Coping with stress is reportedly one of the main reasons for chronic cannabis use. Developing a real-time system that offers cannabis users alternative methods to cope with stress is of interest in medical applications. To develop such a system, it is necessary to design a reliable mechanism for identifying cannabis use sessions in uncontrolled environments using physiological markers captured with wearable sensors. Therefore, the primary objective of this study is to design a system that can identify sessions of cannabis consumption by utilizing one of the most significant biomarkers of stress, Electrodermal Activity (EDA). We conducted a user study to collect physiological sensor data in real-life setting. We then model the cannabis use detection as a supervised learning problem and train a neural network model. To improve the performance of the proposed model for a specific subject, transfer learning techniques were used to retrain the base model on the new user data. Trained model achieved average f1-score of 0.68 and accuracy of 71.58% on the test data from Leave One Subject Out (LOSO) analysis. After applying transfer learning, the retrained model achieved average f1-score of 0.8 and accuracy of 83.61% when detecting the cannabis consumption period for the same subjects.
@inproceedings{cannabisBSN,title={Personalized Modeling and Detection of Moments of Cannabis Use in Free-Living Environments},author={Azghan, Reza Rahimi and Glodosky, Nicholas C and Sah, Ramesh Kumar and Cuttler, Carrie and McLaughlin, Ryan and Cleveland, Michael J and Ghasemzadeh, Hassan},booktitle={2023 IEEE-EMBS International Conference on Wearable and Implantable Body Sensor Networks (BSN)},year={2023},abbr={Conference},organization={IEEE},bibtex_show={true},website={https://ghasemzadeh.com/publication/detection-of-moments-of-cannabis-use-bsn23/}}
Journal
Stress Monitoring in Free-Living Environments
Sah, Ramesh Kumar, Cleveland, Michael J, and Ghasemzadeh, Hassan
IEEE Journal of Biomedical and Health Informatics (JBHI) 2023
Stress monitoring is an important area of research with significant implications for individuals’ physical and mental health. We present a data-driven approach for stress detection based on convolutional neural networks while addressing the problems of the best sensor channel and the lack of knowledge about stress episodes. Our work is the first to present an analysis of stress-related sensor data collected in real-world conditions from individuals diagnosed with Alcohol Use Disorder (AUD) and undergoing treatment to abstain from alcohol. We developed polynomial-time sensor channel selection algorithms to determine the best sensor modality for a machine learning task. We model the time variation in stress labels expressed by the participants as the subjective effects of stress. We addressed the subjective nature of stress by determining the optimal input length around stress events with an iterative search algorithm.
@article{jbhiStressSah,title={Stress Monitoring in Free-Living Environments},author={Sah, Ramesh Kumar and Cleveland, Michael J and Ghasemzadeh, Hassan},journal={IEEE Journal of Biomedical and Health Informatics (JBHI)},year={2023},publisher={IEEE},abbr={Journal},bibtex_show={true},website={https://ghasemzadeh.com/publication/jbhi-stress-monitoring/},url={https://zenodo.org/record/6640290}}
2022
Conference
Comparing the Predictability of Sensor Modalities to Detect Stress from Wearable Sensor Data
Holder, Ryan,Â
Sah, Ramesh Kumar, Cleveland, Michael, and Ghasemzadeh, Hassan
Detecting stress from wearable sensor data enables those struggling with unhealthy stress coping mechanisms to better manage their stress. Previous studies have investigated how mechanisms for detecting stress from sensor data can be optimized, comparing alternative algorithms and approaches to find the best possible outcome. One strategy to make these mechanisms more accessible is to reduce the number of sensors that wearable devices must support. Reducing the number of sensors will enable wearable devices to be a smaller size, require less battery, and last longer, making use of these wearable devices more accessible. To progress towards this more convenient stress detection mechanism, we investigate how learning algorithms perform on singular modalities and compare the outcome with results from multiple modalities. We found that singular modalities performed comparably or better than combined modalities on two stress-detection datasets, suggesting that there is promise for detecting stress with fewer sensor requirements. Our results are acquired from testing with random holdout and leave-one-subject-out validation, using several machine learning techniques. Our results can inspire work on optimizing stress detection with singular modalities to make the benefits of these detection mechanisms more convenient.
