AYDIN
DEMIRCIOGLU

Publications (2,453 citations, h-index 24, i10-index 40)

1. 2026
   Measuring the optimistic bias of cross-validation in radiomics.

   Demircioglu — Scientific Reports — PubMed
2. 2026
   Deep learning-based automatic field of view planning for prostate MRI in oblique coronal and oblique axial planes.

   Quinsten et al. — Scientific Reports — PubMed
3. 2025
   Evaluation metrics in medical imaging AI: fundamentals, pitfalls, misapplications, and recommendations.

   Kocak et al. — Eur J Radiol AI — Publisher
4. 2025
   Reproducibility and interpretability in radiomics: a critical assessment.

   Demircioglu — Diagn Interv Radiol — PubMed
5. 2024
   METhodological RadiomICs Score (METRICS): a quality scoring tool for radiomics research endorsed by EuSoMII.

   Kocak et al. — Insights into Imaging — PubMed
6. 2024
   The effect of feature normalization methods in radiomics.

   Demircioglu — Insights into Imaging — PubMed
7. 2024
   Applying oversampling before cross-validation will lead to high bias in radiomics.

   Demircioglu — Scientific Reports — PubMed
8. 2023
   Are deep models in radiomics performing better than generic models? A systematic review.

   Demircioglu — European Radiology Experimental — PubMed
9. 2022
   Benchmarking feature selection methods in radiomics.

   Demircioglu — Investigative Radiology — PubMed
10. 2022
    The effect of preprocessing filters on predictive performance in radiomics.

    Demircioglu — European Radiology Experimental — PubMed
11. 2022
    The frozen elephant trunk technique: impact of proximalization and the four-sites perfusion technique.

    Tsagakis et al. — Eur J Cardiothorac Surg — PubMed
12. 2022
    Evaluation of the dependence of radiomic features on the machine learning model.

    Demircioglu — Insights into Imaging — PubMed

Code

• radMLBench — 50 public radiomics datasets.
• radFS — feature-selection benchmarking.
• radCV — cross-validation bias studies.

Background

I am currently with the Institute of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen. I work on radiomics and quantitative medical imaging, mainly the reproducibility of radiomic features and the optimistic bias from data leakage in model validation. I also develop deep learning models for scan-range planning and dose reduction. Earlier, I worked on large-scale kernel methods and SVMs at the Institute for Neuroinformatics, Ruhr University Bochum, and before that at several companies on image processing. I hold a PhD in mathematics from the University of Potsdam and a diploma in mathematics from the University of Wuppertal.

Contact

Aydin Demircioglu
Institute of Diagnostic and Interventional Radiology and Neuroradiology
University Hospital Essen
Hufelandstraße 55
45147 Essen
hello@aydindemircioglu.de
Github · Research Gate