-
A Novel 1D Generative Adversarial Network-based Framework for Atrial Fibrillation Detection using Restored Wrist Photoplethysmography Signals
Authors:
Faizul Rakib Sayem,
Mosabber Uddin Ahmed,
Saadia Binte Alam,
Sakib Mahmud,
Md. Mamun Sheikh,
Abdulrahman Alqahtani,
Md Ahasan Atick Faisal,
Muhammad E. H. Chowdhury
Abstract:
Atrial fibrillation (AF) increases the risk of stroke. Electrocardiogram (ECG) is used for AF detection, while photoplethysmography (PPG) is simple to use and appropriate for long-term monitoring. We have developed a novel approach to detect AF from smartwatch-based wrist PPG signals. To the best of our knowledge, this is the first study to employ 1D CycleGAN for the reconstruction of 1D wrist PPG…
▽ More
Atrial fibrillation (AF) increases the risk of stroke. Electrocardiogram (ECG) is used for AF detection, while photoplethysmography (PPG) is simple to use and appropriate for long-term monitoring. We have developed a novel approach to detect AF from smartwatch-based wrist PPG signals. To the best of our knowledge, this is the first study to employ 1D CycleGAN for the reconstruction of 1D wrist PPG signals, since the quality of wrist PPG signals is very poor due to motion artifacts and acquisition site limitations. Our proposed approach is validated on a dataset of 21,278 10s long wrist PPG segments. The dataset was divided into two halves and two experiments were conducted. One half was used for training and the other half for testing and vice versa. Our classification model (Self-AFNet) with 1D-CycleGAN for restoration achieves an accuracy of 96.41% and 97.09% for both splits, respectively. There is a noteworthy improvement in accuracy (2.94% and 5.08% for test splits, respectively) for restored signals as compared to unrestored PPG. AF detection task was also performed using ECG signals with the matched PPG signals to assess the validity of employing reconstructed PPG-based AF classification. Self-AFNet managed to achieve an accuracy of 98.07% and 98.97%, respectively using two ECG splits which is comparable to the performance of AF detection utilizing reconstructed PPG segments. This study reflects that reconstructed wrist PPG signals acquired from wearable devices can be used for reliable detection of atrial fibrillation, which can significantly help in the reduction of the risk of stroke.
△ Less
Submitted 23 December, 2023; v1 submitted 13 November, 2023;
originally announced December 2023.
-
Hurricane and Storm Surges-Induced Power System Vulnerabilities and their Socioeconomic Impact
Authors:
Abodh Poudyal,
Shishir Lamichhane,
Charlotte Wertz,
Sajjad Uddin Mahmud,
Anamika Dubey
Abstract:
This paper introduces a probabilistic framework to quantify community vulnerability towards power losses due to extreme weather events. To analyze the impact of weather events on the power grid, the wind fields of historical hurricanes from 2000 to 2018 on the Texas coast are modeled using their available parameters, and probabilistic storm surge scenarios are constructed utilizing the hurricane c…
▽ More
This paper introduces a probabilistic framework to quantify community vulnerability towards power losses due to extreme weather events. To analyze the impact of weather events on the power grid, the wind fields of historical hurricanes from 2000 to 2018 on the Texas coast are modeled using their available parameters, and probabilistic storm surge scenarios are constructed utilizing the hurricane characteristics. The vulnerability of hurricanes and storm surges is evaluated on a 2000 bus synthetic power grid model on the geographical footprint of Texas. The load losses, obtained via branch and substation outages, are then geographically represented at the county level and integrated with the publicly available Social Vulnerability Index to evaluate the Integrated Community Vulnerability Index (ICVI), which reflects the impacts of these extreme weather events on the socioeconomic and community power systems. The analysis concludes that the compounded impact of power outages due to extreme weather events can amplify the vulnerability of affected communities. Such analysis can help the system planners and operators make an informed decision.
△ Less
Submitted 25 November, 2023;
originally announced November 2023.
