Medical Image Analysis with Machine Learning (MEDiml): Strategies for Simpler Radiomics

Abstract

Radiomics extracts quantitative features describing tumor shape, intensity, and texture from medical images to build models that predict clinical outcomes. However, large feature sets increase the risk of overfitting in small datasets, and building an end-to-end workflow often requires programming expertise. We present MEDiml (Medical Image Analysis with Machine Learning), open-source radiomics software available as a Python package (MEDiml-pkg) and a no-code desktop application (MEDiml-app). Both run on the same computational engine, and code generation exports graphically configured workflows as executable Python code. Feature extraction follows the Image Biomarker Standardization Initiative (IBSI) standards. MEDiml is built around complexity minimization: features are grouped into levels of increasing complexity (morphological, intensity, textural,l and filter-based), a model is trained at each level, and the simplest level not significantly outperformed by more complex ones is retained. Applied to five oncological datasets (2,104 scans), MEDiml achieved AUCs of 0.65 to 0.82, with morphological or intensity features sufficient in four of the five tasks. Resources: www.mediml.app.

Quantum Machine Learning in Medical Imaging: A Primer for Radiologists
Ruben Ortiz Cordero, Gaven K. Smart, Harshada Gothankar +3
arXiv23 Oct 2026PreprintCC BY 4.0Full textQuantum physics & quantum informationarXiv
Phase-edge imaging using q-plate shifts for faster and simpler microscopy
Jigme Zangpo, Hirokazu Kobayashi, Ryo Yasuhara
arXiv21 May 2026PreprintCC BY 4.0Full textOptics & photonicsarXiv
Magic Gate Teleportation: Structure, Useful Resource States, and Simpler Feedforward
Yunzhe Zheng, Allen Zang, Aleksander Kubica
arXiv10 Oct 2026PreprintCC BY-NC-ND 4.0Full textQuantum physics & quantum informationarXiv
Plasmonic Lithography Optical Proximity Correction Method Based on Decomposed Machine Learning Fast Imaging Model
zhenyu xing, Huwen Ding, pengyu ren +1
Optics Express9 Oct 2026Peer-reviewedLicence not statedOptics & photonicsCrossref
Compact and stable representation of real-frequency spectral functions for machine learning
Xinyang Dong, Ido Zemach, Lei Zhang +2
Physical Review Research3 Nov 2026Peer-reviewedCC BY 4.0Atomic, molecular & optical physicsCrossref

Cite

BibTeX RIS

BibTeX

@misc{loutfi2026medical,
  title = {{Medical Image Analysis with Machine Learning (MEDiml): Strategies for Simpler Radiomics}},
  author = {Mahdi Ait Lhaj Loutfi and Simon Provost and Élodie Gillard and Moustafa Amine Bezzahi and Alexander Chum and Ben Coull-Neveu and Juan Duran and Mariem Kallel and Hakima Laribi and Oliver Lefebvre and Nicolas Longchamps and Thanh-Tri Nguyen and Steven Ravary and Ouael Nedjem Eddine Sahbi and Jérémy Dana and Martin Vallières},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.37715},
  eprint = {2610.37715},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.37715},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Medical Image Analysis with Machine Learning (MEDiml): Strategies for Simpler Radiomics
AU  - Mahdi Ait Lhaj Loutfi
AU  - Simon Provost
AU  - Élodie Gillard
AU  - Moustafa Amine Bezzahi
AU  - Alexander Chum
AU  - Ben Coull-Neveu
AU  - Juan Duran
AU  - Mariem Kallel
AU  - Hakima Laribi
AU  - Oliver Lefebvre
AU  - Nicolas Longchamps
AU  - Thanh-Tri Nguyen
AU  - Steven Ravary
AU  - Ouael Nedjem Eddine Sahbi
AU  - Jérémy Dana
AU  - Martin Vallières
PY  - 2026
DA  - 2026/10/27
PB  - arXiv
DO  - 10.48550/arxiv.2610.37715
UR  - https://arxiv.org/abs/2610.37715
AB  - Radiomics extracts quantitative features describing tumor shape, intensity, and texture from medical images to build models that predict clinical outcomes. However, large feature sets increase the risk of overfitting in small datasets, and building an end-to-end workflow often requires programming expertise. We present MEDiml (Medical Image Analysis with Machine Learning), open-source radiomics software available as a Python package (MEDiml-pkg) and a no-code desktop application (MEDiml-app). Both run on the same computational engine, and code generation exports graphically configured workflows as executable Python code. Feature extraction follows the Image Biomarker Standardization Initiative (IBSI) standards. MEDiml is built around complexity minimization: features are grouped into levels of increasing complexity (morphological, intensity, textural,l and filter-based), a model is trained at each level, and the simplest level not significantly outperformed by more complex ones is retained. Applied to five oncological datasets (2,104 scans), MEDiml achieved AUCs of 0.65 to 0.82, with morphological or intensity features sufficient in four of the five tasks. Resources: www.mediml.app.
ER  -

Abstract

Related

Cite