Skip to main content

Rajdeep Nath

Followers

3

Following

1

Public Views

Address: India

less

Illinois Institute of Technology

Lazaros Karagiannidis

Niaz Chalabianloo

Journal of Emerging Computer Technologies

Celal Bayar Üniversitesi

University of Michigan

Philip Gouverneur

Wrocław University of Technology

Interests

Uploads

Papers by Rajdeep Nath

Wearable Health Monitoring System for Older Adults in a Smart Home Environment

2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI), 2021

The advent of IoT has enabled the design of connected and integrated smart health monitoring syst... more The advent of IoT has enabled the design of connected and integrated smart health monitoring systems. These smart health monitoring systems could be realized in a smart home context to render long-term care to the elderly population. In this paper, we present the design of a wearable health monitoring system suitable for older adults in a smart home context. The proposed system offers solutions to monitor the stress, blood pressure, and location of an individual within a smart home environment. The stress detection model proposed in this work uses Electrodermal Activity (EDA), Photoplethysmogram (PPG), and Skin Temperature (ST) sensors embedded in a smart wristband for detecting physiological stress. The stress detection model is trained and tested using stress labels obtained from salivary cortisol which is a clinically established biomarker for physiological stress. A voice-based prototype is also implemented and the feasibility of the proposed system for integration in a smart home environment is analyzed by simulating a data acquisition and streaming scenario. We have also proposed a blood pressure estimation model using PPG signal and advanced regression techniques for integration with the stress detection model in the wearable health monitoring system. Finally, the design of a voice-assisted indoor location system is proposed for integration with the proposed system within a smart home environment. The proposed wearable health monitoring system is an important direction to realize a smart home environment with extensive diagnostic capabilities so that such a system could be useful for rendering long-term and personalized care to the aging population in the comfort of their home.

Machine Learning Based Prediction of Future Stress Events in a Driving Scenario

2021 IEEE 7th World Forum on Internet of Things (WF-IoT), 2021

This paper presents a model for predicting a driver's stress level up to one minute in advance. S... more This paper presents a model for predicting a driver's stress level up to one minute in advance. Successfully predicting future stress would allow stress mitigation to begin before the subject becomes stressed, reducing or possibly avoiding the performance penalties of stress. The proposed model takes features extracted from Galvanic Skin Response (GSR) signals on the foot and hand and Respiration and Electrocardiogram (ECG) signals from the chest of the driver. The data used to train the model was retrieved from an existing database and then processed to create statistical and frequency features. A total of 42 features were extracted from the data and then expanded into a total of 252 features by grouping the data and taking six statistical measurements of each group for each feature. A Random Forest Classifier was trained and evaluated using a leave-one-subject-out testing approach. The model achieved 94% average accuracy on the test data. Results indicate that the model performs well and could be used as part of a vehicle stress prevention system.

Machine Learning-Based Anxiety Detection in Older Adults Using Wristband Sensors and Context Feature

SN Computer Science, 2021

This paper explores a novel method for anxiety detection in older adults using simple wristband s... more This paper explores a novel method for anxiety detection in older adults using simple wristband sensors such as electrodermal activity (EDA) and photoplethysmogram (PPG) and a context-based feature. The proposed method for anxiety detection combines features from a single physiological signal with an experimental context-based feature to improve the performance of the anxiety detection model. The experimental data for this work are obtained from a year-long experiment on 41 healthy older adults (26 females and 15 males) in the age range 60-80 with mean age 73.36 ± 5.25 during a Trier Social Stress Test (TSST) protocol. The anxiety level ground truth was obtained from State-Trait Anxiety Inventory (STAI), which is regarded as the gold standard to measure perceived anxiety. EDA and blood volume pulse (BVP) signals were recorded using a wristworn EDA and PPG sensor, respectively. 47 features were computed from EDA and BVP signal, out of which a final set of 24 significantly correlated features were selected for analysis. The phases of the experimental study are encoded as unique integers to generate the context feature vector. A combination of features from a single sensor with the context feature vector is used for training a machine learning model to distinguish between anxious and not-anxious states. Results and analysis showed that the EDA and BVP machine learning models that combined the context feature along with the physiological features achieved 3.37% and 6.41% higher accuracy, respectively, than the models that used only physiological features. Further, end-to-end processing of EDA and BVP signals was simulated for real-time anxiety level detection. This work demonstrates the practicality of the proposed anxiety detection method in facilitating long-term monitoring of anxiety in older adults using low-cost consumer devices.

Towards Photoplethysmogram Based Non-Invasive Blood Pressure Classification

2018 IEEE International Symposium on Smart Electronic Systems (iSES) (Formerly iNiS), 2018

A novel blood pressure classification model using Phototplethysmogram (PPG) is proposed in this w... more A novel blood pressure classification model using Phototplethysmogram (PPG) is proposed in this work. The proposed model uses signal processing and machine learning algorithms to classify blood pressure in four stages: normal, elevated, stage 1 and stage 2. A total of 83 features were extracted from the PPG signal which includes 71 statistical features and 12 characteristic features. We have used random forest classifier to train and test our predictive model. The proposed method is evaluated on publicly available MIMIC database for 20 different individuals. The database contains raw PPG data for different users and Arterial Blood Pressure (ABP) to calculate the systolic and diastolic blood pressure to be used as the ground truth for training and validation purposes. We have achieved an overall accuracy of 90.8% over the four classes of blood pressure levels. The results indicate that the proposed model will be ideal for integration into a non-invasive blood pressure monitoring system with significant accuracy.