-
Physics-Enhanced TinyML for Real-Time Detection of Ground Magnetic Anomalies
Authors:
Talha Siddique,
MD Shaad Mahmud
Abstract:
Space weather phenomena like geomagnetic disturbances (GMDs) and geomagnetically induced currents (GICs) pose significant risks to critical technological infrastructure. While traditional predictive models, grounded in simulation, hold theoretical robustness, they grapple with challenges, notably the assimilation of imprecise data and extensive computational complexities. In recent years, Tiny Mac…
▽ More
Space weather phenomena like geomagnetic disturbances (GMDs) and geomagnetically induced currents (GICs) pose significant risks to critical technological infrastructure. While traditional predictive models, grounded in simulation, hold theoretical robustness, they grapple with challenges, notably the assimilation of imprecise data and extensive computational complexities. In recent years, Tiny Machine Learning (TinyML) has been adopted to develop Machine Learning (ML)-enabled magnetometer systems for predicting real-time terrestrial magnetic perturbations as a proxy measure for GIC. While TinyML offers efficient, real-time data processing, its intrinsic limitations prevent the utilization of robust methods with high computational needs. This paper developed a physics-guided TinyML framework to address the above challenges. This framework integrates physics-based regularization at the stages of model training and compression, thereby augmenting the reliability of predictions. The developed pruning scheme within the framework harnesses the inherent physical characteristics of the domain, striking a balance between model size and robustness. The study presents empirical results, drawing a comprehensive comparison between the accuracy and reliability of the developed framework and its traditional counterpart. Such a comparative analysis underscores the prospective applicability of the developed framework in conceptualizing robust, ML-enabled magnetometer systems for real-time space weather forecasting.
△ Less
Submitted 19 November, 2023;
originally announced November 2023.
-
A Novel Deep Learning Technique for Morphology Preserved Fetal ECG Extraction from Mother ECG using 1D-CycleGAN
Authors:
Promit Basak,
A. H. M Nazmus Sakib,
Muhammad E. H. Chowdhury,
Nasser Al-Emadi,
Huseyin Cagatay Yalcin,
Shona Pedersen,
Sakib Mahmud,
Serkan Kiranyaz,
Somaya Al-Maadeed
Abstract:
Monitoring the electrical pulse of fetal heart through a non-invasive fetal electrocardiogram (fECG) can easily detect abnormalities in the develo** heart to significantly reduce the infant mortality rate and post-natal complications. Due to the overlap** of maternal and fetal R-peaks, the low amplitude of the fECG, systematic and ambient noises, typical signal extraction methods, such as adap…
▽ More
Monitoring the electrical pulse of fetal heart through a non-invasive fetal electrocardiogram (fECG) can easily detect abnormalities in the develo** heart to significantly reduce the infant mortality rate and post-natal complications. Due to the overlap** of maternal and fetal R-peaks, the low amplitude of the fECG, systematic and ambient noises, typical signal extraction methods, such as adaptive filters, independent component analysis, empirical mode decomposition, etc., are unable to produce satisfactory fECG. While some techniques can produce accurate QRS waves, they often ignore other important aspects of the ECG. Our approach, which is based on 1D CycleGAN, can reconstruct the fECG signal from the mECG signal while maintaining the morphology due to extensive preprocessing and appropriate framework. The performance of our solution was evaluated by combining two available datasets from Physionet, "Abdominal and Direct Fetal ECG Database" and "Fetal electrocardiograms, direct and abdominal with reference heartbeat annotations", where it achieved an average PCC and Spectral-Correlation score of 88.4% and 89.4%, respectively. It detects the fQRS of the signal with accuracy, precision, recall and F1 score of 92.6%, 97.6%, 94.8% and 96.4%, respectively. It can also accurately produce the estimation of fetal heart rate and R-R interval with an error of 0.25% and 0.27%, respectively. The main contribution of our work is that, unlike similar studies, it can retain the morphology of the ECG signal with high fidelity. The accuracy of our solution for fetal heart rate and R-R interval length is comparable to existing state-of-the-art techniques. This makes it a highly effective tool for early diagnosis of fetal heart diseases and regular health checkups of the fetus.
△ Less
Submitted 25 September, 2023;
originally announced October 2023.
-
Spatiotemporal Impact Analysis of Hurricanes and Storm Surges on Power Systems
Authors:
Abodh Poudyal,
Charlotte Wertz,
Amy Mi Nguyen,
Sajjad Uddin Mahmud,
Vibha Gunturi,
Anamika Dubey
Abstract:
This paper develops a spatiotemporal probabilistic impact assessment framework to analyze and quantify the compounding effect of hurricanes and storm surges on the bulk power grid. The probabilistic synthetic hurricane tracks are generated using historical hurricane data, and storm surge scenarios are generated based on observed hurricane parameters. The system losses are modeled using a loss metr…
▽ More
This paper develops a spatiotemporal probabilistic impact assessment framework to analyze and quantify the compounding effect of hurricanes and storm surges on the bulk power grid. The probabilistic synthetic hurricane tracks are generated using historical hurricane data, and storm surge scenarios are generated based on observed hurricane parameters. The system losses are modeled using a loss metric that quantifies the total load loss. The overall simulation is performed on the synthetic Texas 2000-bus system mapped on the geographical footprint of Texas. The results show that power substation inundation due to storm surge creates additional load losses as the hurricane traverses inland.
