About Me
I’m Usama, a dedicated PhD student at Birmingham City University, holding a BSc in Computer Science with AI from the University of Nottingham. With a diverse range of experiences, I’ve cultivated strong teamwork and leadership skills while delving into exciting projects.
Proficient in a variety of programming languages, including Python, Java/Kotlin, Dart, and C/C#, I bring a versatile skill set to the table.
My journey so far has exposed me to diverse environments and multifaceted tasks. The knowledge I’ve acquired, whether through collaborative projects or tackling complex challenges, sets me apart. My unique background and expertise in the field contribute to my distinctive problem-solving and teamwork capabilities.
I invite you to explore my website to learn more about me and my exceptional journey. Let’s connect and discover how I can be a valuable asset to your team.
Projects
SwinCup - Cascaded Swin Transformer for Histological Structures Segmentation in Colorectal Cancer
Developed a new deep learning model for segmenting abnormal gland structures from histology image data.
- Reviewed current work on similar datasets.
- Developed a viable prototype model based on previous architecture designs.
- Implemented a machine learning pipeline to validate the performance of the model on multiple datasets.
- Published an article summarizing the key points of the project in a peer-reviewed journal.
- Technologies used: Python, PyTorch/TensorFlow/Keras, NumPy, OpenCV, grid search/random search/Bayesian optimization, Slurm for workload parallelization.
GNN-based Medical Image Classification
- Developed a new pipeline for applying Graph Neural Networks (GNN) to medical images.
- Conducted a thorough study of hyperparameters to optimize performance on multiple datasets using genetic algorithms.
- Implemented the GNN model using PyTorch and TensorFlow frameworks.
- Preprocessed medical image data using OpenCV and NumPy.
- Visualized results using Matplotlib and other data visualization tools.
- Technologies used: Pytorch Geometric, Optuna, GraphGym, Scikit-learn, Pandas.
Remote Sensing for Mineral Exploration
- Implemented machine learning models for classifying patches and detecting minerals from satellite data.
- Constructed a model training pipeline using PyTorch framework.
- Conducted feature engineering and hyperparameter tuning to improve model performance.
- Preprocessed remote sensing data using rasterio and geopandas libraries.
- Analyzed model predictions and visualized results using Matplotlib and other data visualization tools.
A survey on artificial intelligence in histopathology image analysis
- Reviewed literature for the latest techniques.
- Identified common challenges and limitations in the current histopathology image analysis workflow.
- Analyzed the potential impact of AI on improving diagnostic accuracy, time efficiency, and patient outcomes in histopathology.
- Analyzed frameworks for integrating AI into the histopathology workflow, outlining key considerations and best practices for implementation.
Continuous Authentication using sensor data
- Implemented a series of Machine Learning techniques such as Neural Networks, SVMs, and Siamese networks to authenticate users through their walking using gyroscope sensor data.
- Gained an understanding of one-shot learning and the ability to create ML models with scarce data.
- Formalized theoretical concepts in a real-world project.
A pipeline implementation is explored that tries to enhance the accuracy of Deep Learning models in the detection and segmentation of pneumothorax from a set of chest radiographs.
By cascading a CNN and a U-Net, we ensure that the CNN will filter out all the cases with no pathology, i.e.’Normal’ cases, leaving the U-Net (trained on positive cases only) to focus on arranging a set of filters for the separation of pathology from the infected lungs. The models are cross validated on the same data to ensure consistency. The proposed method demonstrates higher accuracy in both segmenting and detecting the pathology. By implementing it with more complex architectures and integrating in the domain knowledge of radiologists, this methods can be applied in conjunction with other applications to rapidly triage and prioritise cases for the presence of anomalies. The reason I chose this project is because of my great interest in machine learning algorithms and their impact. I wanted to explore theses algorithms and learn more about them and what better cause to do that for than the betterment of human lives.
A Health game developed to spread awareness about epilepsy.