@article{holder2022comparing,title={Comparing the Predictability of Sensor Modalities to Detect Stress from Wearable Sensor Data},author={Holder, Ryan and Sah, Ramesh Kumar and Cleveland, Michael and Ghasemzadeh, Hassan},booktitle={2022 IEEE 19th Annual Consumer Communications \& Networking Conference (CCNC)},pages={557--562},year={2022},pdf={https://drive.google.com/file/d/1alWRUM3GnjkK4WvUQY_khUsciYgjcwvQ/view?usp=sharing},organization={IEEE},abbr={Conference},bibtex_show={true},website={https://ghasemzadeh.com/project/stress-mobile-health/}}
Conference
Continual Learning for Activity Recognition
Sah, Ramesh Kumar, Seyd, Iman Mirzadeh, and Ghasemzadeh, Hassan
The recent success of deep neural networks in prediction tasks on wearable sensor data is evident. However, in more practical online learning scenarios, where new data arrive sequentially, neural networks suffer severely from the “catastrophic forgetting“ problem. In real-world settings, given a pre-trained model on the old data, when we collect new data, it is practically infeasible to re-train the model on both old and new data because the computational costs will increase dramatically as more and more data arrive in time. However, if we fine-tune the model only with the new data because the new data might be different from the old data, the neural network parameters will change to fit the new data. As a result, the new parameters are no longer suitable for the old data. This phenomenon is known as catastrophic forgetting, and continual learning research aims to overcome this problem with minimal computational costs. While most of the continual learning research focuses on computer vision tasks, implications of catastrophic forgetting in wearable computing research and potential avenues to address this problem have remained unexplored. To address this knowledge gap, we study continual learning for activity recognition using wearable sensor data. We show that the catastrophic forgetting problem is a critical challenge for the real-world deployment of machine learning models for wearable sensor data. Moreover, we show that the catastrophic forgetting problem can be alleviated by employing various training techniques.
@article{continualLearning,title={Continual Learning for Activity Recognition},author={Sah, Ramesh Kumar and Seyd, Iman Mirzadeh and Ghasemzadeh, Hassan},booktitle={2022 IEEE Engineering in Medicine and Biology Conference (EMBC)},year={2022},pdf={https://drive.google.com/file/d/1hbaeB0OVxynZj0ih2OmAQsGMbMBBUM0t/view?usp=sharing},abbr={Conference},bibtex_show={true}}
Conference
Stressalyzer: Convolutional Neural Network Framework for Personalized Stress Classification
Sah, Ramesh Kumar, Cleveland, Michael, Habibi, Assal, and Ghasemzadeh, Hassan
Stress detection and monitoring is an active area of research with important implications for an individual’s personal, professional, and social health. Current approaches for stress classification use traditional machine learning algorithms trained on features computed from multiple sensor modalities. These methods are data and computation-intensive, rely on hand-crafted features, and lack reproducibility. These limitations impede the practical use of stress detection and classification systems in the real world. To overcome these shortcomings, we propose Stressalyzer, a novel stress classification and personalization framework from single-modality sensor data without feature computation and selection. Stressalyzer uses only Electrodermal activity (EDA) sensor data while providing competitive results compared to the state-of-the-art techniques that use traditional machine learning models. Our single-channel neural network-based model achieves a classification accuracy of 92.9% and an f 1 score of 0.89 for binary stress classification. Our leave-one-subject-out analysis establishes the subjective nature of stress and shows that personalizing stress models using Stressalyzer significantly im- proves the model performance. Without model personalization, we found a performance decline in 40% of the subjects, suggesting the need for model personalization.
