-
Transfer Learning to Detect COVID-19 Coughs with Incremental Addition of Patient Coughs to Healthy People's Cough Detection Models
Authors:
Sudip Vhaduri,
Seungyeon Paik,
Jessica E Huber
Abstract:
Millions of people have died worldwide from COVID-19. In addition to its high death toll, COVID-19 has led to unbearable suffering for individuals and a huge global burden to the healthcare sector. Therefore, researchers have been trying to develop tools to detect symptoms of this human-transmissible disease remotely to control its rapid spread. Coughing is one of the common symptoms that research…
▽ More
Millions of people have died worldwide from COVID-19. In addition to its high death toll, COVID-19 has led to unbearable suffering for individuals and a huge global burden to the healthcare sector. Therefore, researchers have been trying to develop tools to detect symptoms of this human-transmissible disease remotely to control its rapid spread. Coughing is one of the common symptoms that researchers have been trying to detect objectively from smartphone microphone-sensing. While most of the approaches to detect and track cough symptoms rely on machine learning models developed from a large amount of patient data, this is not possible at the early stage of an outbreak. In this work, we present an incremental transfer learning approach that leverages the relationship between healthy peoples' coughs and COVID-19 patients' coughs to detect COVID-19 coughs with reasonable accuracy using a pre-trained healthy cough detection model and a relatively small set of patient coughs, reducing the need for large patient dataset to train the model. This type of model can be a game changer in detecting the onset of a novel respiratory virus.
△ Less
Submitted 11 November, 2023;
originally announced November 2023.
-
Discovering COVID-19 Coughing and Breathing Patterns from Unlabeled Data Using Contrastive Learning with Varying Pre-Training Domains
Authors:
**** Cai,
Sudip Vhaduri,
Xiao Luo
Abstract:
Rapid discovery of new diseases, such as COVID-19 can enable a timely epidemic response, preventing the large-scale spread and protecting public health. However, limited research efforts have been taken on this problem. In this paper, we propose a contrastive learning-based modeling approach for COVID-19 coughing and breathing pattern discovery from non-COVID coughs. To validate our models, extens…
▽ More
Rapid discovery of new diseases, such as COVID-19 can enable a timely epidemic response, preventing the large-scale spread and protecting public health. However, limited research efforts have been taken on this problem. In this paper, we propose a contrastive learning-based modeling approach for COVID-19 coughing and breathing pattern discovery from non-COVID coughs. To validate our models, extensive experiments have been conducted using four large audio datasets and one image dataset. We further explore the effects of different factors, such as domain relevance and augmentation order on the pre-trained models. Our results show that the proposed model can effectively distinguish COVID-19 coughing and breathing from unlabeled data and labeled non-COVID coughs with an accuracy of up to 0.81 and 0.86, respectively. Findings from this work will guide future research to detect an outbreak of a new disease early.
△ Less
Submitted 2 June, 2023;
originally announced June 2023.
-
Predicting dominant hand from spatiotemporal context varying physiological data
Authors:
Jorge Neira-Garcia,
Sudip Vhaduri
Abstract:
Health metrics from wrist-worn devices demand an automatic dominant hand prediction to keep an accurate operation. The prediction would improve reliability, enhance the consumer experience, and encourage further development of healthcare applications. This paper aims to evaluate the use of physiological and spatiotemporal context information from a two-hand experiment to predict the wrist placemen…
▽ More
Health metrics from wrist-worn devices demand an automatic dominant hand prediction to keep an accurate operation. The prediction would improve reliability, enhance the consumer experience, and encourage further development of healthcare applications. This paper aims to evaluate the use of physiological and spatiotemporal context information from a two-hand experiment to predict the wrist placement of a commercial smartwatch. The main contribution is a methodology to obtain an effective model and features from low sample rate physiological sensors and a self-reported context survey. Results show an effective dominant hand prediction using data from a single subject under real-life conditions.
△ Less
Submitted 8 December, 2022;
originally announced December 2022.
-
Automatic Anomalies Detection in Hydraulic Devices
Authors:
Jose A. Solorio,
Jose M. Garcia,
Sudip Vhaduri
Abstract:
Nowadays, the applications of hydraulic systems are present in a wide variety of devices in both industrial and everyday environments. The implementation and usage of hydraulic systems have been well documented; however, today, this still faces a challenge, the integration of tools that allow more accurate information about the functioning and operation of these systems for proactive decision-maki…
▽ More
Nowadays, the applications of hydraulic systems are present in a wide variety of devices in both industrial and everyday environments. The implementation and usage of hydraulic systems have been well documented; however, today, this still faces a challenge, the integration of tools that allow more accurate information about the functioning and operation of these systems for proactive decision-making. In industrial applications, many sensors and methods exist to measure and determine the status of process variables (e.g., flow, pressure, force). Nevertheless, little has been done to have systems that can provide users with device-health information related to hydraulic devices integrated into the machinery. Implementing artificial intelligence (AI) technologies and machine learning (ML) models in hydraulic system components has been identified as a solution to the challenge many industries currently face: optimizing processes and carrying them out more safely and efficiently. This paper presents a solution for the characterization and estimation of anomalies in one of the most versatile and used devices in hydraulic systems, cylinders. AI and ML models were implemented to determine the current operating status of these hydraulic components and whether they are working correctly or if a failure mode or abnormal condition is present.