△ Less
Submitted 24 November, 2022;
originally announced December 2022.
-
Output Feedback Adaptive Optimal Control of Affine Nonlinear systems with a Linear Measurement Model
Authors:
Tochukwu Elijah Ogri,
S. M. Nahid Mahmud,
Zachary I. Bell,
Rushikesh Kamalapurkar
Abstract:
Real-world control applications in complex and uncertain environments require adaptability to handle model uncertainties and robustness against disturbances. This paper presents an online, output-feedback, critic-only, model-based reinforcement learning architecture that simultaneously learns and implements an optimal controller while maintaining stability during the learning phase. Using multipli…
▽ More
Real-world control applications in complex and uncertain environments require adaptability to handle model uncertainties and robustness against disturbances. This paper presents an online, output-feedback, critic-only, model-based reinforcement learning architecture that simultaneously learns and implements an optimal controller while maintaining stability during the learning phase. Using multiplier matrices, a convenient way to search for observer gains is designed along with a controller that learns from simulated experience to ensure stability and convergence of trajectories of the closed-loop system to a neighborhood of the origin. Local uniform ultimate boundedness of the trajectories is established using a Lyapunov-based analysis and demonstrated through simulation results, under mild excitation conditions.
△ Less
Submitted 3 April, 2023; v1 submitted 12 October, 2022;
originally announced October 2022.
-
Real-Time Machine Learning Enabled Low-Cost Magnetometer System
Authors:
Talha Siddique,
Md. Shaad Mahmud
Abstract:
Geomagnetically Induced Currents (GICs) are one of the most hazardous effects of space weather. The rate of change in ground horizontal magnetic component dBH/dt is used as a proxy measure for GIC. In order to monitor and predict dBH/dt, ground-based fluxgate magnetometers are used. However, baseline correction is crucial before such magnetometer data can be utilized. In this paper, a low-cost Mac…
▽ More
Geomagnetically Induced Currents (GICs) are one of the most hazardous effects of space weather. The rate of change in ground horizontal magnetic component dBH/dt is used as a proxy measure for GIC. In order to monitor and predict dBH/dt, ground-based fluxgate magnetometers are used. However, baseline correction is crucial before such magnetometer data can be utilized. In this paper, a low-cost Machine Learning (ML) enabled magnetometer system has been implemented to perform realtime baseline correction of magnetometer data. The predicted geomagnetic components are then used to derive a forecast for dBH/dt. Two different ML models were deployed, and their real-time and offline prediction accuracy were examined. The localized peaks of the predicted dBH/dt are further validated using binary event analysis.
△ Less
Submitted 16 September, 2022;
originally announced September 2022.
-
Design, Implementation, and Performance Evaluation of a Fiber Bragg Gratings(FBG) based Smart Insole to Measure Plantar Pressure and Temperature
Authors:
Sakib Mahmud,
Amith Khandakar,
Muhammad E. H. Chowdhury,
Mohammed AbdulMoniem,
Mamun Bin Ibne Reaz,
Zaid Bin Mahbub,
Kishor Kumar Sadasivuni,
M Murugappan,
Mohammed Alhatou
Abstract:
Various foot complications can be easily avoided by continuous monitoring of foot plantar temperature and pressure. In this work, the design, characterization, and implementation of a Fiber Bragg Gratings (FBG) based smart insole capable of simultaneously measuring plantar pressure and temperature has been reported. The instrumented insole was tested and verified during static and gait exercises.…
▽ More
Various foot complications can be easily avoided by continuous monitoring of foot plantar temperature and pressure. In this work, the design, characterization, and implementation of a Fiber Bragg Gratings (FBG) based smart insole capable of simultaneously measuring plantar pressure and temperature has been reported. The instrumented insole was tested and verified during static and gait exercises. The paper also provides a comparison of the developed optoelectronic-based solution with a commercially available and widely used plantar pressure measurement and analysis system and a lab-made, electronic sensor-based smart insole that can measure both plantar temperature and pressure. It was shown that even though the commercially available expensive system is very robust and highly precise due to many sensing units on the insole, the developed insole with a smaller number of sensors can provide both plantar temperature and pressure with reasonable precision while displaying both foot pressure and temperature maps, and gait cycle plots in real-time. The FBG-based solution is comfortable and safer than the other systems being compared. Thus, confirming the effectiveness of the proposed solution as an addition to the research area for detecting foot complications non-invasively using foot insoles.