Quantum Annealing for Automated Feature Selection in Stress Detection

2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI), 2021

We present a novel methodology for automated feature subset selection from a pool of physiologica... more We present a novel methodology for automated feature subset selection from a pool of physiological signals using Quantum Annealing (QA). As a case study, we will investigate the effectiveness of QA-based feature selection techniques in selecting the optimal feature subset for stress detection. Features are extracted from four signal sources: foot EDA, hand EDA, ECG, and respiration. The proposed method embeds the feature variables extracted from the physiological signals in a binary quadratic model. The bias of the feature variable is calculated using the Pearson correlation coefficient between the feature variable and the target variable. The weight of the edge connecting the two feature variables is calculated using the Pearson correlation coefficient between two feature variables in the binary quadratic model. Subsequently, D-Wave's clique sampler is used to sample cliques from the binary quadratic model. The underlying solution is then re-sampled to obtain multiple good solutions and the clique with the lowest energy is returned as the optimal solution. The proposed method is compared with commonly used feature selection techniques for stress detection. Results indicate that QA-based feature subset selection performed equally as that of classical techniques. However, under data uncertainty conditions such as limited training data, the performance of quantum annealing for selecting optimum features remained unaffected, whereas a significant decrease in performance is observed with classical feature selection techniques. Preliminary results show the promise of quantum annealing in optimizing the training phase of a machine learning classifier, especially under data uncertainty conditions.

Validating Physiological Stress Detection Model Using Cortisol as Stress Bio Marker

2020 IEEE International Conference on Consumer Electronics (ICCE), 2020

In this work, we have presented the validation of a stress detection model using cortisol as the ... more In this work, we have presented the validation of a stress detection model using cortisol as the stress biomarker. The proposed model uses two physiological signals: Galvanic Skin Response (GSR) and Photoplethysmograph (PPG) to classify stress into two levels. GSR and PPG signals were collected from a total of 13 participants along with saliva samples taken at time points throughout the duration of the experiment. We have used 10 out of the 13 participants to train our model. Data from the remaining 3 participants was used to test the robustness of the model in distinguishing stressed states from non-stressed states. We have achieved an overall accuracy of 92% with the model achieving precision, recall and f1-score of 93%, 99% and 96% respectively in predicting the occurrences of stressful events. Results indicate the promise of the proposed methodology in accurately detecting the presence of stressful events by generalizing the test data coming from a subset of population in contrast to the training data.

A Review of Machine Learning Classification Using Quantum Annealing for Real-World Applications

SN Computer Science, 2021

Optimizing the training of a machine learning pipeline helps in reducing training costs and impro... more Optimizing the training of a machine learning pipeline helps in reducing training costs and improving model performance. One such optimizing strategy is quantum annealing, which is an emerging computing paradigm that has shown potential in optimizing the training of a machine learning model. The implementation of a physical quantum annealer has been realized by D-wave systems and is available to the research community for experiments. Recent experimental results on a variety of machine learning applications using quantum annealing have shown interesting results where the performance of classical machine learning techniques is limited by limited training data and high dimensional features. This article explores the application of D-wave's quantum annealer for optimizing machine learning pipelines for real-world classification problems. We review the application domains on which a physical quantum annealer has been used to train machine learning classifiers. We discuss and analyze the experiments performed on the D-Wave quantum annealer for applications such as image recognition, remote sensing imagery, computational biology, and particle physics. We discuss the possible advantages and the problems for which quantum annealing is likely to be advantageous over classical computation.

Smart Wristband-Based Stress Detection Framework for Older Adults With Cortisol as Stress Biomarker

IEEE Transactions on Consumer Electronics, 2021

In this work, our objective is to design, develop, and evaluate the effectiveness of a stress det... more In this work, our objective is to design, develop, and evaluate the effectiveness of a stress detection model for older adults using a system of wrist-worn sensors. Our system uses four signals, EDA, BVP, IBI, and ST from EDA, PPG, and ST sensors, embedded in a smart wristband, to classify between stressed and not-stressed state. The stress reference is obtained from salivary cortisol measurement, which is a well established clinical biomarker for measuring physiological stress. This work is the result of year-long data collection and analysis of 40 older adults (28 females and 12 males) and age 73.625 ± 5.39. EDA, BVP, IBI, and ST signals were collected during TSST (Trier Social Stress Test), which is a well known experimental protocol to reliably induce stress in humans in a social setting. 47 features were extracted from EDA, BVP, IBI, and ST signals, out of which 27 features were selected using a supervised feature selection method. Results and analysis show that combining the features from all the four signal streams increases the model’s ability to accurately distinguish between the stressed and not-stressed states. The proposed model achieved a macro-average F1-score of 0.92 and an accuracy of 94% in distinguishing between the two states when features from all the four signals were used. Further, we prototype the proposed stress detection model in a consumer end device with voice capabilities, so that users can receive feedback on their vitals and stress levels by querying on voice-enabled consumer devices such as smartphones and smart speakers.