△ Less
Submitted 5 December, 2022;
originally announced December 2022.
-
Opportunistic Multi-Modal User Authentication for Health-Tracking IoT Wearables
Authors:
Alexa Muratyan,
William Cheung,
Sayanton V. Dibbo,
Sudip Vhaduri
Abstract:
With the advancement of technologies, market wearables are becoming increasingly popular with a range of services, including providing access to bank accounts, accessing cars, monitoring patients remotely, among several others. However, often these wearables collect various sensitive personal information of a user with no to limited authentication, e.g., knowledge-based external authentication tec…
▽ More
With the advancement of technologies, market wearables are becoming increasingly popular with a range of services, including providing access to bank accounts, accessing cars, monitoring patients remotely, among several others. However, often these wearables collect various sensitive personal information of a user with no to limited authentication, e.g., knowledge-based external authentication techniques, such as PINs. While most of these external authentication techniques suffer from multiple limitations, including recall burden, human errors, or biases, researchers have started using various physiological and behavioral data, such as gait and heart rate, collected by the wearables to authenticate a wearable user implicitly with a limited accuracy due to sensing and computing constraints of wearables. In this work, we explore the usefulness of blood oxygen saturation SpO2 values collected from the Oximeter device to distinguish a user from others. From a cohort of 25 subjects, we find that 92% of the cases SpO2 can distinguish pairs of users. From detailed modeling and performance analysis, we observe that while SpO2 alone can obtain an average accuracy of 0.69 and F1 score of 0.69, the addition of heart rate (HR) can improve the average identification accuracy by 15% and F1 score by 13%. These results show promise in using SpO2 along with other biometrics to develop implicit continuous authentications for wearables.
△ Less
Submitted 9 October, 2021; v1 submitted 28 September, 2021;
originally announced September 2021.
-
Understanding College Students' Phone Call Behaviors Towards a Sustainable Mobile Health and Wellbeing Solution
Authors:
Yugyeong Kim,
Sudip Vhaduri,
Christian Poellabauer
Abstract:
During the transition from high school to on-campus college life, a student leaves home and starts facing enormous life changes, including meeting new people, more responsibilities, being away from family, and academic challenges. These recent changes lead to an elevation of stress and anxiety, affecting a student's health and wellbeing. With the help of smartphones and their rich collection of se…
▽ More
During the transition from high school to on-campus college life, a student leaves home and starts facing enormous life changes, including meeting new people, more responsibilities, being away from family, and academic challenges. These recent changes lead to an elevation of stress and anxiety, affecting a student's health and wellbeing. With the help of smartphones and their rich collection of sensors, we can continuously monitor various factors that affect students' behavioral patterns, such as communication behaviors associated with their health, wellbeing, and academic success. In this work, we try to assess college students' communication patterns (in terms of phone call duration and frequency) that vary across various geographical contexts (e.g., dormitories, classes, dining) during different times (e.g., epochs of a day, days of a week) using visualization techniques. Findings from this work will help foster the design and delivery of smartphone-based health interventions; thereby, help the students adapt to the changes in life.
△ Less
Submitted 11 November, 2020;
originally announced November 2020.
-
Resource Management Schemes for Cloud-Native Platforms with Computing Containers of Docker and Kubernetes
Authors:
Ying Mao,
Yuqi Fu,
Suwen Gu,
Sudip Vhaduri,
Long Cheng,
Qingzhi Liu
Abstract:
Businesses have made increasing adoption and incorporation of cloud technology into internal processes in the last decade. The cloud-based deployment provides on-demand availability without active management. More recently, the concept of cloud-native application has been proposed and represents an invaluable step toward hel** organizations develop software faster and update it more frequently t…
▽ More
Businesses have made increasing adoption and incorporation of cloud technology into internal processes in the last decade. The cloud-based deployment provides on-demand availability without active management. More recently, the concept of cloud-native application has been proposed and represents an invaluable step toward hel** organizations develop software faster and update it more frequently to achieve dramatic business outcomes. Cloud-native is an approach to build and run applications that exploit the cloud computing delivery model's advantages. It is more about how applications are created and deployed than where. The container-based virtualization technology, such as Docker and Kubernetes, serves as the foundation for cloud-native applications. This paper investigates the performance of two popular computational-intensive applications, big data, and deep learning, in a cloud-native environment. We analyze the system overhead and resource usage for these applications. Through extensive experiments, we show that the completion time reduces by up to 79.4% by changing the default setting and increases by up to 96.7% due to different resource management schemes on two platforms. Additionally, the resource release is delayed by up to 116.7% across different systems. Our work can guide developers, administrators, and researchers to better design and deploy their applications by selecting and configuring a hosting platform.
△ Less
Submitted 20 October, 2020;
originally announced October 2020.
