Research Projects

Project Name: Human Rules - AI Brains: Towards Expert-Guideline Conformance for Machine-Learning-based Segmentations of Clinical Target Volumes

Project Duration: 2020 – 2024 (doctoral project)

Funded by: HIDSS4Health – Kristina Giske, Oliver Jäkel, Martin Frank (KIT) https://www.hidss4health.de/  => Projects 2020 => Surgery & Intervention 4.0

Patient-tailored contours of target volumes are fundamental for radiation treatment planning and thus, the outcome of the cancer treatment. Delineation of clinical target volumes on planning CT scans is challenging for human experts, is extremely time consuming, and shows large variation between observers. State-of-the-art machine learning algorithms reach high accuracy on the automatic segmentation of anatomical structures which is yet not transferable to target volumes without additional constraints. The translation of the expert guidelines into the machine learning realm can advance automated target volume delineation to facilitate its guideline conformance and its clinical use. In this HIDSS4Health-funded doctoral project Alexandra Walter tackles the implications of supervised learning on data prone to inter-observer variabilities and patient-individual differences utilizing mathematical rules given by human experts.

Click here for more details. 

Project Name: High-, moderate- and low-LET ion beams in the treatment of radioresistant tumors: Impact of beam quality, tumor grading, and hypoxic status on radiation response.

Project Duration: 04/2020 - 01/2025 (2nd funding period)

Funded by: Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) -- Project number 319589509

Besides allowing for highly conformal irradiations, ion beams also exhibit an increased radiobiological effectiveness (RBE), which may be suited to improve the effectiveness in the treatment of hypoxic tumors. This project investigates the RBE of experimental prostate tumors in the rat under normoxic (oxygen breathing) and hypoxic (clamping of tumor-supplying artery) conditions after irradiation with photons, protons, helium, carbon and oxygen ions and quantifies the oxygen enhancement ratio (OER). These investigations are accompanied by detailed histological characterization of the radiation response. See also Glowa et al. 2024 (DOI:10.1016/j.ijrobp.2024.05.004)

Click here for more details. 

Project Name: matRad – an open-source dose calculation and treatment planning toolkit

Funding Period: 2022 – 2025

Funding Information: matRad development is currently funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) -- Project No. 443188743: "Sustainable development of the open source radiotherapy dose calculation and optimization toolkit matRad"

matRad is an open source software for radiation treatment planning of intensity-modulated photon, proton, and carbon ion therapy. matRad is developed for educational and research purposes and entirely written in MATLAB. matRad is published on GitHub (https://github.com/e0404/matRad) and is used I numerous in-house, collaborative, and independent international research projects.

matRad implements established dose calculation algorithms for photons, protons and carbon ions as well as non-linear constrained biological treatment plan optimization. It includes open patient and machine data and features a graphical user interface and a powerful scripting API. During the funding period, matRad will be extended to facilitate interfacing of open Monte Carlo dose calculation techniques, inclusion of helium and advanced biological models, and working with matRad from Python. Further, development will be professionalized to follow continuous integration standards and automated tests & builds.

Click here for more details.

Project Name: anthropomorphic abdominal Pancreas Phantom for Ion-beam Therapy (PPIeT)

Project Duration: 05/2021 - 05/2024 (doctoral project)

Pancreatic cancer has a five-year survival rate of around 5 % to 10 %, and imposes challenges in radiotherapy due to different factors like tumor resistance and breathing-induced organ motion. To address this, we designed and constructed an anthropomorphic pancreas phantom for studying dose delivery during carbon ion irradiation. Thanks to the expertise in developing anthropomorphic phantoms in our research group, it was possible to include deformable elements and 3D-printed organs that can reproduce breathing-induced organ motion and allowed for testing various treatment modalities. This research project contributes valuable insights to advance ion-based cancer treatments, emphasizing tailored strategies accounting for organ motion management techniques aimed at improving the outcome of radiotherapy.

Click here for more information about our research group. 

Project Name: In-vivo monitoring of carbon ion radiotherapy delivery (InViMo)

Project Duration:  2019 - until today

Funded by: National Center for Tumor Diseases (NCT), Funding Program "Proof-of-Concept Clinical Trials" (funding period: 2019 - 2023)

During the radiation treatments of patients by carbon ions, a wide spectra of secondary radiation is leaving the patient. We developed a method to track these single secondary ions. By analysing their paths we draw conclusions about possible internal changes in the patient's tissue along the beam path (e.g. filling of the nose cavity), which might be critical for the local controll of the dissease. Since secondary radiation is escaping the treated patient as a by-product of the treatiment, this kind of imaging does not require any additional radiation dose to the patient. Following year-long research on patient models, we are about to start a clinical study at the Heidelberg Ion Beam Therapy Center.

Click here for more details. 

Project Name: Reducing secondary cancer risk by measuring neutron exposure in light ion beam radiotherapy

Project Duration: 2023 - 2026

Funded by: Deutsche Forschungsgemeinschaft (DFG, German Research Foundation), funding number 495217943

With particular relevance for paediatric patients and radiotherapy during pregnancy, the effects of secondary neutrons in proton and light ion beam therapy need to be further investigated. It is, therefore, necessary to improve the current neutron detection techniques, particularly with a focus on the high-energy neutrons generated during particle therapy. For this, this project is currently implementing Fluorescent Nuclear Track Detectors (FNTDs) as a technique to improve secondary neutron dose quantification. Even considering previous reports where these secondary neutron doses are rather low, improving their experimental quantification will help to better estimate the potential long-term risk. As a long-term goal, including this information in the treatment planning systems would make it possible to select improved treatment for the patients.

Click here for more information about the research group

Project Name: Online adaptive radiotherapy for locally advanced lung cancer: assessing the patient benefit in a multi-centric comparison of CBCT- and MRI-guidance approaches

Project Duration: 03/2023 - 02/2028

Funded by: German Cancer Aid, funding number 70114708 => 8. Ausschreibungsrunde (2021)

Within the Priority Program Translational Oncology of the German Cancer Aid, we were successful to obtain funds for a collaborative research project and clinical study for radiotherapy in lung cancer patients. For our project application entitled Online adaptive radiotherapy for locally advanced lung cancer: assessing the patient benefit in a multi-centric comparison of CBCT- and MRI-guidance approaches, we were able to obtain a 1.5M€ grant to perform a clinical trial for MR-guided radiotherapy, which has been initiated at the three participating MR-linac centers at the University Hospitals in Heidelberg, Tübingen and the Ludwigs-Maximilians-University (LMU) in Munich and which will soon be started at the new ETHOS® system at DKFZ.

Click here for more information. 

Project Name: AI-Based Risk Assessment for Radiation-Induced Lung Injury in Lung Cancer SBRT

Project Duration: 2022 - 2024

Funded by: Wilhelm Sander-Stiftung, funding number 3010001119

In the multifaceted field of lung cancer research, distinct biomarkers have been discovered through separate omics analyses, encompassing clinical phenomics, radiomics, dosiomics, and biomolecular omics. These biomarkers provide crucial insights, captured both longitudinally over time and at specific instances. Importantly, the diverse datasets derived from these omics analyses are routinely collected and available when a patient undergoes radiotherapy treatment. Our project seeks to harness this information by developing an AI-based workflow that will integrate these different layers of data. The objective is to accurately predict the risk of radiation-induced lung injury and distinguish it from cancer recurrence. This critical differentiation aids in selecting the optimal clinical therapeutic strategies to manage post-radiotherapy complications, ultimately guiding clinicians toward a more personalized, data-driven approach to lung cancer treatment.

Click here for more information.