Analysis of respiratory sounds, such as coughing and breathing, has emerged as a promising non-invasive approach for the early detection of pulmonary conditions, including COVID-19 and chronic obstructive pulmonary disease (COPD), as well as for managing treatment plans. In this work, we explore audio-based classification of respiratory conditions in terms of breathing patterns using the few-shot learning approach that requires only a few samples to develop models. We experimented with three publicly available datasets of audio recordings of breathing patterns of healthy people and patients with COVID-19 or COPD. Through a detailed evaluation using three types of audio features commonly employed for audio event classification, with varying embedding dimensions and shot numbers, we found that Mel-Frequency Cepstral Coefficients (MFCCs) with 20 embedding dimensions can achieve an average accuracy of around 85% using only 10 shots when classifying the breathing patterns obtained from the three datasets. These findings highlight the potential for developing audio-based screening tools that require only a few samples, which can be utilized for public health diagnostics.

In the healthcare industry, researchers have been developing machine learning models to automate diagnosing patients with respiratory illnesses based on their breathing patterns. However, these models do not consider the demographic biases, particularly sex bias, that often occur when models are trained with a skewed patient dataset. Hence, it is essential in such an important industry to reduce this bias so that models can make fair diagnoses. In this work, we examine the bias in models used to detect breathing patterns of two major respiratory diseases, i.e., chronic obstructive pulmonary disease (COPD) and COVID-19. Using decision tree models trained with audio recordings of breathing patterns obtained from two open-source datasets consisting of 29 COPD and 680 COVID-19-positive patients, we analyze the effect of sex bias on the models. With a threshold optimizer and two constraints (demographic parity and equalized odds) to mitigate the bias, we witness 81.43% (demographic parity difference) and 71.81 % (equalized odds difference) improvements. These findings are statistically significant.

While researchers have been trying to understand the stress and fatigue among pilots, especially pilot trainees, and to develop stress/fatigue models to automate the process of detecting stress/fatigue, they often do not consider biases such as sex in those models. However, in a critical profession like aviation, where the demographic distribution is disproportionately skewed to one sex, it is urgent to mitigate biases for fair and safe model predictions. In this work, we investigate the perceived stress/fatigue of 69 college students, including 40 pilot trainees with around 63% male. We construct models with decision trees first without bias mitigation and then with bias mitigation using a threshold optimizer with demographic parity and equalized odds constraints 30 times with random instances. Using bias mitigation, we achieve improvements of 88.31% (demographic parity difference) and 54.26% (equalized odds difference), which are also found to be statistically significant.

The introduction of the Internet of Things (IoT) has made several emerging applications, from financial transactions to property access, possible through IoT-connected smart wearables (smartwatches). This creates an immediate need for an authentication system that can validate a user seamlessly, compared to knowledge-based approaches. In this work, we present an implicit authentication system that utilizes a bag of on-phone artificial neural network (ANN) models to validate a user based on the availability of three soft-biometrics (heart rate, gait, and breathing patterns) collected from smartphones and Fitbits. We find that using all three biometrics we can achieve an average accuracy of up to$.973 \pm . 004$. Next, we implement the bag of models on smartphones using Google's TensorFlow Lite framework-supportedTFL Authapplication, which requires around 56-65 KB memory and can verify a user in 5 seconds. Finally, we evaluate the systemTFL Authusing two cohorts of 25 subjects in total, and we find that the system has average understandability and importance scores of around 4.0 and 4.3 on a 1 – 5 scale.

Respiratory diseases, such as chronic obstructive pulmonary disease (COPD) and asthma, are two major reasons for people's death across the globe. In addition to these common inflammatory respiratory diseases, some human transmissible respiratory diseases, such as coronaviruses, cause a global pandemic. One major symptom of these inflammatory respiratory diseases is coughing. Identifying coughing using smartphone-microphone recordings is easily doable from a remote setup and can help physicians and researchers early guess a situation for an individual and a community. However, smartphone-microphone recordings can be affected by environmental noises and that can impact the performance of models that are developed to detect coughing from microphone recording. Thereby, in this work, we present a detailed analysis of noise impacts on cough detection models. We develop models using voluntary coughs and other background sounds obtained from three public datasets and test the performance of those models while detecting various types of coughs, including COPD and COVID-19, obtain from three separate datasets in the presence of background noises.

