- Startseite
- Forschung
- Artificial Intelligence in Neuroradiology
- Neuroradiologische Forschung
- Artificial Intelligence in Neuroradiology
- Cardiovascular Imaging
- Multiparametric Imaging and Radiomics
- Machine Learning and Data Science Group
- Experimental Imaging and Image-Guided Therapy
- Forschung mit Racoon
- Data Science
- Kinderradiologie
- Onkologische Bildgebung
- Senologie
- Klinische Studien
- Computed Tomography Research
- Muskuloskelettale Bildgebung
- Translational Imaging
Artificial Intelligence in Neuroradiology
Research Topics
Neurooncology
Our AI research group aims to develop novel methods to improve brain imaging and establish new approaches for non-invasive tumor characterization. Recent advances in processing of imaging data with use of Artificial Intelligence and introduction of quantitative MR imaging sequences provide numerous opportunities for clinicians to improve patient management and therapy planning.
One of our main interests is to automatize brain tumor detection and consecutive segmentation using Deep Learning based techniques to free up radiologists’ resources and enhance reproducibility of imaging data. In combination with newly developed imaging sequences such as MR Fingerprinting, we want to determine tumor biology, non-invasively. Finally, we use Artificial Intelligence for studies regarding non-invasive multiparametric diagnostics and studies evaluating multimodal tumor therapy.
One of our current projects aims to develop methods for distinguishing true tumor progression from therapy-related changes in glioma patients undergoing treatment, with one approach utilizing quantitative MRI imaging and another involving deep learning algorithms, making it more accessible compared to the alternative FET-PET imaging. Additionally, advanced language analysis will be employed to effectively summarize the complex disease courses of glioma patients, thereby enhancing their treatment and resource allocation.
Neurovascular Imaging
Another focus of our research group is the automated detection of neurovascular pathologies with special interest in intracranial aneurysms. Further, we want to use automatically generated segmentations to stratify patient risk and facilitate endovascular or surgical treatment.
Grant support
- Else Kröner-Fresenius-Stiftung (Kai Laukamp & Michael Schönfeld)
- Institut für Diagnostische und Interventionelle Radiologie, University Hospital Cologne (Kai Laukamp & Michael Schönfeld)
- Faculty of Medicine and University Hospital Cologne (Kai Laukamp)
- Cologne Clinician Scientist Programm (David Zopfs)
- Else Kröner-Fresenius-Stiftung (Simon Lennartz)
Collaborations
- Institute for Diagnostic Radiology, Neuroradiology and Nuclear Medicine, Johannes Wesling Klinikum Minden, University Hospitals of the Ruhr-University Bochum, Germany
- Department of Neurosurgery and Stereotaxy, Center for Neurosurgery, Faculty of Medicine and University Hospital Cologne, University of Cologne, Germany
- Institute for Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen, Germany
- Department of Diagnostic and Interventional Radiology, University Hospital Würzburg, Germany
- Department of Diagnostic and Interventional Neuroradiology and Pediatric Neuroradiology, University Hospital Bonn, Germany
- Institute of Diagnostic and Interventional Radiology, University Hospital Bonn, Germany
- Philips GmbH Innovative Technologies, Aachen, Germany
- Lung Cancer Group, Department I of Internal Medicine, Faculty of Medicine and University Hospital Cologne, University of Cologne, Germany
- Monitoring Patients with Glioblastoma by Using a Large Language Model: Accurate Summarization of Radiology Reports with GPT-4. Radiology. 2024