A project initiated to create a mobile game with the main purpose of spreading awareness of epilepsy to the general public and do so in a way that is as engaging and effective as possible. The project specifies that a full release game will be developed. The game will involve different gaming genres to teach the user about various aspects and types of epilepsy in both direct and subtle ways for learning to be as enjoyable and effective as possible.
The project utilised Unity game Engine and C# language, I was also exposed to concepts like Animation, 3D Modeling, and Level Design
I was tasked as the project manager for a team of 4 members. I learned a lot from this experience from managing people to conflict resolution. The project spaned a whole academic year (9 months) throughout which we gathered requirements from the client, presented our ideas, and proceeded to develop based on the chosen pathway. There was a lot to learn here, delegating tasks was something I had to develop to ensure that we are on track to finish on time. Overall, just like any other project, our group project included a lot of ups and downs. The ups improved our overall self-confidence in our knowledge, while the Downs maximized the learning experiences we gained.
Little experiences encountered throughout the process helped us rethink our decisions, improve our skill sets, and gain more knowledge and professionalism.
Experience
Birmingham City University, Birmingham
Visiting Lecturer
October 2021 - Present
- Evaluated and revised plans and course content to achieve student-centered learning.
- Organized lectures around different AI and data visualization techniques.
- Trained a batch of students on using machine learning toolsets.
- Assessed different aspects of machine learning and data visualization design.
University of Nottingham Malaysia
https://www.nottingham.edu.myResearch Intern
June 2019 - August 2019
Developed deep learning model capable of diagnosing chest radio-graphs
Designed a Deep Learning model capable of diagnosing Chest Radio-graphs to potentially help in aiding early recognition of pneumothoraces that are usually diagnosed by a radiologist on a Chest X-ray and can sometimes be very difficult to confirm. • Analyzed data-sets to be tested by looking at previous multi-classification multi-label papers • Researched recent model architectures to design, implement , and test performance on the data-sets • Implemented a series of models to compare their performance using different performance techniques
Languages and Technologies : Python (Pytorch), FastAi, CNNs.
Education
Nottingham University
Bsc Computer Science with Artificial Intelligence
2016 - 2020
During my degree at Nottingham University I learnt lots of key skills such as team work and project management. During the degree, I have been given the chance to explore the fundamental concepts of machine learning and data science, and I have developed a keen interest in more specialized areas including big data, data mining and the internet of things.
Outside academia, I’ve developed effective communication, collaborative and organizational skills through work experience as part of the Activities network in the Student Union at my university. In this role, I was part of a team responsible for handling various events on campus, this included providing designs and posters for the events, and coordinating between societies and other teams for the execution of the events. I’ve also volunteered as a Student Ambassador for the Computer Science school, which saw me encourage potential students to pursue a degree in Computer/Data Science through sharing my own experiences and setting-up coding workshops. I’ve held other positions too, including as a member of the Organization team of the first TEDx event on-campus, and as a member of the Elections committee in the university where I managed the implementation of a new voting system in the university. I’ve been extremely proactive throughout my degree, from being a member of a squash team to fundraising for charities such as Islamic Relief and orphanage Foundations.
Birmingham City Unviversity
PhD Computing
2021 - Present
I am currently a PhD student at Birmingham City University, my research is focused on appling Patch-Based Deep Learning for Enhanced computer Vision I am also a visiting lecturer at the university where I teach a module on applied AI and data visualization. I am also a member of the Data Science Research Group at the university where I am involved in various projects related to data science and machine learning.
A Little More About Me
Alongside my interests in networks and software engineering some of my other interests and hobbies are:
- Hiking
- Gaming
- Photography I enjoy hiking and exploring new places, I have been to many places in the UK and I’m looking forward to exploring more places in the future.
I’m also interested in photography and videography as I like to be creative, I’m mainly enjoy landscape and abstract photography. I like to showcase my work on instagram if you would like to take a look - Usama Zidan.