@article{stressPersonalization,title={Stressalyzer: Convolutional Neural Network Framework for Personalized Stress Classification},author={Sah, Ramesh Kumar and Cleveland, Michael and Habibi, Assal and Ghasemzadeh, Hassan},booktitle={2022 IEEE Engineering in Medicine and Biology Conference (EMBC)},year={2022},pdf={https://drive.google.com/file/d/1frDHECipfU4hkiHgmiYdGQuQwwANOS_4/view?usp=sharing},abbr={Conference},bibtex_show={true},website={https://ghasemzadeh.com/project/stress-mobile-health/}}
Journal
Probabilistic Cascading Classifier for Energy-Efficient Activity Monitoring in Wearables
Advances in embedded systems have given rise to integrating several small-size health monitoring devices within daily human life. This trend led to an ongoing extension of wearable sensors in a broad range of applications. Wearable technologies, which are firmly connected with the human body, utilize sensors and machine learning to describe individuals’ physical or psychological routines through activity recognition and human movement. Since wearables are used all day long, the power consumption of these systems needs to be reasonably low. Current research considers that such machine learning methods are trained with fixed properties, including sensor sampling rate and statistical features computed from the time series data. However, in reality, wearables require continuous reconfiguration of their computational algorithms due to the personalized nature of human gait and movement. Furthermore, computational algorithms must become energy- and memory-efficient due to these embedded sensors’ limited power and memory. In this paper, we propose a resource-efficient framework for real-time, continuous, and on-node human activity recognition. Typically activity recognition problem is a multi-class classification problem. However, we suggest transforming this problem based on MET (Metabolic Equivalent of Task) into a hierarchical classification model, providing personalized structure for each individual. We discuss the design and construction of this new configurable classification paradigm. Our results demonstrate that the proposed probabilistic cascading system accuracy for different personalized scenarios varies between 94.5% and 96.9% in detecting activities using a limited memory, while power usage of the system is reduced by as high as 17.2% compared to the traditional methods.
@article{mahdiBinaryCascading,title={Probabilistic Cascading Classifier for Energy-Efficient Activity Monitoring in Wearables},author={Pedram, Mahdi and Sah, Ramesh Kumar and Rokni, Ali and Nourollahi, Marjan and Ghasemzadeh, Hassan},journal={IEEE Sensors},year={2022},publisher={IEEE Sensors Journal},abbr={Journal},bibtex_show={true},pdf={https://drive.google.com/file/d/1HaLR1tNqHYj6sky558kvVsXaYDJIRFD6/view?usp=sharing},website={https://ghasemzadeh.com/publication/cascading-binary-sensors/}}
Conference
ADARP: A Multi Modal Dataset for Stress and Alcohol Relapse Quantification in Real Life Setting
Sah, Ramesh Kumar, McDonell, Michael, Pendry, Patricia, Parent, Sara, Ghasemzadeh, Hassan, and Cleveland, Michael J
In 2022 IEEE-EMBS International Conference on Wearable and Implantable Body Sensor Networks (BSN) 2022
Stress detection and classification from wearable sensor data is an emerging area of research with significant implications for individuals’ physical and mental health. In this work, we introduce a new dataset, ADARP, which contains physiological data and self-report outcomes collected in real- world ambulatory settings involving individuals diagnosed with alcohol use disorders. We describe the user study, present details of the dataset, establish the significant correlation between physiological data and self-reported outcomes, demonstrate stress classification, and make our dataset public to facilitate research.
@inproceedings{sah2022adarp,title={ADARP: A Multi Modal Dataset for Stress and Alcohol Relapse Quantification in Real Life Setting},author={Sah, Ramesh Kumar and McDonell, Michael and Pendry, Patricia and Parent, Sara and Ghasemzadeh, Hassan and Cleveland, Michael J},booktitle={2022 IEEE-EMBS International Conference on Wearable and Implantable Body Sensor Networks (BSN)},pages={1--4},year={2022},abbr={Conference},organization={IEEE},bibtex_show={true},pdf={https://arxiv.org/pdf/2206.14568.pdf},website={https://ghasemzadeh.com/publication/adarp-dataset/},url={https://zenodo.org/record/6640290}}
2021
arXiv
Stress Classification and Personalization: Getting the most out of
the least
Stress detection and monitoring is an active area of research with important implications for the personal, professional, and social health of an individual. Current approaches for affective state classification use traditional machine learning algorithms with features computed from multiple sensor modalities. These methods are data-intensive and rely on hand-crafted features which impede the practical applicability of these sensor systems in daily lives. To overcome these shortcomings, we propose a novel Convolutional Neural Network (CNN) based stress detection and classification framework without any feature computation using data from only one sensor modality. Our method is competitive and outperforms current state-of-the-art techniques and achieves a classification accuracy of 92.85% and an f1 score of 0.89. Through our leave-one-subject-out analysis, we also show the importance of personalizing stress models.