Machine Learning Based Stress Monitoring in Older Adults Using Wearable Sensors and Cortisol as Stress Biomarker

Journal of Signal Processing Systems, 2021

The objective of this work is to evaluate the effectiveness of a wearable physiological stress mo... more The objective of this work is to evaluate the effectiveness of a wearable physiological stress monitoring system in distinguishing between stressed and non-stressed state in older adults using machine learning techniques. This system utilizes EDA and BVP signal to detect occurrence of stress as indicated by salivary cortisol measurement which is a reliable objective measure of physiological stress. Data of 19 healthy older adults (11 female and 8 male) with mean age 73.15 ± 5.79 were used for this study. EDA and BVP signals were recorded using a finger tip sensor during the Trier Social Stress Test, which is a well known experimental protocol to reliably induce stress in humans in a social setting. 39 statistical measures of the peak characteristic of EDA and BVP signal were extracted. A supervised feature selection algorithm is used to select important features as an input to the machine learning model. Four machine learning algorithms were evaluated based on their performance in classifying between stressed and non-stressed states. Results indicate that the logistic regression performed the best among Random Forest, κ-NN, and Support Vector Machine achieving an macro-average and micro-average f1-score of 0.87 and 0.95 respectively and an AUC score of 0.81. We also evaluated the effectiveness of a novel deep learning Long Short-Term Memory (LSTM) based classifier in distinguishing between stressed and non-stressed state. Results on test data shows that LSTM based classifier achieved an improvement of 6.7% and 2% in terms of macro-average f1-score and micro-average f1-score respectively. Also the AUC score for LSTM classifier is found to be 0.9 which is about 11% higher than the best performing logistic regression model. This work can be used to design a convenient unobtrusive wearable device to monitor stress levels in older adults in their home environment, thereby facilitating aging in place and improving the quality of life.

Machine Learning Based Solutions for Real-Time Stress Monitoring

IEEE Consumer Electronics Magazine, 2020

Stress may be defined as the reaction of the body to regulate itself to changes within the enviro... more Stress may be defined as the reaction of the body to regulate itself to changes within the environment through mental, physical, or emotional responses. Recurrent episodes of acute stress can disturb the physical and mental stability of a person. This subsequently can have a negative effect on work performance and in the long term can increase the risk of physiological disorders like hypertension and psychological illness such as anxiety disorder. Psychological stress is a growing concern for the worldwide population across all age groups. A reliable, costefficient, acute stress detection system could enable its users to better monitor and manage their stress to mitigate its long-term negative effects. In this article, we will review and discuss the literature that has used machine learning based approaches for stress detection. We will also review the existing solutions in the literature that have leveraged the concept of edge computing in providing a potential solution in real-time monitoring of stress.

Productivity enhancement by component-oriented line balancing method

International Journal of Industrial and Systems Engineering, 2018

The problem of balancing a production line has been one of the typical complications in the field... more The problem of balancing a production line has been one of the typical complications in the field of industrial engineering. An unbalanced production line associates with the downfall in productivity. Since productivity is the 'primus motor' of income in any firm, the above problem cannot be over sighted. In this paper a similar problem is being faced by a company which is unable to cope with the increasing demands of their product. The past data of component manufacturing process, resources required and the time taken is collected, when this data is analysed, constraints are observed in the testing machine of the production line. The elimination of bottleneck in this line is needed in order to escalate the productivity. The intended rate of production is achieved through component-oriented line balancing (COLB) approach. In this method the number of workstations is increased and hence bottle neck is removed. The comparison of the data before and after the implementation of component-oriented line balancing shows a simultaneous improvement in the productivity and reduction of labour cost.

Smart Home Environment for Mild Cognitive Impairment Population: Solutions to Improve Care and Quality of Life

IEEE Consumer Electronics Magazine, 2018

Solutions to improve care and quality of life.

Towards Designing Reliable Universal QCA Logic in the Presence of Cell Deposition Defect

2016 29th International Conference on VLSI Design and 2016 15th International Conference on Embedded Systems (VLSID), 2016

This work targets design of a reliable universal logic gate in Quantum-dot cellular automata (r-U... more This work targets design of a reliable universal logic gate in Quantum-dot cellular automata (r-ULG with the single clock zone, r-ULG-II with multiple clock zone) with hybrid cell orientations that realizes majority and minority functions simultaneously with high fault tolerance. The characterization of the defective behaviour of r-ULGs under different cell deposition defects is investigated. The outcomes indicate that the proposed r-ULG provide 75% fault tolerance under single clock zone and 100% under two clock zone.