△ Less
Submitted 25 August, 2022;
originally announced August 2022.
-
Safe Controller for Output Feedback Linear Systems using Model-Based Reinforcement Learning
Authors:
S M Nahid Mahmud,
Moad Abudia,
Scott A Nivison,
Zachary I. Bell,
Rushikesh Kamalapurkar
Abstract:
The objective of this research is to enable safety-critical systems to simultaneously learn and execute optimal control policies in a safe manner to achieve complex autonomy. Learning optimal policies via trial and error, i.e., traditional reinforcement learning, is difficult to implement in safety-critical systems, particularly when task restarts are unavailable. Safe model-based reinforcement le…
▽ More
The objective of this research is to enable safety-critical systems to simultaneously learn and execute optimal control policies in a safe manner to achieve complex autonomy. Learning optimal policies via trial and error, i.e., traditional reinforcement learning, is difficult to implement in safety-critical systems, particularly when task restarts are unavailable. Safe model-based reinforcement learning techniques based on a barrier transformation have recently been developed to address this problem. However, these methods rely on full state feedback, limiting their usability in a real-world environment. In this work, an output-feedback safe model-based reinforcement learning technique based on a novel barrier-aware dynamic state estimator has been designed to address this issue. The developed approach facilitates simultaneous learning and execution of safe control policies for safety-critical linear systems. Simulation results indicate that barrier transformation is an effective approach to achieve online reinforcement learning in safety-critical systems using output feedback.
△ Less
Submitted 4 April, 2022;
originally announced April 2022.
-
A Shallow U-Net Architecture for Reliably Predicting Blood Pressure (BP) from Photoplethysmogram (PPG) and Electrocardiogram (ECG) Signals
Authors:
Sakib Mahmud,
Nabil Ibtehaz,
Amith Khandakar,
Anas Tahir,
Tawsifur Rahman,
Khandaker Reajul Islam,
Md Shafayet Hossain,
M. Sohel Rahman,
Mohammad Tariqul Islam,
Muhammad E. H. Chowdhury
Abstract:
Cardiovascular diseases are the most common causes of death around the world. To detect and treat heart-related diseases, continuous Blood Pressure (BP) monitoring along with many other parameters are required. Several invasive and non-invasive methods have been developed for this purpose. Most existing methods used in the hospitals for continuous monitoring of BP are invasive. On the contrary, cu…
▽ More
Cardiovascular diseases are the most common causes of death around the world. To detect and treat heart-related diseases, continuous Blood Pressure (BP) monitoring along with many other parameters are required. Several invasive and non-invasive methods have been developed for this purpose. Most existing methods used in the hospitals for continuous monitoring of BP are invasive. On the contrary, cuff-based BP monitoring methods, which can predict Systolic Blood Pressure (SBP) and Diastolic Blood Pressure (DBP), cannot be used for continuous monitoring. Several studies attempted to predict BP from non-invasively collectible signals such as Photoplethysmogram (PPG) and Electrocardiogram (ECG), which can be used for continuous monitoring. In this study, we explored the applicability of autoencoders in predicting BP from PPG and ECG signals. The investigation was carried out on 12,000 instances of 942 patients of the MIMIC-II dataset and it was found that a very shallow, one-dimensional autoencoder can extract the relevant features to predict the SBP and DBP with the state-of-the-art performance on a very large dataset. Independent test set from a portion of the MIMIC-II dataset provides an MAE of 2.333 and 0.713 for SBP and DBP, respectively. On an external dataset of forty subjects, the model trained on the MIMIC-II dataset, provides an MAE of 2.728 and 1.166 for SBP and DBP, respectively. For both the cases, the results met British Hypertension Society (BHS) Grade A and surpassed the studies from the current literature.
△ Less
Submitted 12 November, 2021;
originally announced November 2021.
