Tahmina Zebin

Tahmina Zebin
University of East Anglia | UEA · School of Computing Sciences

PhD

About

30
Publications
95,630
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
488
Citations
Citations since 2016
24 Research Items
484 Citations
2016201720182019202020212022050100150
2016201720182019202020212022050100150
2016201720182019202020212022050100150
2016201720182019202020212022050100150
Introduction
Tahmina Zebin currently works at the School of Computing Sciences The University of East Anglia. Tahmina does research in Medical Image Processing, Wearable Sensors data processing, activity recognition with Statistical and Deep Learning Algorithms.
Additional affiliations
April 2017 - February 2020
The University of Manchester
Position
  • Research Associate
October 2016 - March 2017
The University of Manchester
Position
  • Researcher
September 2013 - present
American International University-Bangladesh
Position
  • Professor (Associate)

Publications

Publications (30)
Article
Full-text available
Edge computing aims to integrate computing into everyday settings, enabling the system to be context-aware and private to the user. With the increasing success and popularity of deep learning methods, there is an increased demand to leverage these techniques in mobile and wearable computing scenarios. In this paper, we present an assessment of a de...
Conference Paper
Full-text available
Serious Mental Illnesses (SMIs) including schizophrenia and bipolar disorder are long term conditions which place major burdens on health and social care services. Locomotor activity is altered in many cases of SMI, and so in the long term wearable activity trackers could potentially aid in the early detection of SMI relapse, allowing early and tar...
Preprint
Full-text available
Over the past few years, Domain Name Service (DNS) remained a prime target for hackers as it enables them to gain first entry into networks and gain access to data for exfiltration. Although the DNS over HTTPS (DoH) protocol has desirable properties for internet users such as privacy and security, it also causes a problem in that network administra...
Article
Over the past few years, Domain Name Service (DNS) remained a prime target for hackers as it enables them to gain first entry into networks and gain access to data for exfiltration. Although the DNS over HTTPS (DoH) protocol has desirable properties for internet users such as privacy and security, it also causes a problem in that network administra...
Article
Full-text available
Chest X-rays are playing an important role in the testing and diagnosis of COVID-19 disease in the recent pandemic. However, due to the limited amount of labelled medical images, automated classification of these images for positive and negative cases remains the biggest challenge in their reliable use in diagnosis and disease progression. We appli...
Chapter
Full-text available
A Network Intrusion Detection System is a critical component of every internet connected system due to likely attacks from both external and internal sources. A NIDS is used to detect network born attacks such as denial of service attacks, malware, and intruders that are operating within the system. Neural networks have become an increasingly popul...
Presentation
Full-text available
A presentation on Capsule Neural networks for MSc students
Conference Paper
Full-text available
In recent years machine learning methods for human activity recognition have been found very effective. These classify discriminative features generated from raw input sequences acquired from body-worn inertial sensors. However, it involves an explicit feature extraction stage from the raw data, and although human movements are encoded in a sequenc...
Presentation
Full-text available
Tutorial session on Single layer perceptron and its implementation in python
Chapter
Full-text available
Wearable inertial sensors are currently receiving pronounced interest due to applications in unconstrained daily life settings, ambulatory monitoring and pervasive computing systems. This research focuses on human activity recognition problem, in which inputs are multichannel time series signals acquired from a set of body-worn inertial sensors and...
Conference Paper
Full-text available
Article
Medical image analysis and data compression are rapidly evolving fields with growing applications in healthcare services e.g. teleradiology, teleconsultation, e-health, telemedicine and statistical analysis of medical data. In this paper, a Layered Set Partitioning in Hierarchical Trees (LSPIHT) algorithm for medical data compression and transmissi...
Article
Full-text available
Flexible manipulator systems exhibit many advantages over their traditional (rigid) counterparts. However, they have not been favored in production industries due to its obvious disadvantages in controlling the manipulator. This paper presents theoretical investigation into the dynamic modeling and characterization of a constrained two-link flexibl...
Article
Full-text available
Medical image analysis and data compression are rapidly evolving fields with growing applications in healthcare services e.g. teleradiology, teleconsultation, e-health, telemedicine and statistical analysis of medical data. In this paper, a Layered Set Partitioning in Hierarchical Trees (LSPIHT) algorithm for medical data compression and transmissi...
Article
Full-text available
Flexible manipulator systems exhibit many advantages over their traditional (rigid) counterparts. However, they have not been favored in production industries due to its obvious disadvantages in controlling the manipulator. This paper presents theoretical investigation into the dynamic modeling and characterization of a constrained two-link flexibl...

Questions

Question (1)

Network

Cited By

Projects

Projects (3)
Project
Sensor fusion techniques
Project
Work Package 1: Development of Initial Adaptive Sensing Traditional Sensors: Only designed to collect Data Wearable Clinic: Personalized and Data Responsive Sensor (Opportunities for sensor design and optimization) Electronic Health Record (Time Sparse): Appointment missed--> Pull Data from Wearable WP2 Prediction Algorithm--> Probability of An abnormal Event to occur-->Turn ON and Collect Data From the Sensor remotely. Design and Implementation of the application Peripheral Interface(API): - It will remotely control the Sensor - Integration with EHR - communication with Risk Prediction Algorithm WP2. 2. On Sensor/node signal processing(Unsupervised Machine Learning) Collected data --> Group into Phenotypes--> Transmitted to EHR/PC--> Reduce communication overhead. It will also facilitate: -Adaptive sensing/current level of sensing decision by the sensor itself. -Supported/ Supplemented by external information(Software/ algorithmic output of WP2 & WP3) Target Data Fusion: -Low temporal Resolution Data from EHR - High Temporal Resolution Data From Accelerometer. Map: Trained Machine learning models on sensors (Real time low power clustering of phenotypes) Challenges: -Custom Micro-chip(Not Cost-Effective) -Power consumption(how Low?) -Trust and Privacy of the System.
Project
Our focus in this research is on the use of deep learning approaches for human activity recognition (HAR) scenario, in which inputs are multichannel time series signals acquired from a set of body-worn inertial sensors and outputs are predefined human activities. Here, we present a feature learning method that deploys convolutional neural networks (CNN) to automate feature learning from the raw inputs in a systematic way. The influence of various important hyper-parameters such as number of convolutional layers and kernel size on the performance of CNN was monitored. Experimental results indicate that CNNs achieved significant speed-up in computing and deciding the final class and marginal improvement in overall classification accuracy compared to the baseline models such as Support Vector Machines and Multi-layer perceptron networks.