On the reliability of majority logic structure in quantum-dot cellular automata

Microelectronics Journal, 2016

Quantum-dot cellular automata (QCA) is projected to be a promising nanotechnology due to its extr... more Quantum-dot cellular automata (QCA) is projected to be a promising nanotechnology due to its extremely small feature size and ultra low power consumption. However, acceptance of a QCA design is limited due to its high defect rate. Efficient fault tolerant schemes are, therefore, needed for reliable design. This work targets design of a new fault tolerant scheme around QCA logic primitives which encapsulates two different orientations of QCA cell. A 2 Â 2 array of four rotated (' þ') cells, called complementary tile (CT), is introduced to maximize the throughput. It ensures 100% fault tolerance under single cell missing defect. Two reliable majority voters (RMV), based on the CT, are designed which outperforms the existing majority logic in QCA. The functional characterization and polarization of RMV under different cell deposition (missing/additional) defects are covered. The significance of the clocking in fault tolerance is also investigated with RMV with multi clock zone. The error probability model for the proposed RMV, under cell deposition (missing/additional) defect, is developed to ensure better understanding of reliability in QCA.

Towards the design of hybrid QCA tiles targeting high fault tolerance

Journal of Computational Electronics, 2015

The increasing fabrication cost of CMOS-based computing devices and the ever-approaching limits o... more The increasing fabrication cost of CMOS-based computing devices and the ever-approaching limits of their fabrication have led to the search for feasible options with high device density and low power waste. Quantum-dot cellular automata (QCA) is an emerging technology with such potential to match the design target beyond the limits of state-of-the-art CMOS. But nanotechnologies, like QCA are extremely susceptible to various forms of flaws and variations during fabrication at nano scale. Thus, the exploration of ingenious fault tolerant structure around QCA is gaining high importance. This work targets a new robust QCA tile structure hybridizing rotated and non-rotated cell together resulting lesser kink energy. Different QCA logic primitives (majority/minority logic, fanout tiles, etc.) are made using such hybrid cell structure. The functional characterization using the kink energy and the polarization level of such QCA structures under different cell defects have been thoroughly investigated. The results suggest that the proposed QCA logic primitives have achieved high fault tolerance of 97.43 %. Also, 100 % fault tolerance can be ascertained if the proposed logic circuit drives the correct output with the application of 001, 011 as a primitive test vector only. A comparative per-B Bibhash Sen

Fault masking in Quantum-dot cellular automata using prohibitive logic circuit

2015 19th International Symposium on VLSI Design and Test, 2015

This work targets the challenges to design with uncertainty in QCA architecture via characterizat... more This work targets the challenges to design with uncertainty in QCA architecture via characterization of fabrication faults caused due to particular cell displacement in the QCA logic circuit. One of the worst kind of fault to remove is input generated transient fault, called single event upset (SEU). To show the unpredictability of QCA circuits under such faults, the correct input vectors are filtered beyond which the circuit behaves abnormally. The consequence of errors on the circuit is then analysed and test vector(s) is/are obtained to identify such faults. A prohibitive logic circuit is designed which allows just the right inputs and/or equivalent inputs to the faulty inputs. SEU transient fault can be avoided by the proposed prohibitive logic circuit to make an initiative towards the green computing.

PPG Based Continuous Blood Pressure Monitoring Framework for Smart Home Environment

2020 IEEE 6th World Forum on Internet of Things (WF-IoT), 2020

In this paper, we have proposed a computational framework for continuous blood pressure estimatio... more In this paper, we have proposed a computational framework for continuous blood pressure estimation using Photoplethysmogram (PPG) signal. The proposed framework is evaluated on the publicly available MIMIC Database. The database contains raw PPG data for different users and also the Arterial Blood Pressure (ABP) for calculating the systolic and diastolic blood pressure. Results showed that the Decision Tree Regressor boosted by Adaboost regressor could estimate systolic blood pressure with a mean average error of 2.07 and a standard deviation of 5.97 and the diastolic blood pressure with a mean average error of 1.15 and a standard deviation of 4.05. The results indicate that the proposed framework is an ideal candidate for integrating with the wearable devices for unobtrusive cuff less continuous blood pressure monitoring to provide real time update on the blood pressure of an individual with high degree of accuracy.

IoT based indoor location detection system for smart home environment

2018 IEEE International Conference on Consumer Electronics (ICCE), 2018

Smart home environment is expected to meet the requirements, essentially for the aging population... more Smart home environment is expected to meet the requirements, essentially for the aging population, to support the concept of “aging in place”, to provide reliable care and to ensure safety and proper diagnosis by keeping track of daily living, medical condition of the resident and providing feedback to the caregiver. In order to meet these requirements, smart home of today should support a number of functionalities. One such functionality of a smart home environment is the location detection. In this work, we propose a voice based location detection system which can be integrated in a smart home environment. Our location detection system uses Amazon Echo as the voice interface and HC-SR04 ultrasonic sensor to detect location of specific patients. The proposed location detection will be suitable for large scale application where we may need to keep track of multiple patients. Moreover, the inclusion of voice enabled feature to this system will reduce the burden of learning curve of new technologies on family and caregivers thereby improving the quality of life.