@article{stressMostofLeast,author={Sah, Ramesh Kumar and Ghasemzadeh, Hassan},title={Stress Classification and Personalization: Getting the most out of
the least},year={2021},pdf={https://drive.google.com/file/d/1mJ419kC5YY2jV448u1181MPByD-Xq9rx/view?usp=sharing},eprinttype={arXiv},eprint={2107.05666},timestamp={Tue, 20 Jul 2021 15:08:33 +0200},biburl={https://dblp.org/rec/journals/corr/abs-2107-05666.bib},bibsource={dblp computer science bibliography, https://dblp.org},abbr={arXiv},bibtex_show={true},website={https://ghasemzadeh.com/project/stress-mobile-health/}}
Journal
Associations Between Physiological Signals Captured Using Wearable Sensors and Self-reported Outcomes Among Adults in Alcohol Use Disorder Recovery: Development and Usability Study
Alinia, Parastoo,Â
Sah, Ramesh Kumar, McDonell, Michael, Pendry, Patricia, Parent, Sara, Ghasemzadeh, Hassan, Cleveland, Michael John, and others,
Previous research has highlighted the role of stress in substance misuse and addiction, particularly for relapse risk. Mobile health interventions that incorporate real-time monitoring of physiological markers of stress offer promise for delivering tailored interventions to individuals during high-risk states of heightened stress to prevent alcohol relapse. Before such interventions can be developed, measurements of these processes in ambulatory, real-world settings are needed. This research is a proof-of-concept study to establish the feasibility of using a wearable sensor device to continuously monitor stress in an ambulatory setting. Toward that end, we first aimed to examine the quality of 2 continuously monitored physiological signals—electrodermal activity (EDA) and heart rate variability (HRV)—and show that the data follow standard quality measures according to the literature. Next, we examined the associations between the statistical features extracted from the EDA and HRV signals and self-reported outcomes. We determined that 87.86% (1,032,265/1,174,898) of the EDA signals were clean. A comparison of the frequency of skin conductance responses per minute with previous research confirmed that the physiological signals collected in the ambulatory setting were successful. The results also indicated that the statistical features of the EDA and HRV measures were significantly correlated with the self-reported outcomes, including the number of stressful events marked on the sensor device, positive and negative emotions, and experienced pain and discomfort.
@article{alinia2021associations,title={Associations Between Physiological Signals Captured Using Wearable Sensors and Self-reported Outcomes Among Adults in Alcohol Use Disorder Recovery: Development and Usability Study},author={Alinia, Parastoo and Sah, Ramesh Kumar and McDonell, Michael and Pendry, Patricia and Parent, Sara and Ghasemzadeh, Hassan and Cleveland, Michael John and others},journal={JMIR Formative Research},volume={5},number={7},pages={e27891},year={2021},pdf={https://drive.google.com/file/d/1H8IpfgF4lJ62Wm7wN8mqXT39ehw5TEta/view?usp=sharing},publisher={JMIR Publications Inc., Toronto, Canada},abbr={Journal},bibtex_show={true},website={https://ghasemzadeh.com/project/stress-mobile-health/}}
2020
arXiv
Adversarial Transferability in Wearable Sensor Systems
Machine learning algorithms are increasingly used for inference and decision-making in embedded systems. Data from sensors are used to train machine learning models for various smart functions of embedded and cyber-physical systems ranging from applications in healthcare and emergency response to autonomous vehicles and national security. However, recent studies have shown that machine learning models can be attacked by adding adversarial noise to their inputs. The perturbed inputs are called adversarial examples. Adversarial examples that attack a machine learning model trained in a source domain are also often effective against the model trained in a target domain. This property of adversarial examples is called adversarial transferability and has not been explored in wearable systems to date. In this work, we take the first stride in studying adversarial transferability in wearable systems from four distinct viewpoints: (1) transferability between machine learning models; (2) transferability across users/subjects of the embedded system; (3) transferability across sensor body locations; and (4) transferability across datasets used for model training. We present a set of carefully designed experiments to investigate these transferability scenarios. We also propose a threat model describing the interactions of an adversary with the source and target sensor systems in different transferability settings. In most cases, we found strong untargeted transferability, whereas targeted attacks were less successful with success scores from 0% to 80%. The transferability of adversarial examples depends on many factors such as the inclusion of data from all subjects, sensor body position, number of samples in the dataset, type of learning algorithm, and the distribution of source and target system dataset. The transferability of adversarial examples decreases sharply when the data distribution of the source and target system becomes more distinct. We also provide further guidelines and suggestions for the community for designing robust sensor systems.