When high school students leave their homes for a college education, they often face enormous changes and challenges in life, such as meeting new people, more responsibilities in life, and being away from family and their comfort zones. These sudden changes often lead to an elevation of stress and anxiety, affecting a student’s health and well-being. Situations can even get worse in the age of global pandemics, such as COVID-19, when regular life and social activities are significantly disrupted due to lockdown or stay-at-home orders. Therefore, predicting phone call patterns (a measure of social engagement) based on various factors and activities of a person can be helpful to foster social engagement and promote health and well-being during sudden lifestyle changes. In this work, we investigate a cohort of 370 on-campus college students over three consecutive semesters and breaks between them to find various geo-temporal factors and activities that affect students’ phone call behaviors and develop models that can predict the next call duration with a correlation of up to 0.89 between the actual and predicted duration using individual-level generalized linear models. Findings from this work can further be extended to other populations, and thereby, our findings will enable the design and delivery of new smartphone-based health interventions (guided feedback) to help people to adapt and cope up with situations that affect their lifestyle and social activities.

Modern smartphones and wearables are able to continuously collect significant amounts of sensor data, where such data can be helpful to study a user's mobility or social interaction patterns, but also to deliver various services based on a user's presence at different places during certain times of the day. Therefore, it is important to accurately identify personal places of interest (POIs), such as a user's workplace or home. Such places are usually determined using segmentation of location traces, but frequent gaps in the data (i.e., missing location readings) can result in a large number of small and incomplete segments that should actually be grouped together into a single large segment. This paper presents a segmentation approach that utilizes a user's personal data obtained from multiple sensor sources and devices such as the battery recharge behavior (measured on smartphones), step counts, and sleep patterns (measured by wearables), to opportunistically fill gaps in the user's location traces. Using the data from a mobile crowd sensing study of more than 450 users over a 2-year period, we show that our approach is able to generate fewer, but more complete segments compared to the state of the art.

As the economy progresses and new technologies emerge, more people are struggling with sleep-related difficulties. Poor sleep quality adversely affects people's health and well-being, productivity, academic success, and cognitive capability. These impairments can also affect traffic and industrial safety, and national economic developments. To better tackle these problems, it is important to accurately understand people's sleep quality. In this work, we present approaches to accurately estimate a user's sleep duration, which will facilitate better estimation of sleep quality. We apply generalized linear model (GLM) and generalized linear mixed model (GLMM), which takes person variability into consideration in addition to fixed effects, such as various temporal factors (sleep start time, days of a week, etc.), weather, a user's daily activities and calendar entries to estimate sleep duration. Through our analysis of a longitudinal sensor dataset collected from the smartphones and Fitbits of a cohort of 18 on-campus college students over an extended period of time, we show the feasibility of the work with correlations of up to 0.745 between the pairs of actual and estimated sleep durations.

Non-speech human sounds, such as coughs and snores, and their patterns are associated with different respiratory diseases, including asthma, chronic obstructive pulmonary disease (COPD), as well as other health difficulties such as sleep disorders. Thereby, researchers and physicians have been using coughs and snores as symptoms while reporting and assessing respiratory diseases, their stages, and sleep quality. However, so far, the assessments frequently depend on different types of patient-reported surveys, which inherently suffer from various limitations, such as recall biases, human errors. Therefore, automated detection and reporting of coughs and snores can improve the disease assessment and monitoring. In this paper, we present an automated approach to detect coughs and snores from smartphone-microphones using generalized, semi-personalized and personalized modeling schemes. We analyze three separate datasets and different combinations of three types of nocturnal noises (i.e., sounds from air conditioners (AC), dog barks, and sirens) using the Mel-frequency cepstral coefficient (MFCC) features and different classification techniques. We find that a generalized model with the support vector machine (SVM) classifier can achieve an average accuracy of 0.86 ± 0.14, F1 score of 0.86± 0.13, and area under the receiver operating characteristic curve (AUC-ROC) of 0.94 ± 0.08. These performances can further be improved to an average accuracy of 0.96± 0.08, F1 score of 0.96± 0.08, and AUC-ROC of 0.98 ± 0.04 using the personalized random forest (RF) model. The results show the potential for smartphones to automatically report symptoms of respiratory diseases as well as sleep disorders. Furthermore, we find that our models perform consistently well while testing on separate datasets in the presence of multiple background-noises.