Wearable Health Monitoring System for Older Adults in a Smart Home Environment

2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI), 2021

The advent of IoT has enabled the design of connected and integrated smart health monitoring syst... more The advent of IoT has enabled the design of connected and integrated smart health monitoring systems. These smart health monitoring systems could be realized in a smart home context to render long-term care to the elderly population. In this paper, we present the design of a wearable health monitoring system suitable for older adults in a smart home context. The proposed system offers solutions to monitor the stress, blood pressure, and location of an individual within a smart home environment. The stress detection model proposed in this work uses Electrodermal Activity (EDA), Photoplethysmogram (PPG), and Skin Temperature (ST) sensors embedded in a smart wristband for detecting physiological stress. The stress detection model is trained and tested using stress labels obtained from salivary cortisol which is a clinically established biomarker for physiological stress. A voice-based prototype is also implemented and the feasibility of the proposed system for integration in a smart home environment is analyzed by simulating a data acquisition and streaming scenario. We have also proposed a blood pressure estimation model using PPG signal and advanced regression techniques for integration with the stress detection model in the wearable health monitoring system. Finally, the design of a voice-assisted indoor location system is proposed for integration with the proposed system within a smart home environment. The proposed wearable health monitoring system is an important direction to realize a smart home environment with extensive diagnostic capabilities so that such a system could be useful for rendering long-term and personalized care to the aging population in the comfort of their home.

Machine Learning Based Prediction of Future Stress Events in a Driving Scenario

2021 IEEE 7th World Forum on Internet of Things (WF-IoT), 2021

This paper presents a model for predicting a driver's stress level up to one minute in advance. S... more This paper presents a model for predicting a driver's stress level up to one minute in advance. Successfully predicting future stress would allow stress mitigation to begin before the subject becomes stressed, reducing or possibly avoiding the performance penalties of stress. The proposed model takes features extracted from Galvanic Skin Response (GSR) signals on the foot and hand and Respiration and Electrocardiogram (ECG) signals from the chest of the driver. The data used to train the model was retrieved from an existing database and then processed to create statistical and frequency features. A total of 42 features were extracted from the data and then expanded into a total of 252 features by grouping the data and taking six statistical measurements of each group for each feature. A Random Forest Classifier was trained and evaluated using a leave-one-subject-out testing approach. The model achieved 94% average accuracy on the test data. Results indicate that the model performs well and could be used as part of a vehicle stress prevention system.

Machine Learning-Based Anxiety Detection in Older Adults Using Wristband Sensors and Context Feature

SN Computer Science, 2021

This paper explores a novel method for anxiety detection in older adults using simple wristband s... more This paper explores a novel method for anxiety detection in older adults using simple wristband sensors such as electrodermal activity (EDA) and photoplethysmogram (PPG) and a context-based feature. The proposed method for anxiety detection combines features from a single physiological signal with an experimental context-based feature to improve the performance of the anxiety detection model. The experimental data for this work are obtained from a year-long experiment on 41 healthy older adults (26 females and 15 males) in the age range 60-80 with mean age 73.36 ± 5.25 during a Trier Social Stress Test (TSST) protocol. The anxiety level ground truth was obtained from State-Trait Anxiety Inventory (STAI), which is regarded as the gold standard to measure perceived anxiety. EDA and blood volume pulse (BVP) signals were recorded using a wristworn EDA and PPG sensor, respectively. 47 features were computed from EDA and BVP signal, out of which a final set of 24 significantly correlated features were selected for analysis. The phases of the experimental study are encoded as unique integers to generate the context feature vector. A combination of features from a single sensor with the context feature vector is used for training a machine learning model to distinguish between anxious and not-anxious states. Results and analysis showed that the EDA and BVP machine learning models that combined the context feature along with the physiological features achieved 3.37% and 6.41% higher accuracy, respectively, than the models that used only physiological features. Further, end-to-end processing of EDA and BVP signals was simulated for real-time anxiety level detection. This work demonstrates the practicality of the proposed anxiety detection method in facilitating long-term monitoring of anxiety in older adults using low-cost consumer devices.

Towards Photoplethysmogram Based Non-Invasive Blood Pressure Classification

2018 IEEE International Symposium on Smart Electronic Systems (iSES) (Formerly iNiS), 2018

A novel blood pressure classification model using Phototplethysmogram (PPG) is proposed in this w... more A novel blood pressure classification model using Phototplethysmogram (PPG) is proposed in this work. The proposed model uses signal processing and machine learning algorithms to classify blood pressure in four stages: normal, elevated, stage 1 and stage 2. A total of 83 features were extracted from the PPG signal which includes 71 statistical features and 12 characteristic features. We have used random forest classifier to train and test our predictive model. The proposed method is evaluated on publicly available MIMIC database for 20 different individuals. The database contains raw PPG data for different users and Arterial Blood Pressure (ABP) to calculate the systolic and diastolic blood pressure to be used as the ground truth for training and validation purposes. We have achieved an overall accuracy of 90.8% over the four classes of blood pressure levels. The results indicate that the proposed model will be ideal for integration into a non-invasive blood pressure monitoring system with significant accuracy.