-
Safety aware model-based reinforcement learning for optimal control of a class of output-feedback nonlinear systems
Authors:
S M Nahid Mahmud,
Moad Abudia,
Scott A Nivison,
Zachary I. Bell,
Rushikesh Kamalapurkar
Abstract:
The ability to learn and execute optimal control policies safely is critical to realization of complex autonomy, especially where task restarts are not available and/or the systems are safety-critical. Safety requirements are often expressed in terms of state and/or control constraints. Methods such as barrier transformation and control barrier functions have been successfully used, in conjunction…
▽ More
The ability to learn and execute optimal control policies safely is critical to realization of complex autonomy, especially where task restarts are not available and/or the systems are safety-critical. Safety requirements are often expressed in terms of state and/or control constraints. Methods such as barrier transformation and control barrier functions have been successfully used, in conjunction with model-based reinforcement learning, for safe learning in systems under state constraints, to learn the optimal control policy. However, existing barrier-based safe learning methods rely on full state feedback. In this paper, an output-feedback safe model-based reinforcement learning technique is developed that utilizes a novel dynamic state estimator to implement simultaneous learning and control for a class of safety-critical systems with partially observable state.
△ Less
Submitted 1 October, 2021;
originally announced October 2021.
-
QUCoughScope: An Artificially Intelligent Mobile Application to Detect Asymptomatic COVID-19 Patients using Cough and Breathing Sounds
Authors:
Muhammad E. H. Chowdhury,
Nabil Ibtehaz,
Tawsifur Rahman,
Yosra Magdi Salih Mekki,
Yazan Qibalwey,
Sakib Mahmud,
Maymouna Ezeddin,
Susu Zughaier,
Sumaya Ali S A Al-Maadeed
Abstract:
In the break of COVID-19 pandemic, mass testing has become essential to reduce the spread of the virus. Several recent studies suggest that a significant number of COVID-19 patients display no physical symptoms whatsoever. Therefore, it is unlikely that these patients will undergo COVID-19 test, which increases their chances of unintentionally spreading the virus. Currently, the primary diagnostic…
▽ More
In the break of COVID-19 pandemic, mass testing has become essential to reduce the spread of the virus. Several recent studies suggest that a significant number of COVID-19 patients display no physical symptoms whatsoever. Therefore, it is unlikely that these patients will undergo COVID-19 test, which increases their chances of unintentionally spreading the virus. Currently, the primary diagnostic tool to detect COVID-19 is RT-PCR test on collected respiratory specimens from the suspected case. This requires patients to travel to a laboratory facility to be tested, thereby potentially infecting others along the way.It is evident from recent researches that asymptomatic COVID-19 patients cough and breath in a different way than the healthy people. Several research groups have created mobile and web-platform for crowdsourcing the symptoms, cough and breathing sounds from healthy, COVID-19 and Non-COVID patients. Some of these data repositories were made public. We have received such a repository from Cambridge University team under data-sharing agreement, where we have cough and breathing sound samples for 582 and 141 healthy and COVID-19 patients, respectively. 87 COVID-19 patients were asymptomatic, while rest of them have cough. We have developed an Android application to automatically screen COVID-19 from the comfort of people homes. Test subjects can simply download a mobile application, enter their symptoms, record an audio clip of their cough and breath, and upload the data anonymously to our servers. Our backend server converts the audio clip to spectrogram and then apply our state-of-the-art machine learning model to classify between cough sounds produced by COVID-19 patients, as opposed to healthy subjects or those with other respiratory conditions. The system can detect asymptomatic COVID-19 patients with a sensitivity more than 91%.
△ Less
Submitted 20 March, 2021;
originally announced March 2021.
