed by the internet of things (IoT). These cyber-physical systems are the basis of the fourth industrial revolution which is indexed by industry 4.0. In particular, industry 4.0 relies heavily on the IoT and smart sensors such as smart energy meters. The reliability and security represent the main challenges that face the industry 4.0 implementation. This paper introduces a new infrastructure based on machine learning to analyze and monitor the output data of the smart meters to investigate if this data is real data or fake. The fake data are due to the hacking and the inefficient meters. The industrial environment affects the efficiency of the meters by temperature, humidity, and noise signals. Furthermore, the proposed infrastructure validates the amount of data loss via communication channels and the internet connection. The decision tree is utilized as an effective machine learning algorithm to carry out both regression and classification for the meters’ data. The data monitoring is carried based on the industrial digital twins’ platform. The proposed infrastructure results provide a reliable and effective industrial decision that enhances the investments in industry 4.0.

A Hybrid Semantic Knowledge Integration and Sharing Approach for Distributed Smart Environments

Distributed systems provide smart functionality to everyday objects with the help of wireless sensors using the internet. Since the last decade, the industry is struggling to develop efficient and intelligent protocols to integrate a huge number of smart objects in distributed computing environments. However, the main challenge for smart and distributed system designers lies in the integration of a large number of heterogeneous components for faster, cheaper, and more efficient functionalities. To deal with this issue, practitioners are using edge computing along with server and desktop technology for the development of smart applications by using Service-Oriented Architecture (SOA) where every smart object offers its functionality as a service, enabling other objects to interact with them dynamically. In order to make such a system, researchers have considered context-awareness and Quality of Service (QoS) attributes of device services. However, context modeling is a complicated task since it could include everything around the applications. Moreover, it is also important to consider non-functional interactions that may have an impact on the behavior of the complete system. In this regard, various research dimensions are explored. However, rich context-aware modeling, QoS, user priorities, grouping, and value type direction along with uncertainty are not considered properly while modeling of incomplete or partial domain knowledge during ontology engineering, resulting in low accuracy of results. In this paper, we present a semantic and logic-based formal framework (hybrid) to find the best service among many candidate services by considering the limitations of existing frameworks. Experimental results of the proposed framework show the improvement of the discovered results.

A Low-Cost Visible Light Positioning System for Indoor Positioning

Currently, a high percentage of the world’s population lives in urban areas, and this proportion will increase in the coming decades. In this context, indoor positioning systems (IPSs) have been a topic of great interest for researchers. On the other hand, Visible Light Communication (VLC) systems have advantages over RF technologies; for instance, they do not need satellite signals or the absence of electromagnetic interference to achieve positioning. Nowadays, in the context of Indoor Positioning (IPS), Visible Light Positioning (VLP) systems have become a strong alternative to RF-based systems, allowing the reduction in costs and time to market. This paper shows a low cost VLP solution for indoor systems. This includes multiple programmable beacons and a receiver which can be plugged to a smartphone running a specific app. The position information will be quickly and securely available through the interchange between the receiver and any configurable LED-beacon which is strategically disposed in an area. The implementation is simple, inexpensive, and no direct communication with any data server is required.

SNAPS: Sensor Analytics Point Solutions for Detection and Decision Support Systems

In this review, we discuss the role of sensor analytics point solutions (SNAPS), a reduced complexity machine-assisted decision support tool. We summarize the approaches used for mobile phone-based chemical/biological sensors, including general hardware and software requirements for signal transduction and acquisition. We introduce SNAPS, part of a platform approach to converge sensor data and analytics. The platform is designed to consist of a portfolio of modular tools which may lend itself to dynamic composability by enabling context-specific selection of relevant units, resulting in case-based working modules. SNAPS is an element of this platform where data analytics, statistical characterization and algorithms may be delivered to the data either via embedded systems in devices, or sourced, in near real-time, from mist, fog or cloud computing resources. Convergence of the physical systems with the cyber components paves the path for SNAPS to progress to higher levels of artificial reasoning tools (ART) and emerge as data-informed decision support, as a service for general societal needs. Proof of concept examples of SNAPS are demonstrated both for quantitative data and qualitative data, each operated using a mobile device (smartphone or tablet) for data acquisition and analytics. We discuss the challenges and opportunities for SNAPS, centered around the value to users/stakeholders and the key performance indicators users may find helpful, for these types of machine-assisted tools.

