Intelligent Sensing and Vision (IntSaV)

• Professor Massoud Zolgharni (Head)
• Dr Sama Aleshaiker
• Dr Obed Brew
• Dr Laden Husamaldin
• Professor Jonathan Loo
• Dr Jin Luo (School of Biomedical Sciences)
• Dr Nasser Matoorianpour
• Dr Ikram Ur Rehman
• Dr Nagham Saeed

The aim of Eugenio’s research is in the development of a novel system for electroglottographic measurements where deep learning models are employed to allow the system to self-calibrate for specific users, provide self-deployability and discard unwanted signals.

Prior to joining the doctoral community at UWL, Eugenio obtained a BA in interpreting and translation at UNINT University of Rome where his studies involved phonetics, phonology and anatomy of phonation. He then joined UWL where he obtained a BSc (Hons) in Applied Sound Engineering and successively an MSc in Digital Audio Engineering. As part of his MSc, Eugenio worked with electroglottography for the conversion of voice signals into control signals in music technologies applications which laid the foundations of his current research.

Eugenio’s academic experience outlines the interdisciplinary nature of his research which combines the study of human phonation anatomy, signal processing, computing and electronics.

Contact email: 21261215@student.uwl.ac.uk

Robert’s research is aimed at developing automatic quality scores pipeline that enable cardiologist’s consistent assessment of cardiac diagnosis. The PhD research makes use of machine learning models and clinical practice to elicit precise throughput for echocardiograms.

This cross-disciplinary research which includes experts in the health sectors, the use of high-end computing framework will provide, automatic cardiac view detection, real-time quality scores and advance precise diagnosis for clinical pathologies.

Prior to his time at UWL, Robert earned his MSc degree in automation and control from Edinburg Napier University and built his career in cutting-edge technology research and development. His cross-disciplinary experience includes embedded applications (SoC) for medical devices, industrial automation, guidance and navigational systems. His current interest includes cognitive systems, kinematic modelling and clinical AI.

Contact email: robbie.labs@uwl.ac.uk

Beth’s current research aims to develop a deep learning model capable of taking accurate, reproducible and bias-resistant Doppler wave measurements.

She recently completed an MSc in Software Engineering at the University of West London and, before commencing her PhD, was Head of Computing at a secondary school.

Beth’s interests are in machine learning, artificial intelligence and their application to solve complex problems with the potential to enhance quality of life, especially in healthcare.

contact email: elisabeth.lane@uwl.ac.uk

Henrique Ribeiro is a PhD candidate in artificial intelligence directed towards healthcare, currently researching and developing a Deep Neural Network model capable of accurately predicting heart failure using signal processing.

He recently completed his MSc in Software Engineering at the University of West London, where during his academic progression, he acquired knowledge in automation, robotics and software engineering.

Before commencing his MSc, Henrique was an IT consultant at a major retail company.

Contact email: henrique.demeloribeiro@uwl.ac.uk

Jamie’s research aims to use IoT-enabled technology and machine learning to devise methods of medical intervention pertaining to those suffering from neurological disorders. Jamie is a postgraduate student currently studying for a PhD in Computer Science, focusing on IoT (Internet of Things), real-time data processing and machine learning.

He completed his undergraduate degree with the University of West London in 2018, attaining a first-class degree in Computer Science. The following year Jamie completed an MSc in Software Engineering and was introduced to IoT-enabled technology and the potential advancements it can lead to regarding current healthcare practices.

Contact email: pordjam@uwl.ac.uk

Jevgeni is currently conducting research on Explainable Artificial Intelligence (XAI) techniques in an Echocardiographic context, which aims to find the best methods that could explain predictions of the clinical models.

Even though AI expands its use cases and fields, it is stuck in front of the wall of trust in the fields that involve critical decisions, such as medicine.

Clinical experts require more understanding and explanations behind the models’ predictions before using them to diagnose patients. Jevgeni’s research aims to find ways that could efficiently give such explanations to clinical experts.

Jevgeni completed an accredited BSc programme in Computer Science at UWL and gained professional experience as a Software Engineer at an international company before joining IntSaV.

Contact email: jevgeni.jevsikov@uwl.ac.uk

Adriano Platania earned his Masters in Information Systems from the University of West London, obtaining a distinction after achieving first honours with a degree in Computer Science in 2019.

He is currently a PhD candidate at the same university. His research is focused on developing deep learning models in computer vision.

Before his time at UWL, Adriano was a commercially focused professional with extensive travel industry experience in agency and client-side roles. Adriano loves math, philosophy, interior design, impressionism, and puzzle-solving.

Contact email: gaetano.platania@uwl.ac.uk

Eman received her B.Sc. degree in Computer Science from the Islamic University of Gaza (IUG), Palestine and a BEng. degree (with honours) in Software Engineering from the University of Palestine (UP).

She received the M.Sc. degree in Computer Information Systems with distinction from Jordan. She has a long experience working in the academic fields related to computer science. Before joining the UWL, Eman was a lecturer in the Faculty of Information Technology at UP, and she was the Head of the Multimedia Department at the faculty of IT since 2016.

