Pritam Sarkar

Followers

Following

Public Views

Interests

Uploads

Papers by Pritam Sarkar

CardioGAN: Attentive Generative Adversarial Network with Dual Discriminators for Synthesis of ECG from PPG

Proceedings of the AAAI Conference on Artificial Intelligence

Electrocardiogram (ECG) is the electrical measurement of cardiac activity, whereas Photoplethysmo... more Electrocardiogram (ECG) is the electrical measurement of cardiac activity, whereas Photoplethysmogram (PPG) is the optical measurement of volumetric changes in blood circulation. While both signals are used for heart rate monitoring, from a medical perspective, ECG is more useful as it carries additional cardiac information. Despite many attempts toward incorporating ECG sensing in smartwatches or similar wearable devices for continuous and reliable cardiac monitoring, PPG sensors are the main feasible sensing solution available. In order to tackle this problem, we propose CardioGAN, an adversarial model which takes PPG as input and generates ECG as output. The proposed network utilizes an attention-based generator to learn local salient features, as well as dual discriminators to preserve the integrity of generated data in both time and frequency domains. Our experiments show that the ECG generated by CardioGAN provides more reliable heart rate measurements compared to the original...

Download

Computer-Aided Diagnosis using Class-Weighted Deep Neural Network

2019 18th IEEE International Conference On Machine Learning And Applications (ICMLA)

Computer-aided diagnosis has become a major focal point of Artificial Intelligence. Interpreting ... more Computer-aided diagnosis has become a major focal point of Artificial Intelligence. Interpreting medical images is often time-consuming and requires significant human expertise. Hence, there is an increasing demand to use machine learning techniques to correctly classify different medical images captured by mammography, CT scans, and MRI among others. This paper presents a deep learning method for computer-aided differential diagnosis of benign and malignant breast cancer tumors by avoiding potential errors caused by poor feature selection as well as class imbalances in the dataset. We design, develop and test an end-to-end convolutional neural network architecture for two different breast cancer datasets of fine needle aspiration biopsy samples, and show that our network outperforms the state of the art. Furthermore, we have introduced a loss coefficient which can be adjusted to fine-tune the performance of our network. The proposed method can be used to support oncologists in the detection of breast cancer with high confidence.

Self-Supervised Audio-Visual Representation Learning with Relaxed Cross-Modal Synchronicity

arXiv (Cornell University), Nov 9, 2021

We present CrissCross, a self-supervised framework for learning audiovisual representations. A no... more We present CrissCross, a self-supervised framework for learning audiovisual representations. A novel notion is introduced in our framework whereby in addition to learning the intra-modal and standard 'synchronous' cross-modal relations, CrissCross also learns 'asynchronous' cross-modal relationships. We perform in-depth studies showing that by relaxing the temporal synchronicity between the audio and visual modalities, the network learns strong generalized representations useful for a variety of downstream tasks. To pretrain our proposed solution, we use 3 different datasets with varying sizes, Kinetics-Sound, Kinetics400, and Au-dioSet. The learned representations are evaluated on a number of downstream tasks namely action recognition, sound classification, and action retrieval. Our experiments show that CrissCross either outperforms or achieves performances on par with the current state-of-the-art self-supervised methods on action recognition and action retrieval with UCF101 and HMDB51, as well as sound classification with ESC50 and DCASE. Moreover, CrissCross outperforms fully-supervised pretraining while pretrained on Kinetics-Sound. The codes, pretrained models, and supplementary material are available on the project website.

Download

Detection of maternal and fetal stress from the electrocardiogram with self-supervised representation learning

Scientific Reports, 2021

In the pregnant mother and her fetus, chronic prenatal stress results in entrainment of the fetal... more In the pregnant mother and her fetus, chronic prenatal stress results in entrainment of the fetal heartbeat by the maternal heartbeat, quantified by the fetal stress index (FSI). Deep learning (DL) is capable of pattern detection in complex medical data with high accuracy in noisy real-life environments, but little is known about DL’s utility in non-invasive biometric monitoring during pregnancy. A recently established self-supervised learning (SSL) approach to DL provides emotional recognition from electrocardiogram (ECG). We hypothesized that SSL will identify chronically stressed mother-fetus dyads from the raw maternal abdominal electrocardiograms (aECG), containing fetal and maternal ECG. Chronically stressed mothers and controls matched at enrolment at 32 weeks of gestation were studied. We validated the chronic stress exposure by psychological inventory, maternal hair cortisol and FSI. We tested two variants of SSL architecture, one trained on the generic ECG features for emo...

Download

Suppression and controlling of higher order modes in microstrip antenna with composite resonator

2017 8th Annual Industrial Automation and Electromechanical Engineering Conference (IEMECON), 2017

Suppression of higher order harmonic modes of the resonating frequency in a microstrip antenna ha... more Suppression of higher order harmonic modes of the resonating frequency in a microstrip antenna has been proposed in the paper. Controlling harmonics in the microstrip antenna has been implemented by integrating composite resonators at the microstrip feed line which resulted in elimination of higher order modes occurring between the fundamental resonating frequency and the second harmonics of the microstrip antenna. The integration of the composite resonator in the microstrip feed line to suppress and control the higher order modes is the prime novelty of the paper.