@article{advTransferability,author={Sah, Ramesh Kumar and Ghasemzadeh, Hassan},title={Adversarial Transferability in Wearable Sensor Systems},journal={CoRR},volume={abs/2003.07982},year={2020},pdf={https://drive.google.com/file/d/1HWuSxue_OL9LSmAOGVfu_LzHX2UZm-Ov/view?usp=sharing},eprinttype={arXiv},eprint={2003.07982},timestamp={Thu, 17 Dec 2020 16:38:47 +0100},biburl={https://dblp.org/rec/journals/corr/abs-2003-07982.bib},bibsource={dblp computer science bibliography, https://dblp.org},abbr={arXiv},bibtex_show={true},website={https://ghasemzadeh.com/project/computational-autonomy/}}
Conference
Poster: Mobile Health for Alcohol Recovery and Relapse
Sah, Ramesh Kumar, Ghasemzadeh, Hassan, Habibi, Assal, McDonell, Michael, Patricia, Pendry, and Cleveland, Michael
Alcohol related disorder has increasingly become a serious public health issue. Stress detection and intervention is considered a key element in a treatment strategy towards preventing alcohol dependent individuals from relapsing. In this paper, we present a proof-of-concept approach to study the usability of a wearable device and the viability of a mobile health application to prevent alcohol relapse by detecting moments of stress and providing adaptive interventions in real-time.
@article{sah2020poster,title={Poster: Mobile Health for Alcohol Recovery and Relapse},author={Sah, Ramesh Kumar and Ghasemzadeh, Hassan and Habibi, Assal and McDonell, Michael and Patricia, Pendry and Cleveland, Michael},booktitle={2020 IEEE/ACM International Conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE)},pages={18--19},year={2020},pdf={https://drive.google.com/file/d/1QvzvsC5_u7Bo_tvebpwtHluVy8Ikq2Je/view?usp=sharing},organization={IEEE},abbr={Conference},bibtex_show={true},website={https://ghasemzadeh.com/project/stress-mobile-health/}}
2019
Conference
Adar: Adversarial activity recognition in wearables
Recent advances in machine learning and deep neural networks have led to the realization of many important applications in the area of personalized medicine. Whether it is detecting activities of daily living or analyzing images for cancerous cells, machine learning algorithms have become the dominant choice for such emerging applications. In particular, the state-of-the-art algorithms used for human activity recognition (HAR) using wearable inertial sensors utilize machine learning algorithms to detect health events and to make predictions from sensor data. Currently, however, there remains a gap in research on whether or not and how activity recognition algorithms may become the subject of adversarial attacks. In this paper, we take the first strides on (1) investigating methods of generating adversarial example in the context of HAR systems; (2) studying the vulnerability of activity recognition models to adversarial examples in feature and signal domain; and (3) investigating the effects of adversarial training on HAR systems. We introduce Adar, a novel computational framework for optimization-driven creation of adversarial examples in sensor-based activity recognition systems. Through extensive analysis based on real sensor data collected with human subjects, we found that simple evasion attacks are able to decrease the accuracy of a deep neural network from 95.1% to 3.4% and from 93.1% to 16.8% in the case of a convolutional neural network. With adversarial training, the robustness of the deep neural network increased on the adversarial examples by 49.1% in the worst case while the accuracy on clean samples decreased by 13.2%.
@article{sah2019adar,title={Adar: Adversarial activity recognition in wearables},author={Sah, Ramesh Kumar and Ghasemzadeh, Hassan},booktitle={2019 IEEE/ACM International Conference on Computer-Aided Design (ICCAD)},pages={1--8},year={2019},pdf={https://drive.google.com/file/d/1RhtE7mxM-6nsqePUdJqZVyWEQMoH1MNj/view?usp=sharing},organization={IEEE},abbr={Conference},bibtex_show={true},website={https://ghasemzadeh.com/project/computational-autonomy/}}