Quantum Annealing for Automated Feature Selection in Stress Detection

2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI), 2021

We present a novel methodology for automated feature subset selection from a pool of physiologica... more We present a novel methodology for automated feature subset selection from a pool of physiological signals using Quantum Annealing (QA). As a case study, we will investigate the effectiveness of QA-based feature selection techniques in selecting the optimal feature subset for stress detection. Features are extracted from four signal sources: foot EDA, hand EDA, ECG, and respiration. The proposed method embeds the feature variables extracted from the physiological signals in a binary quadratic model. The bias of the feature variable is calculated using the Pearson correlation coefficient between the feature variable and the target variable. The weight of the edge connecting the two feature variables is calculated using the Pearson correlation coefficient between two feature variables in the binary quadratic model. Subsequently, D-Wave's clique sampler is used to sample cliques from the binary quadratic model. The underlying solution is then re-sampled to obtain multiple good solutions and the clique with the lowest energy is returned as the optimal solution. The proposed method is compared with commonly used feature selection techniques for stress detection. Results indicate that QA-based feature subset selection performed equally as that of classical techniques. However, under data uncertainty conditions such as limited training data, the performance of quantum annealing for selecting optimum features remained unaffected, whereas a significant decrease in performance is observed with classical feature selection techniques. Preliminary results show the promise of quantum annealing in optimizing the training phase of a machine learning classifier, especially under data uncertainty conditions.

Validating Physiological Stress Detection Model Using Cortisol as Stress Bio Marker

2020 IEEE International Conference on Consumer Electronics (ICCE), 2020

In this work, we have presented the validation of a stress detection model using cortisol as the ... more In this work, we have presented the validation of a stress detection model using cortisol as the stress biomarker. The proposed model uses two physiological signals: Galvanic Skin Response (GSR) and Photoplethysmograph (PPG) to classify stress into two levels. GSR and PPG signals were collected from a total of 13 participants along with saliva samples taken at time points throughout the duration of the experiment. We have used 10 out of the 13 participants to train our model. Data from the remaining 3 participants was used to test the robustness of the model in distinguishing stressed states from non-stressed states. We have achieved an overall accuracy of 92% with the model achieving precision, recall and f1-score of 93%, 99% and 96% respectively in predicting the occurrences of stressful events. Results indicate the promise of the proposed methodology in accurately detecting the presence of stressful events by generalizing the test data coming from a subset of population in contrast to the training data.

A Review of Machine Learning Classification Using Quantum Annealing for Real-World Applications

SN Computer Science, 2021

Optimizing the training of a machine learning pipeline helps in reducing training costs and impro... more Optimizing the training of a machine learning pipeline helps in reducing training costs and improving model performance. One such optimizing strategy is quantum annealing, which is an emerging computing paradigm that has shown potential in optimizing the training of a machine learning model. The implementation of a physical quantum annealer has been realized by D-wave systems and is available to the research community for experiments. Recent experimental results on a variety of machine learning applications using quantum annealing have shown interesting results where the performance of classical machine learning techniques is limited by limited training data and high dimensional features. This article explores the application of D-wave's quantum annealer for optimizing machine learning pipelines for real-world classification problems. We review the application domains on which a physical quantum annealer has been used to train machine learning classifiers. We discuss and analyze the experiments performed on the D-Wave quantum annealer for applications such as image recognition, remote sensing imagery, computational biology, and particle physics. We discuss the possible advantages and the problems for which quantum annealing is likely to be advantageous over classical computation.

Smart Wristband-Based Stress Detection Framework for Older Adults With Cortisol as Stress Biomarker

IEEE Transactions on Consumer Electronics, 2021

In this work, our objective is to design, develop, and evaluate the effectiveness of a stress det... more In this work, our objective is to design, develop, and evaluate the effectiveness of a stress detection model for older adults using a system of wrist-worn sensors. Our system uses four signals, EDA, BVP, IBI, and ST from EDA, PPG, and ST sensors, embedded in a smart wristband, to classify between stressed and not-stressed state. The stress reference is obtained from salivary cortisol measurement, which is a well established clinical biomarker for measuring physiological stress. This work is the result of year-long data collection and analysis of 40 older adults (28 females and 12 males) and age 73.625 ± 5.39. EDA, BVP, IBI, and ST signals were collected during TSST (Trier Social Stress Test), which is a well known experimental protocol to reliably induce stress in humans in a social setting. 47 features were extracted from EDA, BVP, IBI, and ST signals, out of which 27 features were selected using a supervised feature selection method. Results and analysis show that combining the features from all the four signal streams increases the model’s ability to accurately distinguish between the stressed and not-stressed states. The proposed model achieved a macro-average F1-score of 0.92 and an accuracy of 94% in distinguishing between the two states when features from all the four signals were used. Further, we prototype the proposed stress detection model in a consumer end device with voice capabilities, so that users can receive feedback on their vitals and stress levels by querying on voice-enabled consumer devices such as smartphones and smart speakers.

