AI-assisted cancer image analysis

With global cancer incidence projected to rise by 47% by 2040, the demand for medical imaging in cancer care will increase substantially. This will worsen the existing shortage of diagnostic staff, making efficiency improvements crucial to maintaining accessible healthcare.

Radiology departments already generate vast amounts of CT scans of the abdomen and thorax for cancer patients, and interpreting these scans is both time-consuming and complex. AI-assisted reading has the potential to enhance the efficiency and quality of image-based medical diagnostics. However, most certified AI tools only evaluate scans from a single timepoint, overlooking the valuable insights gained from prior imaging—an essential component of radiologists’ assessments, especially for metastatic cancer, where follow-up imaging is crucial to evaluate treatment response.

In this research line, we aim to develop, validate, and deploy AI algorithms for cancer patients capable of combing data from multiple timepoints and integrate patients' history into the decision-making process. This research line is led by Alessa Hering.

Oncology AI group

Projects

COMFORT

We aim to improve the efficiency of kidney and prostate cancer screening by using artificial intelligence.

Read more →
ULS23

Building and Evaluating Universal Lesion Segmentation Models for Computed Tomography

Read more →
TotalReg

A foundation model for CT image registration

Read more →
OncoFuture

Improving tumor follow-up assessment

Read more →

People

Alessa Hering

Alessa Hering

Assistant Professor

Bram van Ginneken

Bram van Ginneken

Professor, Scientific Co-Director

Myrthe Buser

Myrthe Buser

Postdoctoral Researcher

Max de Grauw

Max de Grauw

PhD Candidate

Luc Builtjes

Luc Builtjes

PhD Candidate

Lena Philipp

Lena Philipp

PhD Candidate

Sarah de Boer

Sarah de Boer

PhD Candidate

Hartmut Häntze

Hartmut Häntze

PhD Candidate

Xuan Loc Pham

Xuan Loc Pham

PhD Candidate

Rianne Weber

Rianne Weber

PhD Candidate

Niels Rocholl

Niels Rocholl

PhD Candidate

Jan Tagscherer

Jan Tagscherer

PhD Candidate

Jen Dusseljee

Jen Dusseljee

Master Student

Publications

  • M. Grauw and B. Ginneken, "Semi-supervised 3D universal lesion segmentation in CT thorax-abdomen scans", European Congress of Radiology, 2022.
  • M. Grauw, B. Ginneken, B. Geisler, E. Smit, M. Rooij, S. Schalekamp and M. Prokop, "Deep learning universal lesion segmentation for automated RECIST measurements on CT: comparison to manual assessment by radiologists", European Congress of Radiology, 2022.
  • A. Hering, S. Hager, J. Moltz, N. Lessmann, S. Heldmann and B. van Ginneken, "CNN-based Lung CT Registration with Multiple Anatomical Constraints", Medical Image Analysis, 2021;72:102139.