-
Context-Dependent Implicit Authentication for Wearable Device User
Authors:
William Cheung,
Sudip Vhaduri
Abstract:
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 fin…
▽ More
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 $0.94 \pm 0.07$, $F_1$ score of $0.93 \pm 0.08$, an equal error rate (EER) of about $0.06$ at a lower confidence threshold of 0.52, which shows the promise of this work.
△ Less
Submitted 25 August, 2020;
originally announced August 2020.
-
Continuous Authentication of Wearable Device Users from Heart Rate, Gait, and Breathing Data
Authors:
William Cheung,
Sudip Vhaduri
Abstract:
The security of private information is becoming the bedrock of an increasingly digitized society. While the users are flooded with passwords and PINs, these gold-standard explicit authentications are becoming less popular and valuable. Recent biometric-based authentication methods, such as facial or finger recognition, are getting popular due to their higher accuracy. However, these hard-biometric…
▽ More
The security of private information is becoming the bedrock of an increasingly digitized society. While the users are flooded with passwords and PINs, these gold-standard explicit authentications are becoming less popular and valuable. Recent biometric-based authentication methods, such as facial or finger recognition, are getting popular due to their higher accuracy. However, these hard-biometric-based systems require dedicated devices with powerful sensors and authentication models, which are often limited to most of the market wearables. Still, market wearables are collecting various private information of a user and are becoming an integral part of life: accessing cars, bank accounts, etc. Therefore, time demands a burden-free implicit authentication mechanism for wearables using the less-informative soft-biometric data that are easily obtainable from modern market wearables. In this work, we present a context-dependent soft-biometric-based authentication system for wearables devices using heart rate, gait, and breathing audio signals. From our detailed analysis using the "leave-one-out" validation, we find that a lighter $k$-Nearest Neighbor ($k$-NN) model with $k = 2$ can obtain an average accuracy of $0.93 \pm 0.06$, $F_1$ score $0.93 \pm 0.03$, and {\em false positive rate} (FPR) below $0.08$ at 50\% level of confidence, which shows the promise of this work.
△ Less
Submitted 24 August, 2020;
originally announced August 2020.
-
Adherence to Personal Health Devices: A Case Study in Diabetes Management
Authors:
Sudip Vhaduri,
Temiloluwa Prioleau
Abstract:
Personal health devices can enable continuous monitoring of health parameters. However, the benefit of these devices is often directly related to the frequency of use. Therefore, adherence to personal health devices is critical. This paper takes a data mining approach to study continuous glucose monitor use in diabetes management. We evaluate two independent datasets from a total of 44 subjects fo…
▽ More
Personal health devices can enable continuous monitoring of health parameters. However, the benefit of these devices is often directly related to the frequency of use. Therefore, adherence to personal health devices is critical. This paper takes a data mining approach to study continuous glucose monitor use in diabetes management. We evaluate two independent datasets from a total of 44 subjects for 60 - 270 days. Our results show that: 1) missed target goals (i.e. suboptimal outcomes) is a factor that is associated with wearing behavior of personal health devices, and 2) longer duration of non-adherence, identified through missing data or data gaps, is significantly associated with poorer outcomes. More specifically, we found that up to 33% of data gaps occurred when users were in abnormal blood glucose categories. The longest data gaps occurred in the most severe (i.e. very low / very high) glucose categories. Additionally, subjects with poorly-controlled diabetes had longer average data gap duration than subjects with well-controlled diabetes. This work contributes to the literature on the design of context-aware systems that can leverage data-driven approaches to understand factors that influence non-wearing behavior. The results can also support targeted interventions to improve health outcomes.
△ Less
Submitted 29 May, 2020;
originally announced June 2020.
-
Summary: Multi-modal Biometric-based Implicit Authentication of Wearable Device Users
Authors:
Sudip Vhaduri,
Christian Poellabauer
Abstract:
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 trans…
▽ More
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.
△ Less
Submitted 15 July, 2019;
originally announced July 2019.
-
Biometric-Based Wearable User Authentication During Sedentary and Non-sedentary Periods
Authors:
Sudip Vhaduri,
Christian Poellabauer
Abstract:
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 the recent advances in mobile sensing, wearables have become a rich collection of portable sensors and are able to provide various types of services including health and fitness tracking, financial transaction…
▽ More
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 the recent advances in mobile sensing, wearables have become a rich collection of portable sensors and are able to provide various types of services including health and fitness tracking, financial transactions, and unlocking smart locks and vehicles. Existing explicit authentication approaches (i.e., PINs or pattern locks) suffer from several limitations including limited display size, shoulder surfing, and 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-grained minute-level biometrics: behavioral (step counts), physiological (heart rate), and hybrid (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 authenticate subjects with average accuracy values of around 92% and 88% during sedentary and non-sedentary periods, respectively. Our findings also show that (a) behavioral biometrics do not work well during sedentary periods and (b) hybrid biometrics typically perform better than other biometrics.
△ Less
Submitted 16 November, 2018;
originally announced November 2018.
