Nusret Ipek

Bioscience Engineering, Computer Vision and Statistical Modelling

Automated behaviour inference for group-housed animals
Deep learning for pharmaceutical quality control

Email: nusret.ipek@ugent.be
 Phone: +32 474 02 36 34
Location: Leuven, Belgium

Profile





Portrait of Nusret Ipek
  • PhD candidate in Bioscience Engineering specialising in computer vision, statistical modelling, and machine learning for behaviour inference and pharmaceutical quality control.
  • Built pipelines for detecting agonistic interactions and deriving social hierarchies from videos and pairwise data; applied deep learning for automated inspection with sub-second inference.
  • Integrates computer vision with mathematical/statistical modelling to convert big data into reliable pairwise interpretations and quality-control signals.
  • Delivers validated, deployable analyses in academic and industrial settings.

Technical skills

  • Languages: Python; C and C++; Julia; R; Java; Bash; SAS
  • ML & CV: PyTorch; torchvision; YOLO family; OpenCV; scikit-learn; Hugging Face Transformers
  • Data & Tools: SQL; Git; Linux; Docker; NumPy; pandas; CUDA
  • Geospatial: QGIS; ArcGIS; GeoPandas

Education

Ghent University

PhD in Bioscience Engineering: Mathematical Modelling

2021 to 2025 (ongoing)

Research in computer vision, machine learning and mathematical modelling for behaviour inference.
AI-driven techniques for pharmaceutical quality control.

Thesis: Understanding Hierarchy Formation in Animal Groups through Probabilistic Modelling and Computer Vision

  • Supervisors: Jan Verwaeren (BIOVISM); Bernard De Baets (KERMIT); Frank Tuyttens (ILVO)
  • Expected defense: December 2025 - January 2026

KU Leuven

MSc in Statistics and Data Science (Magna Cum Laude)

2019 to 2021

European Master in Official Statistics (EMOS).
Training in machine learning, time series, inference, statistical modelling, and GIS.

Arizona State University, Tempe

BSE in Engineering Management (Summa Cum Laude)

2013 to 2017

Specialised in Business Analytics.
Awarded the Moeur Award for exceptional academic achievement.

Experience

Euthority Project

Text Mining and Machine Learning

2019 to 2021

  • Scraped and normalised domestic and supranational court documents; built a reproducible pipeline.
  • Trained legal-text classifiers, results reported in a peer-reviewed article (SSRN: 10.2139/ssrn.3734430).
  • Evaluated and deployed large language models (LLMs) for document classification.

Flemish Statistical Authority

Data Specialist Intern

2020

  • Analysed large-scale business website data; applied NLP to derive cost-effective innovation indicators.
  • Developed classifiers for business innovation estimators achieved 0.9 F1-score on held-out data.

Eurostat Coding Labs

Data Analyst Intern

2020

  • Developed a spatial model to predict population distribution using mobile network operator data for Belgium.
  • Outperformed baseline grid estimates, confirmed through cross-validation and map-based validation.

Özel Fittings Ltd.

Project Management Specialist

2017 to 2018

  • Led customer satisfaction initiatives across the distribution network
  • Optimised the distribution network and cost-to-serve model, cutting handling steps and routes.

Arizona State University, Admission Services

Student Assistant

2013 to 2014

  • Processed student records with attention to confidentiality and GDPR-like safeguards.
  • Authored SOP updates that reduced response time.

Publications

  1. AniDomNet: A sequential pairwise model for inferring dynamic animal dominance hierarchies
    • Headline: Sequential model operates at dyadic level relationships to infer dominance hierarchies and predict future interactions.
    • Journal: Methods in Ecology and Evolution
    • Citation: Ipek, N., Tuyttens, F. A. M., De Beats, B., & Verwaeren, J. (2025). Methods in Ecology and Evolution.
    • DOI:10.1111/2041-210X.70118
  2. Quantifying agonistic interactions between group-housed animals using computer vision
    • Headline: Computer vision pipeline to detect agonistic interactions and derive dominance hierarchies from video.
    • Journal: Scientific Reports
    • Citation: Ipek, N., Van Damme, L. G., Tuyttens, F. A. M., & Verwaeren, J. (2023). Scientific Reports, 13, 14138.
    • DOI: 10.1038/s41598-023-41104-6
  3. Automated particle inspection of continuously freeze-dried products using computer vision
    • Headline: YOLOv7-based inspection of freeze-dried products with sub-second per-vial inference.
    • Journal: International Journal of Pharmaceutics
    • Citation: Hervé, Q., Ipek, N., Verwaeren, J., & De Beer, T. (2024). International Journal of Pharmaceutics, 664, 124629.
    • DOI:10.1016/j.ijpharm.2024.124629
  4. A deep learning approach to perform defect classification of freeze-dried products
    • Headline: Real-time classification of cosmetic defects in high-resolution vial images.
    • Journal: International Journal of Pharmaceutics
    • Citation: Hervé, Q., Ipek, N., Verwaeren, J., & De Beer, T. (2025). International Journal of Pharmaceutics, 670, 125127.
    • DOI: 10.1016/j.ijpharm.2024.125127
  5. Cage enrichment to minimise aggression in part-time group-housed female breeding rabbits
    • Headline: Evaluation of alfalfa blocks with or without wooden panels for aggression reduction in does.
    • Journal: Frontiers in Veterinary Science
    • Citation: Van Damme, L. G., Ipek, N., Verwaeren, J., Delezie, E., & Tuyttens, F. A. M. (2024). Frontiers in Veterinary Science, 11, 1401021.
    • DOI:10.3389/fvets.2024.1401021

Awards

  • 2017 Moeur Award, Arizona State University.
  • 2022 BOF Scholarship, Ghent University.

Certifications

  • 2017 Lean Green Belt.
  • 2017 Six Sigma Green Belt.
  • 2022 Alibaba Cloud Computing Associate (ACA).

Languages

  • Turkish: native (CEFR equivalent C2)
  • English: C1 (CEFR)
  • Dutch: A2 (CEFR)
  • Chinese: A2 (CEFR)

Nusret Ipek — nusret.ipek@ugent.be