Machine Learning Based Stress Monitoring in Older Adults Using Wearable Sensors and Cortisol as Stress Biomarker

Journal of Signal Processing Systems, 2021

The objective of this work is to evaluate the effectiveness of a wearable physiological stress mo... more The objective of this work is to evaluate the effectiveness of a wearable physiological stress monitoring system in distinguishing between stressed and non-stressed state in older adults using machine learning techniques. This system utilizes EDA and BVP signal to detect occurrence of stress as indicated by salivary cortisol measurement which is a reliable objective measure of physiological stress. Data of 19 healthy older adults (11 female and 8 male) with mean age 73.15 ± 5.79 were used for this study. EDA and BVP signals were recorded using a finger tip sensor during the Trier Social Stress Test, which is a well known experimental protocol to reliably induce stress in humans in a social setting. 39 statistical measures of the peak characteristic of EDA and BVP signal were extracted. A supervised feature selection algorithm is used to select important features as an input to the machine learning model. Four machine learning algorithms were evaluated based on their performance in classifying between stressed and non-stressed states. Results indicate that the logistic regression performed the best among Random Forest, κ-NN, and Support Vector Machine achieving an macro-average and micro-average f1-score of 0.87 and 0.95 respectively and an AUC score of 0.81. We also evaluated the effectiveness of a novel deep learning Long Short-Term Memory (LSTM) based classifier in distinguishing between stressed and non-stressed state. Results on test data shows that LSTM based classifier achieved an improvement of 6.7% and 2% in terms of macro-average f1-score and micro-average f1-score respectively. Also the AUC score for LSTM classifier is found to be 0.9 which is about 11% higher than the best performing logistic regression model. This work can be used to design a convenient unobtrusive wearable device to monitor stress levels in older adults in their home environment, thereby facilitating aging in place and improving the quality of life.

Machine Learning Based Solutions for Real-Time Stress Monitoring

IEEE Consumer Electronics Magazine, 2020

Stress may be defined as the reaction of the body to regulate itself to changes within the enviro... more Stress may be defined as the reaction of the body to regulate itself to changes within the environment through mental, physical, or emotional responses. Recurrent episodes of acute stress can disturb the physical and mental stability of a person. This subsequently can have a negative effect on work performance and in the long term can increase the risk of physiological disorders like hypertension and psychological illness such as anxiety disorder. Psychological stress is a growing concern for the worldwide population across all age groups. A reliable, costefficient, acute stress detection system could enable its users to better monitor and manage their stress to mitigate its long-term negative effects. In this article, we will review and discuss the literature that has used machine learning based approaches for stress detection. We will also review the existing solutions in the literature that have leveraged the concept of edge computing in providing a potential solution in real-time monitoring of stress.

Productivity enhancement by component-oriented line balancing method

International Journal of Industrial and Systems Engineering, 2018

The problem of balancing a production line has been one of the typical complications in the field... more The problem of balancing a production line has been one of the typical complications in the field of industrial engineering. An unbalanced production line associates with the downfall in productivity. Since productivity is the 'primus motor' of income in any firm, the above problem cannot be over sighted. In this paper a similar problem is being faced by a company which is unable to cope with the increasing demands of their product. The past data of component manufacturing process, resources required and the time taken is collected, when this data is analysed, constraints are observed in the testing machine of the production line. The elimination of bottleneck in this line is needed in order to escalate the productivity. The intended rate of production is achieved through component-oriented line balancing (COLB) approach. In this method the number of workstations is increased and hence bottle neck is removed. The comparison of the data before and after the implementation of component-oriented line balancing shows a simultaneous improvement in the productivity and reduction of labour cost.

Smart Home Environment for Mild Cognitive Impairment Population: Solutions to Improve Care and Quality of Life

IEEE Consumer Electronics Magazine, 2018

Solutions to improve care and quality of life.

Towards Designing Reliable Universal QCA Logic in the Presence of Cell Deposition Defect

2016 29th International Conference on VLSI Design and 2016 15th International Conference on Embedded Systems (VLSID), 2016

This work targets design of a reliable universal logic gate in Quantum-dot cellular automata (r-U... more This work targets design of a reliable universal logic gate in Quantum-dot cellular automata (r-ULG with the single clock zone, r-ULG-II with multiple clock zone) with hybrid cell orientations that realizes majority and minority functions simultaneously with high fault tolerance. The characterization of the defective behaviour of r-ULGs under different cell deposition defects is investigated. The outcomes indicate that the proposed r-ULG provide 75% fault tolerance under single clock zone and 100% under two clock zone.