Smart Petri Nets Temperature Control Framework for Reducing Building Energy Consumption

Energy consumption is steadily increasing in the Kingdom of Saudi Arabia (KSA), which imposes continuous strains on the electrical load. Furthermore, consumption rationalization measures do not seem to improve the situation in any way. Therefore, the implementation of energy saving policies become an urgent need. This paper targets developing a smart energy-saving framework for integrating new advanced technologies and conventional Air Conditioning (AC) systems to achieve a comfortable environment, optimum energy efficiency and profitability. In this paper, a three-stage smart control framework, which allows controlling room temperature according to the user’s preferences, is implemented. The first stage is a user identification process. In the second stage, a Petri Nets (PN) model monitors users and sends their preferred temperatures to the third stage. A PID controller is implemented in the third stage to regulate room temperatures. The interconnected sensing and actuating devices in this smart environment are configured to provide users with comfort and energy saving functionality. Experimental results show the good performances and features of the proposed approach. The proposed smart framework reduces the energy consumption of the current ON/OFF controller (219.09 kW) by a significant amount which reaches (116.58 kW) by ratio about 46.79%. Reducing energy consumption is one of these important features in addition to system reactivity and user comfort.

Brain Inspired Dynamic System for the Quality of Service Control over the Long-Haul Nonlinear Fiber-Optic Link

Brain-inspired intelligence using the cognitive dynamic system (CDS) concept is proposed to control the quality-of-service (QoS) over a long-haul fiber-optic link that is nonlinear and with non-Gaussian channel noise. Digital techniques such as digital-back-propagation (DBP) assume that the fiber optic link parameters, such as loss, dispersion, and nonlinear coefficients, are known at the receiver. However, the proposed CDS does not need to know about the fiber optic link physical parameters, and it can improve the bit error rate (BER) or enhance the data rate based on information extracted from the fiber optic link. The information extraction (Bayesian statistical modeling) using intelligent perception processing on the received data, or using the previously extracted models in the model library, is carried out to estimate the transmitted data in the receiver. Then, the BER is sent to the executive through the main feedback channel and the executive produces actions on the physical system/signal to ensure that the BER is continuously under the forward-error-correction (FEC) threshold. Therefore, the proposed CDS is an intelligent and adaptive system that can mitigate disturbance in the fiber optic link (especially in an optical network) using prediction in the perceptor and/or doing proper actions in the executive based on BER and the internal reward. A simplified CDS was implemented for nonlinear fiber optic systems based on orthogonal frequency division multiplexing (OFDM) to show how the proposed CDS can bring noticeable improvement in the system’s performance. As a result, enhancement of the data rate by 12.5% and the Q-factor improvement of 2.74 dB were achieved in comparison to the conventional system (i.e., the system without smart brain).

The Pathway to Intelligence: Using Stimuli-Responsive Materials as Building Blocks for Constructing Smart and Functional Systems

Systems that are intelligent have the ability to sense their surroundings, analyze, and respond accordingly. In nature, many biological systems are considered intelligent (e.g., humans, animals, and cells). For man-made systems, artificial intelligence is achieved by massively sophisticated electronic machines (e.g., computers and robots operated by advanced algorithms). On the other hand, freestanding materials (i.e., not tethered to a power supply) are usually passive and static. Hence, herein, the question is asked: can materials be fabricated so that they are intelligent? One promising approach is to use stimuli-responsive materials; these "smart" materials use the energy supplied by a stimulus available from the surrounding for performing a corresponding action. After decades of research, many interesting stimuli-responsive materials that can sense and perform smart functions have been developed. Classes of functions discussed include practical functions (e.g., targeting and motion), regulatory functions (e.g., self-regulation and amplification), and analytical processing functions (e.g., memory and computing). The pathway toward creating truly intelligent materials can involve incorporating a combination of these different types of functions into a single integrated system by using stimuli-responsive materials as the basic building blocks.

Performing Logical Operations with Stimuli-Responsive Building Blocks

Chemical logic gates can be fabricated by synthesizing molecules that have the ability to detect external stimuli (e.g., temperature or pH) and provide logical outputs. It is, however, challenging to fabricate a system that consists of many logic gates using this method: complex molecules can be difficult to synthesize and these logic gates typically cannot be integrated together. Here, we fabricated different types of logic gates by assembling a combination of different types of stimuli-responsive hydrogels that change their size under the influence of one type of stimulus. Importantly, the preparation of these stimuli-responsive hydrogels is widely reported and technically simple. Through designing the geometry of the systems, we fabricated the YES, NOT, OR, AND, NOR, and NAND gates. Although the hydrogels respond to different types of stimuli, their outputs are the same: a change in size of the hydrogel. Hence, we show that the logic gates can be integrated easily (e.g., by connecting an AND gate to an OR gate). In addition, we fabricated a standalone system with the size of a normal drug tablet (i.e., a "smart tablet") that can analyze (or diagnose) different stimuli and control the release of a chemical (or drug) via the logic gates.

