Yasser El Jarida

PhD Student, Medical AI

UM6P College of Computing, Ben Guerir, Morocco

Education

PhD in Computer Science (Medical AI) — UM6P College of Computing, Ben Guerir (2024–Present)
Computer Engineering: Big Data and Cloud Computing — ENSET, Mohammedia (2021–2024)
Preparatory Classes in Mathematics and Physics (CPGE) — Ibn Abdoun High School, Khouribga (2019–2021)
Baccalaureate in Mathematical Sciences A — Abou El Kacem Ezzayani High School, Khenifra (2018–2019)

Keywords: Single-Cell Analysis, AI in Healthcare, Medical Imaging

Awarded the highly competitive BCDI scholarship to conduct an 8-month advanced research internship.
Spearheading early-stage research investigating the intersection of machine learning and cellular biology, with a primary focus on identifying predictive markers for diabetes.

Keywords: Transformers, Cross-Attention, PyTorch, ECG Analysis, Surgical Phase Detection, Synthetic Data

Conducting advanced research on AI for healthcare, focusing on cardiology (ECG analysis) and surgical workflow optimization via intelligent video analysis.
Designed and implemented a Boundary-Aware FACT (Frame–Action Cross-Attention Temporal) model for surgical phase segmentation, achieving 2nd place at the MICCAI 2025 OMNIA SICS155 Challenge.
Introduced a lightweight boundary supervision head and a boundary-weighted temporal smoothing loss, improving segmentation accuracy by +1.3% and F1-score by +1.5 without changing inference cost.
Explored transformer-based architectures (FACT, SurgFormer, TimeSformer, VideoMAE‑v2) for temporal understanding and phase recognition in surgical videos.
Collaborated with clinical experts to ensure interpretability, efficiency, and workflow alignment of deployed models.
Prior work: synthetic datasets and CNN regression (ResNet50, EfficientNet‑B0, InceptionV3) for instant particle size distribution (PSD); published at CVPR 2025 SynData4CV.

Keywords: Python, YOLOv8, SAM, CVAT, DeepFill v2, ResNet50, Streamlit

Built a computer vision pipeline with YOLOv8 + SAM for detection, segmentation, and reflectivity assessment of CSP mirrors.
Improved data quality with image inpainting (DeepFill v2), achieving R² = 0.94 for reflectivity prediction (ResNet50).
Deployed a Streamlit dashboard for interactive evaluation and visualization.

Keywords: Python, TensorFlow, CNNs, YOLOv8

Implemented vision-based violence detection using YOLOv8 with extensive tuning and validation.
Analyzed precision, recall, and F1 to guide iterative improvements.

Awarded Canadian International Development Scholarships 2030 (BCDI 2030), 2026
2nd Place — MICCAI 2025 OMNIA SICS155 Surgical Phase Recognition Challenge, 2025
Accepted Paper — CVPR 2025 Workshop (SynData4CV), 2025
1st Place — Hackathon: Blockchain and AI at the Service of Health, 2023

Advanced detection and segmentation of parabolic trough collector and Fresnel mirrors for CSP maintenance using YOLOv8 and segment anything model. M. Boujoudar, A. Moulay Taj, Y. El Jarida, I. Bouarfa, A. Mouaky, M. El Ydrissi, E. G. Bennouna, H. Ghennioui. The Visual Computer (Springer), 2026.
Development and Internal Validation of an AI-Driven Model for 5-Year Cardiovascular Disease Risk Prediction in the Canadian Longitudinal Study on Aging. M. Bahani, Y. Iraqi, J. Delfrate, Y. El Jarida, A. Nolin-Lapalme, A. Sowa, R. Avram. Journal of the American College of Cardiology (JACC), 2025.
Instant Particle Size Distribution Measurement Using CNNs Trained on Synthetic Data. Y. El Jarida, Y. Iraqi, L. Mekouar. CVPR 2025 Workshop: SynData4CV, 2025.

Designed a boundary-aware extension of the FACT (Frame–Action Cross-Attention Temporal) model for surgical phase segmentation, introducing auxiliary boundary supervision to improve temporal consistency and boundary precision.
Achieved 2nd place (82% test accuracy) on the official SICS-155 leaderboard, outperforming standard Transformer and MS-TCN baselines.
Implemented an efficient boundary head and a boundary-weighted total-variation loss that reduced over-segmentation and sharpened phase transitions without inference overhead.
Conducted extensive hyperparameter sweeps and ablation studies demonstrating consistent accuracy and F1-score improvements (+1.3% accuracy, +1.5 F1).

Created a realistic, high-quality synthetic dataset using Blender to train CNN models (ResNet50, EfficientNet-B0, InceptionV3) for predicting particle size distributions (PSD).
Evaluated the effectiveness of synthetic data for accurately solving PSD prediction tasks, demonstrating the potential of using generated data to replace or supplement real-world samples.

Developing robust and generalizable foundation models based on Vision Transformer architectures for detailed ECG data interpretation and precise cardiac diagnostics.
Addressing key challenges in ECG analysis, including handling noisy signals, detection of irregular cardiac rhythms, and enhancing model interpretability to improve clinical reliability.

Human Research: Data or Specimens Only Research (Basic Course) — CITI Program / MIT Affiliates
Fundamentals of Accelerated Computing with CUDA Python — NVIDIA
Neural Networks and Deep Learning — DeepLearning.ai
Supervised Machine Learning: Regression and Classification — DeepLearning.ai