Happy Driver: Investigating the Effect of Mood on Preferred Style of Driving in Self-Driving Cars

Proceedings of the 9th International Conference on Human-Agent Interaction, 2021

Figure 1: In an online experiment (N=182), participants experienced three driving styles paired w... more Figure 1: In an online experiment (N=182), participants experienced three driving styles paired with three moods, with the hypothesis that matched pairings (eg. calm mood-conservative driving style) would increase driver satisfaction. Exemplar still images of the mood videos in the top row from left to right [18, 24, 32].

Download

Frequency Tuning Bandwidth Enhancement in Compact Microstrip Antenna using Saw Tooth Shaped DGS

2018 IEEE 9th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON), 2018

In the paper a saw tooth shaped Defective Ground Structure (DGS) is implemented in the ground sur... more In the paper a saw tooth shaped Defective Ground Structure (DGS) is implemented in the ground surface of the antenna to achieve the GSM frequency band. The length of the defect is then modified in order to attain resonance of the microstrip antenna at the Bluetooth frequency band and the Wimax frequency band. Furthermore to elevate the performance of the proposed microstrip antenna at the GSM application frequency band and the Wimax application frequency band, dual DGS is integrated in the ground plane. Thus improved bandwidth and improved gain is achieved in the GSM application frequency band and the Wimax application frequency band for proposed microstrip antenna.

Classification of Cognitive Load and Expertise for Adaptive Simulation using Deep Multitask Learning

2019 8th International Conference on Affective Computing and Intelligent Interaction (ACII), 2019

Simulations are a pedagogical means of enabling a risk-free way for healthcare practitioners to l... more Simulations are a pedagogical means of enabling a risk-free way for healthcare practitioners to learn, maintain, or enhance their knowledge and skills. Such simulations should provide an optimum amount of cognitive load to the learner and be tailored to their levels of expertise. However, most current simulations are a one-type-fits-all tool used to train different learners regardless of their existing skills, expertise, and ability to handle cognitive load. To address this problem, we propose an end-to-end framework for a trauma simulation that actively classifies a participant's level of cognitive load and expertise for the development of a dynamically adaptive simulation. To facilitate this solution, trauma simulations were developed for the collection of electrocardiogram (ECG) signals of both novice and expert practitioners. A multitask deep neural network was developed to utilize this data and classify high and low cognitive load, as well as expert and novice participants. A leave-one-subject-out (LOSO) validation was used to evaluate the effectiveness of our model, achieving an accuracy of 89.4% and 96.6% for classification of cognitive load and expertise, respectively.

Download

Self-Supervised Learning for ECG-Based Emotion Recognition

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020

We present an electrocardiogram (ECG)-based emotion recognition system using self-supervised lear... more We present an electrocardiogram (ECG)-based emotion recognition system using self-supervised learning. Our proposed architecture consists of two main networks, a signal transformation recognition network and an emotion recognition network. First, unlabelled data are used to successfully train the former network to detect specific predetermined signal transformations in the self-supervised learning step. Next, the weights of the convolutional layers of this network are transferred to the emotion recognition network, and two dense layers are trained in order to classify arousal and valence scores. We show that our self-supervised approach helps the model learn the ECG feature manifold required for emotion recognition, performing equal or better than the fully-supervised version of the model. Our proposed method outperforms the state-of-the-art in ECG-based emotion recognition with two publicly available datasets, SWELL and AMIGOS. Further analysis highlights the advantage of our self-supervised approach in requiring significantly less data to achieve acceptable results.

Download

Toward Dynamically Adaptive Simulation: Multimodal Classification of User Expertise Using Wearable Devices

Sensors, 2019

Simulation-based training has been proven to be a highly effective pedagogical strategy. However,... more Simulation-based training has been proven to be a highly effective pedagogical strategy. However, misalignment between the participant’s level of expertise and the difficulty of the simulation has been shown to have significant negative impact on learning outcomes. To ensure that learning outcomes are achieved, we propose a novel framework for adaptive simulation with the goal of identifying the level of expertise of the learner, and dynamically modulating the simulation complexity to match the learner’s capability. To facilitate the development of this framework, we investigate the classification of expertise using biological signals monitored through wearable sensors. Trauma simulations were developed in which electrocardiogram (ECG) and galvanic skin response (GSR) signals of both novice and expert trauma responders were collected. These signals were then utilized to classify the responders’ expertise, successive to feature extraction and selection, using a number of machine lear...

Download

Elimination of higher order modes in wearable microstrip antenna

2016 IEEE 7th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON), 2016

Reduction of higher order harmonic modes of the fundamental operating frequency in a wearable mic... more Reduction of higher order harmonic modes of the fundamental operating frequency in a wearable microstrip antenna has been proposed. Controlling harmonics in the microstrip antenna has been implemented by integrating simple I shaped resonator at the microstrip feed line. As a result higher order modes occurring between the fundamental resonating frequency and the second harmonics of the wearable microstrip antenna have been successfully suppressed.

Fixed accuracy confidence interval on the common variance in compound symmetric multivariate normal sampling

Sequential Analysis, 2019

This work deals with construction of fixed accuracy confidence intervals for the common variance ... more This work deals with construction of fixed accuracy confidence intervals for the common variance in compound symmetric multivariate normal distribution by adopting a Stein-type two-stage setup and a completely sequential setup. Related asymptotics are developed and the procedures are evaluated by simulation study.

Download

Pritam Sarkar

Uploads

Papers by Pritam Sarkar

Log In