-
COVID-19 Infection Localization and Severity Grading from Chest X-ray Images
Authors:
Anas M. Tahir,
Muhammad E. H. Chowdhury,
Amith Khandakar,
Tawsifur Rahman,
Yazan Qiblawey,
Uzair Khurshid,
Serkan Kiranyaz,
Nabil Ibtehaz,
M Shohel Rahman,
Somaya Al-Madeed,
Khaled Hameed,
Tahir Hamid,
Sakib Mahmud,
Maymouna Ezeddin
Abstract:
Coronavirus disease 2019 (COVID-19) has been the main agenda of the whole world, since it came into sight in December 2019 as it has significantly affected the world economy and healthcare system. Given the effects of COVID-19 on pulmonary tissues, chest radiographic imaging has become a necessity for screening and monitoring the disease. Numerous studies have proposed Deep Learning approaches for…
▽ More
Coronavirus disease 2019 (COVID-19) has been the main agenda of the whole world, since it came into sight in December 2019 as it has significantly affected the world economy and healthcare system. Given the effects of COVID-19 on pulmonary tissues, chest radiographic imaging has become a necessity for screening and monitoring the disease. Numerous studies have proposed Deep Learning approaches for the automatic diagnosis of COVID-19. Although these methods achieved astonishing performance in detection, they have used limited chest X-ray (CXR) repositories for evaluation, usually with a few hundred COVID-19 CXR images only. Thus, such data scarcity prevents reliable evaluation with the potential of overfitting. In addition, most studies showed no or limited capability in infection localization and severity grading of COVID-19 pneumonia. In this study, we address this urgent need by proposing a systematic and unified approach for lung segmentation and COVID-19 localization with infection quantification from CXR images. To accomplish this, we have constructed the largest benchmark dataset with 33,920 CXR images, including 11,956 COVID-19 samples, where the annotation of ground-truth lung segmentation masks is performed on CXRs by a novel human-machine collaborative approach. An extensive set of experiments was performed using the state-of-the-art segmentation networks, U-Net, U-Net++, and Feature Pyramid Networks (FPN). The developed network, after an extensive iterative process, reached a superior performance for lung region segmentation with Intersection over Union (IoU) of 96.11% and Dice Similarity Coefficient (DSC) of 97.99%. Furthermore, COVID-19 infections of various shapes and types were reliably localized with 83.05% IoU and 88.21% DSC. Finally, the proposed approach has achieved an outstanding COVID-19 detection performance with both sensitivity and specificity values above 99%.
△ Less
Submitted 14 March, 2021;
originally announced March 2021.
-
Detection and severity classification of COVID-19 in CT images using deep learning
Authors:
Yazan Qiblawey,
Anas Tahir,
Muhammad E. H. Chowdhury,
Amith Khandakar,
Serkan Kiranyaz,
Tawsifur Rahman,
Nabil Ibtehaz,
Sakib Mahmud,
Somaya Al-Madeed,
Farayi Musharavati
Abstract:
Since the breakout of coronavirus disease (COVID-19), the computer-aided diagnosis has become a necessity to prevent the spread of the virus. Detecting COVID-19 at an early stage is essential to reduce the mortality risk of the patients. In this study, a cascaded system is proposed to segment the lung, detect, localize, and quantify COVID-19 infections from computed tomography (CT) images Furtherm…
▽ More
Since the breakout of coronavirus disease (COVID-19), the computer-aided diagnosis has become a necessity to prevent the spread of the virus. Detecting COVID-19 at an early stage is essential to reduce the mortality risk of the patients. In this study, a cascaded system is proposed to segment the lung, detect, localize, and quantify COVID-19 infections from computed tomography (CT) images Furthermore, the system classifies the severity of COVID-19 as mild, moderate, severe, or critical based on the percentage of infected lungs. An extensive set of experiments were performed using state-of-the-art deep Encoder-Decoder Convolutional Neural Networks (ED-CNNs), UNet, and Feature Pyramid Network (FPN), with different backbone (encoder) structures using the variants of DenseNet and ResNet. The conducted experiments showed the best performance for lung region segmentation with Dice Similarity Coefficient (DSC) of 97.19% and Intersection over Union (IoU) of 95.10% using U-Net model with the DenseNet 161 encoder. Furthermore, the proposed system achieved an elegant performance for COVID-19 infection segmentation with a DSC of 94.13% and IoU of 91.85% using the FPN model with the DenseNet201 encoder. The achieved performance is significantly superior to previous methods for COVID-19 lesion localization. Besides, the proposed system can reliably localize infection of various shapes and sizes, especially small infection regions, which are rarely considered in recent studies. Moreover, the proposed system achieved high COVID-19 detection performance with 99.64% sensitivity and 98.72% specificity. Finally, the system was able to discriminate between different severity levels of COVID-19 infection over a dataset of 1,110 subjects with sensitivity values of 98.3%, 71.2%, 77.8%, and 100% for mild, moderate, severe, and critical infections, respectively.
△ Less
Submitted 15 February, 2021;
originally announced February 2021.