She has a great commitment to keep learning and developing her professional skills and research skills. She enjoys solving challenging research problems. Her research interests include Data Science, Artificial Intelligence, Machine Learning and Deep Learning. She has publications related to AI and Deep Learning.

Contact email: 21452392@student.uwl.ac.uk

Aamir Anwar earned an undergraduate degree in Software Engineering from City University of Science Information Technology, Peshawar Pakistan, and a Master’s degree in Software Engineering from Bahria University, Islamabad Pakistan.

Mr Anwar held more than six years of industrial and academia experience in different positions. Currently, he is a PhD candidate at UWL. Prior to commencing his PhD, he was a Software Engineering lecturer at Shaheed Zulfikar Ali Bhutto Institute of Science and Technology (SZABIST), Islamabad Pakistan.

Aamir’s research areas include Knowledge Engineering, Machine Learning, Artificial Intelligence in Healthcare and E-Learning.

Contact email: 21452391@student.uwl.ac.uk

• Miss Noha Ghatwary (PhD student) – joint supervision with the University of Lincoln
• Miss Alya Amer (PhD student) – joint supervision with the University of Lincoln
• Mr Ben Docking (MRes student) – joint supervision with the University of Lincoln

Project summary

This cross-disciplinary research proposal aims to develop a real-time and fully automated pipeline for cardiologists to accurately assess all cardiac functions.

A combination of engineering expertise (software development, artificial intelligence, parallel programming, GPUs, medical image acquisition and processing), basic science (mathematical modelling, algorithm development, statistics), and clinical experience (cardiology, echocardiography, Quality Index scoring) will be used.

Research team

• Mr Robert Labs
• Dr Massoud Zolgharni

The main aim of this project is to develop a deep neural network to detect one particular type of irregular heart rhythm (i.e., atrial fibrillation) using ECG recordings.

The primary objective is to investigate the feasibility of using a variety of different DNN models and adopt the most suitable architecture for the intended application.

The secondary objective is to validate the accuracy of the developed model using a large dataset of annotated data by the experts.

Finally, the third objective is to refine the model parameters to achieve cardiologist-level arrhythmia detection and classification using ECG signals.

Research team

• Mr Henrique de Melo Ribeiro
• Dr Massoud Zolgharni

The aim of this study is to arrive at a reliable and robust framework to automatically segment the left ventricle through a 2D echocardiographic image sequence, providing a degree of assistance for physicians in their analysis of echo images.

To achieve the aim of this study, certain objectives are needed, such as:

1. to tackle the challenging and difficult nature of echocardiographic images with the ability to provide a final shape contouring of the left ventricle borders in a good shape connectivity, close to the gold standard delineation by physicians
2. to provide an automated segmentation would lessen the tedious time consumed to segment a number of frames for a large number of patients, leaving more time for the physician to concentrate on the evaluation of some measures out of the segmented region to assess the heart functionality.

Research team

• Miss Neda Azarmehr
• Dr Massoud Zolgharni
• Professor Xujiong Ye (University of Lincoln)

The main aim of this project is to develop a deep learning model to make reproducible, bias-resistant, and multi-beat tissue Doppler measurements.

We will evaluate the performance of the automated technology using a large dataset of real-world images of cardiac images.

Research team

• Miss Elisabeth Sarah Lane
• Dr Massoud Zolgharni

This research is aiming to improve the performance and extend the usage electroglottography (EGG).

EGG provides valuable information about the voice production, but it has some significant drawbacks that impede its use in a wider range of applications.

This research project is aiming to overpass the limitations of EGG by investigating the designing of a novel self-deploying EGG sensor, advance its performance by using artificial intelligence practices and improve users’ mobility by incorporating wireless networking design in the system.

• Mr Eugenio Donati
• Dr Christos Chousidis
• Dr Massoud Zolgharni

This project aims to use Internet of Things (IoT) medical sensor technology, real-time data gathering and big data analytics to aid patients who suffer with epilepsy, by real-time detection of epileptic seizures.

Research team

• Mr Jamie Pordoy
• Dr Massoud Zolgharni

Project description

This research project aims to use Internet of Things (IoT) medical sensor technology, real-time data gathering and big data analytics to aid patients who suffer with epilepsy, by real-time detection of epileptic seizures.

Epileptic seizures are the causation of the neurological disorder epilepsy, which is the second foremost cause of dynamical brain disorder affecting over 65 million people (Iasemidis, 2003).

An overload of electrical activity between communicating neurons causes a temporal imbalance of neurological activity, thus breaking out in the form of seizure release, often leaving the patient with a loss of anatomical motor functions, clarity of memory as well as semi paralysis.

The main of this study is to investigate the feasibility of the real-time strain imaging in cardiac images by implementing the tracking algorithms on Graphics Processing Units (GPU).

Research team

• Mr Ben Docking
• Dr Massoud Zolgharni
• Professor Xujiong Ye (University of Lincoln)