Concurrent recording of RF pulses and gradient fields - comprehensive field monitoring for MRI

Reconstruction of MRI data is based on exact knowledge of all magnetic field dynamics, since the interplay of RF and gradient pulses generates the signal, defines the contrast and forms the basis of resolution in spatial and spectral dimensions. Deviations caused by various sources, such as system imperfections, delays, eddy currents, drifts or externally induced fields, can therefore critically limit the accuracy of MRI examinations. This is true especially at ultra-high fields, because many error terms scale with the main field strength, and higher available SNR renders even smaller errors relevant. Higher baseline field also often requires higher acquisition bandwidths and faster signal encoding, increasing hardware demands and the severity of many types of hardware imperfection. To address field imperfections comprehensively, in this work we propose to expand the concept of magnetic field monitoring to also encompass the recording of RF fields. In this way, all dynamic magnetic fields relevant for spin evolution are covered, including low- to audio-frequency magnetic fields as produced by main magnets, gradients and shim systems, as well as RF pulses generated with single- and multiple-channel transmission systems. The proposed approach permits field measurements concurrently with actual MRI procedures on a strict common time base. The combined measurement is achieved with an array of miniaturized field probes that measure low- to audio-frequency fields via (19) F NMR and simultaneously pick up RF pulses in the MRI system's (1) H transmit band. Field recordings can form the basis of system calibration, retrospective correction of imaging data or closed-loop feedback correction, all of which hold potential to render MRI more robust and relax hardware requirements. The proposed approach is demonstrated for a range of imaging methods performed on a 7 T human MRI system, including accelerated multiple-channel RF pulses. Copyright © 2015 John Wiley & Sons, Ltd.

Developing Global Leaders for Research, Regulation, and Stewardship of Crop Protection Chemistry in the 21st Century

To provide sufficient food and fiber to the increasing global population, the technologies associated with crop protection are growing ever more sophisticated but, at the same time, societal expectations for the safe use of crop protection chemistry tools are also increasing. The goal of this perspective is to highlight the key issues that face future leaders in crop protection, based on presentations made during a symposium titled "Developing Global Leaders for Research, Regulation and Stewardship of Crop Protection Chemistry in the 21st Century", held in conjunction with the IUPAC 13th International Congress of Pesticide Chemistry in San Francisco, CA, USA, during August 2014. The presentations highlighted the fact that leaders in crop protection must have a good basic scientific training and understand new and evolving technologies, are aware of the needs of both developed and developing countries, and have good communication skills. Concern is expressed over the apparent lack of resources to meet these needs, and ideas are put forward to remedy these deficiencies.

Automated personalized feedback for physical activity and dietary behavior change with mobile phones: a randomized controlled trial on adults

BACKGROUND

A dramatic rise in health-tracking apps for mobile phones has occurred recently. Rich user interfaces make manual logging of users' behaviors easier and more pleasant, and sensors make tracking effortless. To date, however, feedback technologies have been limited to providing overall statistics, attractive visualization of tracked data, or simple tailoring based on age, gender, and overall calorie or activity information. There are a lack of systems that can perform automated translation of behavioral data into specific actionable suggestions that promote healthier lifestyle without any human involvement.

OBJECTIVE

MyBehavior, a mobile phone app, was designed to process tracked physical activity and eating behavior data in order to provide personalized, actionable, low-effort suggestions that are contextualized to the user's environment and previous behavior. This study investigated the technical feasibility of implementing an automated feedback system, the impact of the suggestions on user physical activity and eating behavior, and user perceptions of the automatically generated suggestions.

METHODS

MyBehavior was designed to (1) use a combination of automatic and manual logging to track physical activity (eg, walking, running, gym), user location, and food, (2) automatically analyze activity and food logs to identify frequent and nonfrequent behaviors, and (3) use a standard machine-learning, decision-making algorithm, called multi-armed bandit (MAB), to generate personalized suggestions that ask users to either continue, avoid, or make small changes to existing behaviors to help users reach behavioral goals. We enrolled 17 participants, all motivated to self-monitor and improve their fitness, in a pilot study of MyBehavior. In a randomized two-group trial, investigators randomly assigned participants to receive either MyBehavior's personalized suggestions (n=9) or nonpersonalized suggestions (n=8), created by professionals, from a mobile phone app over 3 weeks. Daily activity level and dietary intake was monitored from logged data. At the end of the study, an in-person survey was conducted that asked users to subjectively rate their intention to follow MyBehavior suggestions.

RESULTS

In qualitative daily diary, interview, and survey data, users reported MyBehavior suggestions to be highly actionable and stated that they intended to follow the suggestions. MyBehavior users walked significantly more than the control group over the 3 weeks of the study (P=.05). Although some MyBehavior users chose lower-calorie foods, the between-group difference was not significant (P=.15). In a poststudy survey, users rated MyBehavior's personalized suggestions more positively than the nonpersonalized, generic suggestions created by professionals (P<.001).

CONCLUSIONS

MyBehavior is a simple-to-use mobile phone app with preliminary evidence of efficacy. To the best of our knowledge, MyBehavior represents the first attempt to create personalized, contextualized, actionable suggestions automatically from self-tracked information (ie, manual food logging and automatic tracking of activity). Lessons learned about the difficulty of manual logging and usability concerns, as well as future directions, are discussed.

TRIAL REGISTRATION

ClinicalTrials.gov NCT02359981; https://clinicaltrials.gov/ct2/show/NCT02359981 (Archived by WebCite at http://www.webcitation.org/6YCeoN8nv).