About Me
I’m a PhD candidate in Bioscience Engineering with a passion for blending machine learning, programming, data science, and computer vision to solve real-world challenges. With a strong foundation in mathematical modeling, I’ve worked on a range of projects, from improving customer satisfaction through data-driven solutions to enhancing statistical models for government institutions.
Most recently, my research focuses on developing algorithms to characterize hierarchical structures based on pair-wise interaction data, with a primary focus on animal studies. In addition to this core work, I also apply my expertise in AI to pharmaceutical quality control; developing innovative approaches to ensure the safety and precision of pharmaceutical processes using computer vision technology. Driven by curiosity and a commitment to excellence, I’m always looking for new ways to apply data and machine learning to push the boundaries of bioscience and engineering. I have worked with a range of different technologies and frameworks and I am always looking for opportunities to work with something new. The things that I currently have the most experience working with are: Python, PyTorch, C, C++, R, Julia, Java, SAS, QGIS, ArcGIS, git and AMPL.
Education
Ghent University
Ph.D. in Bioscience Engineering: Mathematical Modelling
2021-2025 (ongoing)
Focused in computer vision and machine learning. My research focuses on inferring behavior from videos and developing algorithms to characterize hierarchical structures based on pair-wise interaction data, primarily in animal studies. Additionally, I apply AI-driven techniques to pharmaceutical quality control, leveraging computer vision to ensure precision and safety in pharmaceutical processes.
KU Leuven
MSc in Statistics and Data Science (Magna Cum Laude)
2019 - 2021
Specialized in European Master Official Statistics (EMOS). The degree provided a solid theoretical and practical background in areas such as machine learning, time-series, inference, statistical modeling, and GIS.
Arizona State University, Tempe
BSE in Engineering Management
(Summa Cum Laude)
2013 - 2017
Specialized in Business Analytics. Received Moeur Award for exceptional academic achievement.
Experience
Euthority Project
Text Mining and Machine Learning
2019 - 2021
Developed machine learning models for analyzing legal documents, scrapping domestic and supranational court documents, and applying deep learning techniques to extract insights from legal text.
Ozel Fittings LTD
Project Management Specialist
2017 - 2018
Led projects to improve customer satisfaction and developed cost-reductive solutions within the distribution network.
ASU Admission Services
Student Worker
2013 - 2014
Managed sensitive personal data while assisting the administration in resolving student-related issues. Provided direct support to prospective students via phone and email.
Projects
A lexicon and rule-based sentiment analysis tool adapted for the Julia programming language.
VaderSentiment.jl is a port of the original VADER sentiment analysis tool implemented in Python. It is designed to analyze sentiment in text, particularly in social media, and provides nuanced sentiment scores based on contextual elements.
Flemish Statistical Authority (VSA)
Data Specialist Intern
2020
Utilized big data from Flemish business websites and applied natural language processing techniques to develop cost-effective official statistics on innovation.
EUROSTAT Coding Labs
Data Analyst Intern
2020
Developed a spatial model to predict population distribution using mobile network operator data, with a focus on applications in Belgium.
Treasures 4 Teachers, USA
Student Intern
2017
Focused on system design improvements for a non-profit organization, utilizing Six Sigma methodology to optimize processes.
Publications
Quantifying agonistic interactions between group-housed animals to derive social hierarchies using computer vision: a case study with commercially group-housed rabbits
https://www.nature.com/articles/s41598-023-41104-6This study develops a computer vision-based pipeline for detecting agonistic interactions between group-housed farm animals, specifically breeding rabbits, achieving 77% precision and 85% recall. The method enables the construction of socio-matrices and the derivation of dominance hierarchies with minimal human intervention.
Ipek, N., Van Damme, L. G., Tuyttens, F. A., & Verwaeren, J. (2023). Quantifying agonistic interactions between group-housed animals to derive social hierarchies using computer vision: a case study with commercially group-housed rabbits. Scientific Reports, 13(1), 14138.
Automated particle inspection of continuously freeze-dried products using computer vision
https://www.sciencedirect.com/science/article/pii/S0378517324008639This paper presents the use of computer vision in the pharmaceutical industry for the automated inspection of freeze-dried products. The YOLOv7 model achieved a particle detection precision of up to 88.9%, significantly outperforming manual inspection, with a processing time under 1 second per vial.
Herve, Q., Ipek, N., Verwaeren, J., & De Beer, T. (2024). Automated particle inspection of continuously freeze-dried products using computer vision. International Journal of Pharmaceutics, 664, 124629.
A deep learning approach to perform defect classification of freeze-dried products
https://www.sciencedirect.com/science/article/pii/S0378517324013619This study utilizes deep learning to classify cosmetic defects in freeze-dried products using high-resolution images. The best model achieved perfect precision and recall for critical defects, with a prediction time under 50 ms, improving the efficiency of quality control in the pharmaceutical industry.
Herve, Q., Ipek, N., Verwaeren, J., & De Beer, T. (2025). A deep learning approach to perform defect classification of freeze-dried products. International Journal of Pharmaceutics, 670, 125127.
Cage enrichment to minimize aggression in part-time group-housed female breeding rabbits
https://www.frontiersin.org/journals/veterinary-science/articles/10.3389/fvets.2024.1401021/fullThis study investigates the effect of cage enrichment on reducing aggression in group-housed female breeding rabbits. The results show that alfalfa blocks, with or without wooden panels, slightly reduce the number of injured does compared to controls, though the challenge of minimizing aggression remains significant in part-time group housing.
Van Damme, L. G., Ipek, N., Verwaeren, J., Delezie, E., & Tuyttens, F. A. (2024). Cage enrichment to minimize aggression in part-time group-housed female breeding rabbits. Frontiers in Veterinary Science, 11, 1401021.
Certifications
- 2017: Lean Green Belt Certificate
- 2017: Six Sigma Green Belt Certificate
- 2022: Alibaba Cloud Computing Associate (ACA)