-
Online inverse reinforcement learning with limited data
Authors:
Ryan Self,
S M Nahid Mahmud,
Katrine Hareland,
Rushikesh Kamalapurkar
Abstract:
This paper addresses the problem of online inverse reinforcement learning for systems with limited data and uncertain dynamics. In the developed approach, the state and control trajectories are recorded online by observing an agent perform a task, and reward function estimation is performed in real-time using a novel inverse reinforcement learning approach. Parameter estimation is performed concur…
▽ More
This paper addresses the problem of online inverse reinforcement learning for systems with limited data and uncertain dynamics. In the developed approach, the state and control trajectories are recorded online by observing an agent perform a task, and reward function estimation is performed in real-time using a novel inverse reinforcement learning approach. Parameter estimation is performed concurrently to help compensate for uncertainties in the agent's dynamics. Data insufficiency is resolved by develo** a data-driven update law to estimate the optimal feedback controller. The estimated controller can then be queried to artificially create additional data to drive reward function estimation.
△ Less
Submitted 18 August, 2020;
originally announced August 2020.
-
Safe Model-Based Reinforcement Learning for Systems with Parametric Uncertainties
Authors:
S M Nahid Mahmud,
Scott A Nivison,
Zachary I. Bell,
Rushikesh Kamalapurkar
Abstract:
Reinforcement learning has been established over the past decade as an effective tool to find optimal control policies for dynamical systems, with recent focus on approaches that guarantee safety during the learning and/or execution phases. In general, safety guarantees are critical in reinforcement learning when the system is safety-critical and/or task restarts are not practically feasible. In o…
▽ More
Reinforcement learning has been established over the past decade as an effective tool to find optimal control policies for dynamical systems, with recent focus on approaches that guarantee safety during the learning and/or execution phases. In general, safety guarantees are critical in reinforcement learning when the system is safety-critical and/or task restarts are not practically feasible. In optimal control theory, safety requirements are often expressed in terms of state and/or control constraints. In recent years, reinforcement learning approaches that rely on persistent excitation have been combined with a barrier transformation to learn the optimal control policies under state constraints. To soften the excitation requirements, model-based reinforcement learning methods that rely on exact model knowledge have also been integrated with the barrier transformation framework. The objective of this paper is to develop safe reinforcement learning method for deterministic nonlinear systems, with parametric uncertainties in the model, to learn approximate constrained optimal policies without relying on stringent excitation conditions. To that end, a model-based reinforcement learning technique that utilizes a novel filtered concurrent learning method, along with a barrier transformation, is developed in this paper to realize simultaneous learning of unknown model parameters and approximate optimal state-constrained control policies for safety-critical systems.
△ Less
Submitted 5 October, 2021; v1 submitted 24 July, 2020;
originally announced July 2020.
-
PPG2ABP: Translating Photoplethysmogram (PPG) Signals to Arterial Blood Pressure (ABP) Waveforms using Fully Convolutional Neural Networks
Authors:
Nabil Ibtehaz,
Sakib Mahmud,
Muhammad E. H. Chowdhury,
Amith Khandakar,
Mohamed Arselene Ayari,
Anas Tahir,
M. Sohel Rahman
Abstract:
Cardiovascular diseases are one of the most severe causes of mortality, taking a heavy toll of lives annually throughout the world. The continuous monitoring of blood pressure seems to be the most viable option, but this demands an invasive process, bringing about several layers of complexities. This motivates us to develop a method to predict the continuous arterial blood pressure (ABP) waveform…
▽ More
Cardiovascular diseases are one of the most severe causes of mortality, taking a heavy toll of lives annually throughout the world. The continuous monitoring of blood pressure seems to be the most viable option, but this demands an invasive process, bringing about several layers of complexities. This motivates us to develop a method to predict the continuous arterial blood pressure (ABP) waveform through a non-invasive approach using photoplethysmogram (PPG) signals. In addition we explore the advantage of deep learning as it would free us from sticking to ideally shaped PPG signals only, by making handcrafted feature computation irrelevant, which is a shortcoming of the existing approaches. Thus, we present, PPG2ABP, a deep learning based method, that manages to predict the continuous ABP waveform from the input PPG signal, with a mean absolute error of 4.604 mmHg, preserving the shape, magnitude and phase in unison. However, the more astounding success of PPG2ABP turns out to be that the computed values of DBP, MAP and SBP from the predicted ABP waveform outperforms the existing works under several metrics, despite that PPG2ABP is not explicitly trained to do so.
△ Less
Submitted 26 September, 2022; v1 submitted 4 May, 2020;
originally announced May 2020.
