Arnab Barua - Academia.edu (original) (raw)

Papers by Arnab Barua

2023 IEEE 13th International Conference on System Engineering and Technology (ICSET)

Sensors & Diagnostics

An increase in microsensor deflection with an increase in blood viscosity during coagulation.

Entropy

Cell decision making refers to the process by which cells gather information from their local mic... more Cell decision making refers to the process by which cells gather information from their local microenvironment and regulate their internal states to create appropriate responses. Microenvironmental cell sensing plays a key role in this process. Our hypothesis is that cell decision-making regulation is dictated by Bayesian learning. In this article, we explore the implications of this hypothesis for internal state temporal evolution. By using a timescale separation between internal and external variables on the mesoscopic scale, we derive a hierarchical Fokker–Planck equation for cell-microenvironment dynamics. By combining this with the Bayesian learning hypothesis, we find that changes in microenvironmental entropy dominate the cell state probability distribution. Finally, we use these ideas to understand how cell sensing impacts cell decision making. Notably, our formalism allows us to understand cell state dynamics even without exact biochemical information about cell sensing pro...

Frontiers in Neurorobotics

Ankle joint power is usually determined by a complex process that involves heavy equipment and co... more Ankle joint power is usually determined by a complex process that involves heavy equipment and complex biomechanical models. Instead of using heavy equipment, we proposed effective machine learning (ML) and deep learning (DL) models to estimate the ankle joint power using force myography (FMG) sensors. In this study, FMG signals were collected from nine young, healthy participants. The task was to walk on a special treadmill for five different velocities with a respective duration of 1 min. FMG signals were collected from an FMG strap that consists of 8 force resisting sensor (FSR) sensors. The strap was positioned around the lower leg. The ground truth value for ankle joint power was determined with the help of a complex biomechanical model. At first, the predictors' value was preprocessed using a rolling mean filter. Following, three sets of features were formed where the first set includes raw FMG signals, and the other two sets contained time-domain and frequency-domain feat...

arXiv (Cornell University), Jun 13, 2016

Biosensors

Many studies have explored divergent deep neural networks in human activity recognition (HAR) usi... more Many studies have explored divergent deep neural networks in human activity recognition (HAR) using a single accelerometer sensor. Multiple types of deep neural networks, such as convolutional neural networks (CNN), long short-term memory (LSTM), or their hybridization (CNN-LSTM), have been implemented. However, the sensor orientation problem poses challenges in HAR, and the length of windows as inputs for the deep neural networks has mostly been adopted arbitrarily. This paper explores the effect of window lengths with orientation invariant heuristic features on the performance of 1D-CNN-LSTM in recognizing six human activities; sitting, lying, walking and running at three different speeds using data from an accelerometer sensor encapsulated into a smartphone. Forty-two participants performed the six mentioned activities by keeping smartphones in their pants pockets with arbitrary orientation. We conducted an inter-participant evaluation using 1D-CNN-LSTM architecture. We found tha...

2018 International Conference on Innovations in Science, Engineering and Technology (ICISET)

Due to the availability of various sensors in the smartphones, used by millions of people for com... more Due to the availability of various sensors in the smartphones, used by millions of people for communication, a new research arena is identified for data mining and machine learning. This paper aims to recognise ten human activities, i.e., sitting, walking, jogging, lying, walking upstairs and downstairs, cycling, standing, squatting in a toilet and fallen down, through smartphone sensors. For the implementation of our models, we collected labeled Gyroscope data, Accelerometer data, Temperature data and Humidity data from three users regarding their daily activities and summarised in 1Hz frequency. Then we used our training dataset to deduct a model for the prediction of activity recognition. Our work is noble in term of our system of data collection along with recognition of new activities with higher accuracy in recognition. These works have a wide range of applications as it may predict disease related to physical activities, monitor physical activities and elderly care.

2019 22nd International Conference on Computer and Information Technology (ICCIT)

Being concerned about the rising rate of the usage level of smartphone for last few years, resear... more Being concerned about the rising rate of the usage level of smartphone for last few years, researchers are striving to let out the disguised assets of smartphones. Regarding this phrase, the embedding of a variety of sensors such as accelerometer sensor, gyroscope sensor, humidity sensor etc. has attained the considerable attention of researchers for facilitating the human activity recognition task employing the sensor’s applications. In this paper, we practiced the Long Short-Term Memory a.k.a. LSTM deep learning model with an aim to recognize thirteen human activities stated as walking, walking upstairs, walking downstairs, sitting, standing, jogging, squatting in the toilet, fallen down, lying, cycling, drinking, eating and genital itching. We opted for these activities with an intention of early diagnosing of diabetes in near future. We amassed data from four sensors stated accelerometer sensor, gyroscope sensor, humidity sensor and temperature sensor subjecting ten volunteers implying a frequency of 10Hz. The data was attained using an android application which was developed by us for the purpose of accumulation of sensor data from the smartphone.