As market wearables are becoming popular with a range of services, including making financial transactions, accessing cars, etc. that they provide based on various private information of a user, security of this information is becoming very important. However, users are often flooded with PINs and passwords in this internet of things (IoT) world. Additionally, hard-biometric, such as facial or finger recognition, based authentications are not adaptable for market wearables due to their limited sensing and computation capabilities. Therefore, it is a time demand to develop a burden-free implicit authentication mechanism for wearables using the less-informative soft-biometric data that are easily obtainable from the market wearables. In this work, we present a context-dependent soft-biometric-based wearable authentication system utilizing the heart rate, gait, and breathing audio signals. From our detailed analysis, we find that a binary support vector machine (SVM) with radial basis function (RBF) kernel can achieve an average accuracy of 0.94 ± 0.07, F1score of 0.93 ± 0.08, an equal error rate (EER) of about 0.06 at a lower confidence threshold of 0.52, which shows the promise of this work.

The Internet of Things (IoT) is increasingly empowering people with an interconnected world of physical objects ranging from smart buildings to portable smart devices, such as wearables. With recent advances in mobile sensing, wearables have become a rich collection of portable sensors and are able to provide various types of services, including tracking of health and fitness, making financial transactions, and unlocking smart locks and vehicles. Most of these services are delivered based on users' confidential and personal data, which are stored on these wearables. Existing explicit authentication approaches (i.e., PINs or pattern locks) for wearables suffer from several limitations, including small or no displays, risk of shoulder surfing, and users' recall burden. Oftentimes, users completely disable security features out of convenience. Therefore, there is a need for a burden-free (implicit) authentication mechanism for wearable device users based on easily obtainable biometric data. In this paper, we present an implicit wearable device user authentication mechanism using combinations of three types of coarse-grain minute-level biometrics: behavioral (step counts), physiological (heart rate), and hybrid (calorie burn and metabolic equivalent of task). From our analysis of over 400 Fitbit users from a 17-month long health study, we are able to authenticate subjects with average accuracy values of around .93 (sedentary) and .90 (non-sedentary) with equal error rates of .05 using binary SVM classifiers. Our findings also show that the hybrid biometrics perform better than other biometrics and behavioral biometrics do not have a significant impact, even during non-sedentary periods.

As the economy progresses and new technology emerges, more people are struggling with sleep-related difficulties. Researchers have previously identified that smartphone use may be associated with poor sleep quality. However, smartphones have become an indispensable part of modern life. Therefore, it is important to investigate the potential impacts of smartphone use patterns on an individual's health. In this paper, we investigate sleep quality variations between two sets of pre-sleep phone use patterns: phone use before bed-time and phone use during bed-time (before sleep). Our analysis, based on a multi-year mobile crowdsensed data collection effort on more than 400 college students, shows significant sleep quality variations when a phone is used in either of these two usage patterns compared to when it is not used. However, the results also show that phone use during bed-time leads to a significantly worse sleep quality. We expect that these findings will be useful for individuals, public authorities, and smartphone developers to improve smartphone users' sleep quality.

It is becoming increasingly important to accurately detect a user's presence at certain locations during certain times of the day, e.g., to study the user's patterns with respect to mobility, behavior, or social interactions and to enable the delivery of targeted services. However, instead of geographic locations, it is often more important to determine a locale that is relevant to the user, e.g., the place of work, home, homes of family and friends, social gathering places, etc. These significant personal places can be determined through analysis, e.g., via segmentation of location traces into a discrete sequence of places. However, segmentation of traces with many gaps (e.g., due to loss of network connectivity or GPS signal) results in a large number of small segments, where many of these segments actually belong together. This work proposes a novel segmentation approach that opportunistically fills gaps in a user's location trace by borrowing location data from other co-located users utilizing the power of mobile crowd sensing and computing (MCSC) paradigm. Through our analysis of four separate large-scale crowd sensing study datasets, we show that our approach yields more and larger segments than the state-of-the-art, where each segment accurately represents the presence of a user at a significant personal place.

