Open to opportunities

Zainab Ehab

@zainabehab

Message

Data scientist specializing in medical imaging, ML, and end-to-end data workflows.

Egypt

Message

What I'm looking for

I seek roles applying ML and computer vision to healthcare/medical imaging, with opportunities for research impact, end-to-end project ownership, and collaborative, learning-focused teams.

I am a data scientist with a strong focus on medical imaging and machine learning, combining academic research and applied project experience to deliver high‑accuracy diagnostic models. I build end-to-end solutions from data collection and preprocessing to model training, evaluation, and visualization.

My published research used few-shot learning with MobileNet and prototypical networks to detect autism from limited facial images, achieving 95% accuracy on 5-way 5-shot tasks. I have developed multiple medical ML projects — including brain and lung cancer classifiers and a heart–lung digital twin — that demonstrate robust performance and careful handling of imbalanced and noisy data.

I also teach applied Python, SQL, Power BI, Excel and Tableau, and emphasize industry-focused case studies and full data workflows to prepare students for real-world data roles. I seek opportunities to apply my expertise in healthcare AI, medical imaging, and data-driven product development.

Experience

Work history, roles, and key accomplishments

Current

Data Analysis Instructor

Current

National Telecommunication Institute

Sep 2025 - Present (1 month)

Teach applied Python, SQL, Power BI, Excel and Tableau and guide students through end-to-end data workflows using industry-focused case studies to prepare them for practical data roles.

Python SQL Power BI Tableau Excel Data Visualization ETL Teaching Case Study Development

Undergraduate Researcher

Egypt Japan University of Science and Technology

Jun 2025 - Present (4 months)

Built an integrated heart–lung digital twin by segmenting MRI/CT scans with U-Net GAN and combining 3D anatomical reconstruction and clinical metadata to achieve 96% disease prediction accuracy.

U Net Blender Python

Researcher (Few-Shot Learning)

Novel Intelligent and Leading Emerging Sciences Conference

Oct 2024 - Present (1 year)

Developed a few-shot learning model using MobileNet within a prototypical network for autism detection from limited facial images, achieving 95% accuracy on 5-way 5-shot tasks via transfer learning and augmentation.

Few Shot Learning MobileNet Prototypical Networks Transfer Learning PyTorch Episodic Training Computer Vision

Machine Learning Engineer

Independent Projects

Delivered multiple ML pipelines including lung cancer classification (improved accuracy 50%→88%), brain tumor classification (95.79% accuracy), and RN-SMOTE for noise-filtered oversampling boosting F1-scores across models.

Python scikit learn Keras OpenCV PyTorch SMOTE Model Evaluation

Education

Degrees, certifications, and relevant coursework

Egypt Japan University of Science and Technology

Bachelor of Computer Science, Artificial Intelligence and Data Science

2021 - 2025

Bachelor of Computer Science with a focus on Artificial Intelligence and Data Science, covering coursework in data structures, algorithms, cloud computing, big data, NLP, databases, data mining, and computer vision.