On the reliability of majority logic structure in quantum-dot cellular automata

Microelectronics Journal, 2016

Quantum-dot cellular automata (QCA) is projected to be a promising nanotechnology due to its extr... more Quantum-dot cellular automata (QCA) is projected to be a promising nanotechnology due to its extremely small feature size and ultra low power consumption. However, acceptance of a QCA design is limited due to its high defect rate. Efficient fault tolerant schemes are, therefore, needed for reliable design. This work targets design of a new fault tolerant scheme around QCA logic primitives which encapsulates two different orientations of QCA cell. A 2 Â 2 array of four rotated (' þ') cells, called complementary tile (CT), is introduced to maximize the throughput. It ensures 100% fault tolerance under single cell missing defect. Two reliable majority voters (RMV), based on the CT, are designed which outperforms the existing majority logic in QCA. The functional characterization and polarization of RMV under different cell deposition (missing/additional) defects are covered. The significance of the clocking in fault tolerance is also investigated with RMV with multi clock zone. The error probability model for the proposed RMV, under cell deposition (missing/additional) defect, is developed to ensure better understanding of reliability in QCA.

Towards the design of hybrid QCA tiles targeting high fault tolerance

Journal of Computational Electronics, 2015

The increasing fabrication cost of CMOS-based computing devices and the ever-approaching limits o... more The increasing fabrication cost of CMOS-based computing devices and the ever-approaching limits of their fabrication have led to the search for feasible options with high device density and low power waste. Quantum-dot cellular automata (QCA) is an emerging technology with such potential to match the design target beyond the limits of state-of-the-art CMOS. But nanotechnologies, like QCA are extremely susceptible to various forms of flaws and variations during fabrication at nano scale. Thus, the exploration of ingenious fault tolerant structure around QCA is gaining high importance. This work targets a new robust QCA tile structure hybridizing rotated and non-rotated cell together resulting lesser kink energy. Different QCA logic primitives (majority/minority logic, fanout tiles, etc.) are made using such hybrid cell structure. The functional characterization using the kink energy and the polarization level of such QCA structures under different cell defects have been thoroughly investigated. The results suggest that the proposed QCA logic primitives have achieved high fault tolerance of 97.43 %. Also, 100 % fault tolerance can be ascertained if the proposed logic circuit drives the correct output with the application of 001, 011 as a primitive test vector only. A comparative per-B Bibhash Sen

Fault masking in Quantum-dot cellular automata using prohibitive logic circuit

2015 19th International Symposium on VLSI Design and Test, 2015

This work targets the challenges to design with uncertainty in QCA architecture via characterizat... more This work targets the challenges to design with uncertainty in QCA architecture via characterization of fabrication faults caused due to particular cell displacement in the QCA logic circuit. One of the worst kind of fault to remove is input generated transient fault, called single event upset (SEU). To show the unpredictability of QCA circuits under such faults, the correct input vectors are filtered beyond which the circuit behaves abnormally. The consequence of errors on the circuit is then analysed and test vector(s) is/are obtained to identify such faults. A prohibitive logic circuit is designed which allows just the right inputs and/or equivalent inputs to the faulty inputs. SEU transient fault can be avoided by the proposed prohibitive logic circuit to make an initiative towards the green computing.

PPG Based Continuous Blood Pressure Monitoring Framework for Smart Home Environment

2020 IEEE 6th World Forum on Internet of Things (WF-IoT), 2020

In this paper, we have proposed a computational framework for continuous blood pressure estimatio... more In this paper, we have proposed a computational framework for continuous blood pressure estimation using Photoplethysmogram (PPG) signal. The proposed framework is evaluated on the publicly available MIMIC Database. The database contains raw PPG data for different users and also the Arterial Blood Pressure (ABP) for calculating the systolic and diastolic blood pressure. Results showed that the Decision Tree Regressor boosted by Adaboost regressor could estimate systolic blood pressure with a mean average error of 2.07 and a standard deviation of 5.97 and the diastolic blood pressure with a mean average error of 1.15 and a standard deviation of 4.05. The results indicate that the proposed framework is an ideal candidate for integrating with the wearable devices for unobtrusive cuff less continuous blood pressure monitoring to provide real time update on the blood pressure of an individual with high degree of accuracy.

IoT based indoor location detection system for smart home environment

2018 IEEE International Conference on Consumer Electronics (ICCE), 2018

Smart home environment is expected to meet the requirements, essentially for the aging population... more Smart home environment is expected to meet the requirements, essentially for the aging population, to support the concept of “aging in place”, to provide reliable care and to ensure safety and proper diagnosis by keeping track of daily living, medical condition of the resident and providing feedback to the caregiver. In order to meet these requirements, smart home of today should support a number of functionalities. One such functionality of a smart home environment is the location detection. In this work, we propose a voice based location detection system which can be integrated in a smart home environment. Our location detection system uses Amazon Echo as the voice interface and HC-SR04 ultrasonic sensor to detect location of specific patients. The proposed location detection will be suitable for large scale application where we may need to keep track of multiple patients. Moreover, the inclusion of voice enabled feature to this system will reduce the burden of learning curve of new technologies on family and caregivers thereby improving the quality of life.