-
Representing Realistic Human Driver Behaviors using a Finite Size Gaussian Process Kernel Bank
Authors:
Hossein Nourkhiz Mahjoub,
Arash Raftari,
Rodolfo Valiente,
Yaser P. Fallah,
Syed K. Mahmud
Abstract:
The performance of cooperative vehicular applications is tightly dependent on the reliability of the underneath Vehicle-to-Everything (V2X) communication technology. V2X standards, such as Dedicated Short-Range Communications (DSRC) and Cellular-V2X (C-V2X), which are passing their research phase before being mandated in the US, are supposed to serve as reliable circulatory systems for the time-cr…
▽ More
The performance of cooperative vehicular applications is tightly dependent on the reliability of the underneath Vehicle-to-Everything (V2X) communication technology. V2X standards, such as Dedicated Short-Range Communications (DSRC) and Cellular-V2X (C-V2X), which are passing their research phase before being mandated in the US, are supposed to serve as reliable circulatory systems for the time-critical information in vehicular networks; however, they are still heavily suffering from scalability issues in real traffic scenarios. The technology-agnostic notion of Model-Based Communications (MBC) has been proposed in our previous works as a promising paradigm to address the scalability issue and its performance, while acquiring different modeling strategies, has been vastly studied. In this work, the modeling capabilities of a powerful non-parametric Bayesian inference scheme, i.e., Gaussian Processes (GPs), is investigated within the MBC context with more details. Our observations reveal an important potential strength of GP-based MBC scheme, i.e., its capability of accurately modeling different driving behavioral patterns by utilizing only a limited size GP kernel bank. This interesting aspect of integrating GP inference with MBC framework, which has been verified in this work using realistic driving data sets, introduces this architecture as a strong and appealing candidate to address the scalability challenge. The results confirm that our proposed approach over-performs the state of the art research in terms of the required communication rate and GP kernel bank size.
△ Less
Submitted 30 March, 2020;
originally announced March 2020.
-
SafeNet: An Assistive Solution to Assess Incoming Threats for Premises
Authors:
Shahinur Alam,
Md Sultan Mahmud,
Mohammed Yeasin
Abstract:
An assistive solution to assess incoming threats (e.g., robbery, burglary, gun violence) for homes will enhance the safety of the people with or without disabilities. This paper presents "SafeNet"- an integrated assistive system to generate context-oriented image descriptions to assess incoming threats. The key functionality of the system includes the detection and identification of human and gene…
▽ More
An assistive solution to assess incoming threats (e.g., robbery, burglary, gun violence) for homes will enhance the safety of the people with or without disabilities. This paper presents "SafeNet"- an integrated assistive system to generate context-oriented image descriptions to assess incoming threats. The key functionality of the system includes the detection and identification of human and generating image descriptions from the real-time video streams obtained from the cameras placed in strategic locations around the house. In this paper, we focus on develo** a robust model called "SafeNet" to generate image descriptions. To interact with the system, we implemented a dialog enabled interface for creating a personalized profile from face images or videos of friends/families. To improve computational efficiency, we apply change detection to filter out frames that do not have any activity and use Faster-RCNN to detect the human presence and extract faces using Multitask Cascaded Convolutional Networks (MTCNN). Subsequently, we apply LBP/FaceNet to identify a person. SafeNet sends image descriptions to the users with an MMS containing a person's name if any match found or as "Unknown", scene image, facial description, and contextual information. SafeNet identifies friends/families/caregiver versus intruders/unknown with an average F-score 0.97 and generates image descriptions from 10 classes with an average F-measure 0.97.
△ Less
Submitted 26 January, 2020;
originally announced February 2020.
-
Online Simultaneous State and Parameter Estimation
Authors:
Ryan Self,
Moad Abudia,
S. M. Nahid Mahmud,
Rushikesh Kamalapurkar
Abstract:
In this paper, a concurrent learning based adaptive observer is developed for a class of second-order nonlinear time-invariant systems with uncertain dynamics. The developed technique results in uniformly ultimately bounded state and parameter estimation errors. As opposed to persistent excitation which is required for parameter convergence in traditional adaptive control methods, the developed te…
▽ More
In this paper, a concurrent learning based adaptive observer is developed for a class of second-order nonlinear time-invariant systems with uncertain dynamics. The developed technique results in uniformly ultimately bounded state and parameter estimation errors. As opposed to persistent excitation which is required for parameter convergence in traditional adaptive control methods, the developed technique only requires excitation over a finite time interval to achieve parameter convergence. Simulation results in both noise-free and noisy environments are presented to validate the design.
△ Less
Submitted 13 October, 2020; v1 submitted 21 March, 2017;
originally announced March 2017.