- Highly compressed SENSE accelerated relaxation-enhanced angiography without contrast and triggering (REACT) for fast non-contrast enhanced magnetic resonance angiography of the neck: Clinical evaluation in patients with acute ischemic stroke at 3 tesla. Magn Reson Imaging. 2024
- Potential of GPT-4 for Detecting Errors in Radiology Reports: Implications for Reporting Accuracy. Radiology. 2024
- GPT-4 for Automated Determination of Radiological Study and Protocol based on Radiology Request Forms: A Feasibility Study. Radiology 2023
- Automated Color-Coding of Lesion Changes in Contrast-Enhanced 3D T1-Weighted Sequences for MRI Follow-up of Brain Metastases. AJNR Am J Neuroradiol. 2022
- Virtual non-contrast reconstructions improve differentiation between vascular enhancement and calcifications in stereotactic planning CT scans of cystic intracranial tumors. Eur J Radiol. 2022
- Two-dimensional CT measurements enable assessment of body composition on head and neck CT. Zopfs D, Pinto Dos Santos D, Kottlors J, Reimer RP, Lennartz S, Kloeckner R, Schlaak M, Theurich S, Kabbasch C, Schlamann M, Große Hokamp N. Eur Radiol. 2022
- Differentiation of Intracerebral Tumor Entities with Quantitative Contrast Attenuation and Iodine Mapping in Dual-Layer Computed Tomography. Diagnostics. 2022
- Contrast-Enhanced Black Blood MRI Sequence Is Superior to Conventional T1 Sequence in Automated Detection of Brain Metastases by Convolutional Neural Networks. Diagnostics 2021
- Fully Automated MR Detection and Segmentation of Brain Metastases in Non-small Cell Lung Cancer Using Deep Learning. J Magn Reson Imaging. 2021
- Deep learning assistance increases the detection sensitivity of radiologists for secondary intracranial aneurysms in subarachnoid hemorrhage. Neuroradiology 2021
- Automated Detection and Segmentation of Brain Metastases in Malignant Melanoma: Evaluation of a Dedicated Deep Learning Model. Am J Neuroradiol. 2021
- Fully automated detection and segmentation of intracranial aneurysms in subarachnoid hemorrhage on CTA using deep learning. Sci Rep. 2020
- Primary Central Nervous System Lymphoma: Clinical Evaluation of Automated Segmentation on Multiparametric MRI Using Deep Learning. J Magn Reson Imaging. 2020
- Automated Meningioma Segmentation in Multiparametric MRI: Comparable Effectiveness of a Deep Learning Model and Manual Segmentation. Clin Neuroradiol. 2020
- MRI Follow-up of Astrocytoma: Automated Coregistration and Color-Coding of FLAIR Sequences Improves Diagnostic Accuracy With Comparable Reading Time. J Magn Reson Imaging. 2020
- Accuracy of Radiomics-Based Feature Analysis on Multiparametric Magnetic Resonance Images for Noninvasive Meningioma Grading. World Neurosurg. 2019
- Fully automated detection and segmentation of meningiomas using deep learning on routine multiparametric MRI. Eur Radiol. 2019
- Follow-up MRI in multiple sclerosis patients: automated co-registration and lesion color-coding improves diagnostic accuracy and reduces reading time. Eur Radiol. 2019
- Clinical Evaluation of a Multiparametric Deep Learning Model for Glioblastoma Segmentation Using Heterogeneous Magnetic Resonance Imaging Data From Clinical Routine. Invest Radiol. 2018
Senior Advisors
Univ.-Prof. Dr. Jan Borggrefe, MD (Johannes Wesling Klinikum Minden - University Hospitals of the Ruhr-University Bochum)
Univ.-Prof. Dr. Norbert Galldiks, MD (University Hospital Cologne)
Dr. Jan-Michael Werner, MD (University Hospital Cologne)
Prof. Dr. Christoph Kabbasch, MD (University Hospital Cologne)
Priv.-Doz. Dr. Simon Lennartz, MD (University Hospital Cologne)
Dr. David Zopfs, MD (University Hospital Cologne)
Group members
Dr. Liliana Caldeira, PhD
Dr. Mirjam Schöneck, PhD
Robert Hahnfeldt, MD
Thomas Schömig, MD
Dr. Thomas Dratsch, MD
Dr. Carsten Gietzen, MD
Dr. Lukas Goertz, MD
Dr. Jan-Peter Grunz, MD
Dr. Stephanie Jünger, MD
Dr. Rahil Shahzad, PhD
Dr. Marco Timmer, MD
Mr. Frank Thiele, MSc
Dr. Nicolas Lopez Armbruster, MD