Brac University, 2020

Nowadays people exchange their personal information and interact with companions and close relati... more Nowadays people exchange their personal information and interact with companions and close relatives in a way that is revolutionized. In any case, the majority of them don’t have the foggiest idea how to utilize, where to click, where not to, where to remark, and where not to. A considerable lot of them are posting on Facebook anything they desire and wish. This posting, fellowship, and so on once in a while brings shocking occasions like identity theft, phishing, Cyber-wrongdoing and so on. So, Social media security has captured a great concern among the public and authority. At present, many features have been added to reduce the risk of hacking information. It is widely acknowledged that these features have played an important role in the security system. The essential focal point of our paper is on the safety implications of consumers posting their own Facebook information. We have made a survey containing 44 inquiries dependent on Facebook clients’ propensity and different things. We have looked at the ongoing information security rupture on Facebook through certain data mining substances. We have targeted three questions about victims of malware, identity theft, and phishing. From, our dataset we will know how many were victims of the three target parameters. We have implemented machine learning algorithms like ANN, XGBoost, SVM, Random Forest, Decision Tree, Gaussian Naive Bayes, Logistic Regression to identify the percentage of how many Facebook accounts are at risk and safety. Moreover, we will compare the best possible approach and worst approach among the algorithms to find the result. Among the models, we see ANN providing us the best result for the three labels with 89.89%, 94.94%, and 86.86%. This research illustrates how different machine learning algorithms predict the risk of Facebook users and which algorithm is most and least suitable to use in this scenario.

2021 IEEE 94th Vehicular Technology Conference (VTC2021-Fall), 2021

Camouflage is an interesting adaptation (for survivability) by organisms in terms of different ag... more Camouflage is an interesting adaptation (for survivability) by organisms in terms of different aggregation or fusion of colourations. Understanding these camouflage strategies in the presence of arsenic on transparent/semi-transparent species is pretty challenging. Previously, several researchers have demonstrated that colouration or pigmentation strategy in an organism is a strategy to merge with the environment to escape from predatory threats. Our study was done on a semi-transparent freshwater prawn species which exhibits a strategy of pigment droplets on its exoskeleton. Unlike previous studies, our findings robustly indicate the fact that pigment droplets are not the only reason for colouration. The pigment droplets rather regulate the darkness of the exoskeleton. However, the transparency of the abdominal muscles additionally plays a crucial role in creating a background of the pigment droplets. The transparency muscles allow light to pass through the abdomen, thus creating a...

Facebook has revolutionized the way people communicate with peers and close relatives, these user... more Facebook has revolutionized the way people communicate with peers and close relatives, these users share personal information with Facebook. In any case, the majority of them don't have the foggiest idea how to utilize, where to click, where not to, where to remark, and where not to. A considerable lot of them are posting on Facebook anything they desire and wish. This posting, fellowship, and so on once in a while brings shocking occasions like identity theft, phishing, Cyber-wrongdoing, and so on. So, Social media security has captured a great concern among the public and authority. At present, many features have been added to reduce the risk of hacking information. It is widely acknowledged that these features have played an important role in the security system. The essential focal point of this paper is on the security ramifications of clients sharing their own data on Facebook. We have made a survey containing 44 inquiries dependent on Facebook clients' propensity and ...

2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI), 2019

Evolution of sensor-based data accumulation process regulated the fact-finding stratum concerning... more Evolution of sensor-based data accumulation process regulated the fact-finding stratum concerning the scrutiny of human-centered motion towards an exemplary appearance. Such evolution enriched the fabrication of a variety of functional datasets, which facilitates the exploration of facts of various research domain. The MHEALTH is such a secondary dataset, that was prepared so as to facilitate the exploration regarding Human Activity Recognition (HAR). This paper performs a comparative study on human activity recognition process in terms of employment of two different data preprocessing methods accompanied by five fashionable classifiers entitled as Random Forest (RF), Support Vector Machine (SVM), Naive Bayes (NB), Multilayer Perceptron (MLP) and Deep Convolutional Neural Network (CNN). We employed the MHEALTH dataset to realize the human activity recognition process. The dataset encompasses information regarding 12 human activities subjecting 10 volunteers. Sensors were employed for the data accumulation process namely Accelerometer, Gyroscope and Magnetometer. Data preprocessing methods, that were employed on the mentioned dataset are elimination of null label instances and uniformity of unbalanced classes. The goal of this study is to analyze performance of different classifiers in terms of the mentioned data preprocessing methods and also identifying the process for which the classifiers exhibit superior accuracy.

Entropy, 2021

The way that progenitor cell fate decisions and the associated environmental sensing are regulate... more The way that progenitor cell fate decisions and the associated environmental sensing are regulated to ensure the robustness of the spatial and temporal order in which cells are generated towards a fully differentiating tissue still remains elusive. Here, we investigate how cells regulate their sensing intensity and radius to guarantee the required thermodynamic robustness of a differentiated tissue. In particular, we are interested in finding the conditions where dedifferentiation at cell level is possible (microscopic reversibility), but tissue maintains its spatial order and differentiation integrity (macroscopic irreversibility). In order to tackle this, we exploit the recently postulated Least microEnvironmental Uncertainty Principle (LEUP) to develop a theory of stochastic thermodynamics for cell differentiation. To assess the predictive and explanatory power of our theory, we challenge it against the avian photoreceptor mosaic data. By calibrating a single parameter, the LEUP ...

Advanced Industrial Wastewater Treatment and Reclamation of Water, 2021