Wearables, such as Fitbit, Apple Watch, and Microsoft Band, with their rich collection of sensors, facilitate the tracking of healthcare- and wellness-related metrics. However, the assessment of the physiological metrics collected by these devices could also be useful in identifying the user of the wearable, e.g., to detect unauthorized use or to correctly associate the data to a user if wearables are shared among multiple users. Further, researchers and healthcare providers often rely on these smart wearables to monitor research subjects and patients in their natural environments over extended periods of time. Here, it is important to associate the sensed data with the corresponding user and to detect if a device is being used by an unauthorized individual, to ensure study compliance. Existing one-time authentication approaches using credentials (e.g., passwords, certificates) or trait-based biometrics (e.g., face, fingerprints, iris, voice) might fail, since such credentials can easily be shared among users. In this paper, we present a continuous and reliable wearable-user authentication mechanism using coarse-grain minute-level physical activity (step counts) and physiological data (heart rate, calorie burn, and metabolic equivalent of task). From our analysis of 421 Fitbit users from a two-year long health study, we are able to statistically distinguish nearly 100% of the subject-pairs and to identify subjects with an average accuracy of 92.97%.

It is becoming increasingly important to accurately detect a user's presence at certain locations during certain times of the day, e.g., to study the user's patterns with respect to mobility, behavior, or social interactions and to enable the delivery of targeted services. However, instead of geographic locations, it is often more important to determine a locale that is important to the user, e.g., the place of work, home, homes of family and friends, social gathering places, etc. These significant personal places can be determined through analysis, e.g., via segmentation of location traces into a discrete sequence of places. However, segmentation of traces with many gaps (e.g., due to loss of network or GPS signal) results in a large number of small segments, where many of these segments actually belong together. This work proposes a new segmentation approach that opportunistically fills gaps in location traces with the help of data from other (co-located) users. Using data from 195 users, collected over a 2-year period, we show that this approach yields fewer and larger segments, where each segment accurately represents the presence of a user at a significant personal place.

We investigate needs, challenges, and opportunities in visualizing time-series sensor data on stress to inform the design of just-in-time adaptive interventions (JITAIs). We identify seven key challenges: massive volume and variety of data, complexity in identifying stressors, scalability of space, multifaceted relationship between stress and time, a need for representation at multiple granularities, interperson variability, and limited understanding of JITAI design requirements due to its novelty. We propose four new visualizations based on one million minutes of sensor data (n=70). We evaluate our visualizations with stress researchers (n=6) to gain first insights into its usability and usefulness in JITAI design. Our results indicate that spatio-temporal visualizations help identify and explain between- and within-person variability in stress patterns and contextual visualizations enable decisions regarding the timing, content, and modality of intervention. Interestingly, a granular representation is considered informative but noise-prone; an abstract representation is the preferred starting point for designing JITAIs.

Driving is known to be a daily stressor. Measurement of driver's stress in real-time can enable better stress management by increasing self-awareness. Recent advances in sensing technology has made it feasible to continuously assess driver's stress in real-time, but it requires equipping the driver with these sensors and/or instrumenting the car. In this paper, we present "GStress", a model to estimate driver's stress using only smartphone GPS traces. The GStress model is developed and evaluated from data collected in a mobile health user study where 10 participants wore physiological sensors for 7 days ( for an average of 10.45 hours/day) in their natural environment. Each participant engaged in 10 or more driving episodes, resulting in a total of 37 hours of driving data. We find that major driving events such as stops, turns, and braking increase stress of the driver. We quantify their impact on stress and thus construct our GStress model by training a Generalized Linear Mixed Model (GLMM) on our data. We evaluate the applicability of GStress in predicting stress from GPS traces, and obtain a correlation of 0.72. By obviating any burden on the driver or the car, we believe, GStress can make driver's stress assessment ubiquitous.

In a wireless ad hoc network, the main issue of a good broadcast protocol is to attain maximum reachability with minimal packet forwarding. Existing protocols address this issue by utilizing the knowledge of up to 2-hop neighbors to approximate an MCDS (minimum connected dominating set) via heuristics derived from techniques known as Self pruning and Dominant pruning. Our experiments show that, using these greedy choice heuristics result in a biased load distribution throughout the network. Some nodes become heavily loaded and consequently packets through those nodes, whether unicast or broadcast, experience significantly larger delay. Contention and collision also increase at some regions, while they are relatively low at other regions. In this paper we address these issues, and propose various methods to evenly distribute the load caused by broadcast packets. Our algorithms take various reactive measures to dynamically include less loaded nodes in the forward list, while maintaining total number of packet forwards low. Detailed simulation using ns-2 shows fair scheduling of resources and significant improvement in distribution of packet forwarding load, packet delay, latency and overall performance.