AI@HZDR in the Research Field HEALTH

Artificial intelligence is transforming science as rapidly as it is reshaping other areas of our lives, and its impact on medicine is particularly pronounced. In the operating room, real-time algorithms are being developed to ensure that surgeries go as smoothly as possible. Research in this area is also being conducted at the National Center for Tumor Diseases (NCT) in Dresden, where experts are developing AI methods to recognize and evaluate patterns in the vast amounts of data generated by clinics and doctors’ offices. Similarly, HZDR’s Center for Advanced Systems Understanding is contributing its expertise to two European projects, while HZDR researchers at OncoRay – the National Center for Radiation Research in Oncology – are working to further improve the effectiveness of modern proton therapy.


Cancer Irradiation with Highest Precision

Foto: PT2030 Illustration 1 ©Copyright: Christian Richter (HZDR/OncoRay)

Online-adaptive Protonentherapie der nächsten Generation | Bild: HZDR/Christian Richter

Proton beams destroy tumors very effectively while sparing healthy tissue. However, this relatively new form of therapy cannot yet react precisely enough to movements during treatment or anatomical changes over the course of the therapy, which usually lasts several weeks.

This is why HZDR scientists are working closely with colleagues from the National Center for Radiation Research in Oncology - OncoRay to develop a fully automated feedback loop supported by AI algorithms for continuous monitoring of the therapy. The aim is to immediately adapt proton irradiation to anatomical changes before and during each individual treatment. For example, if the filling of the bladder, bowel or rectum changes, if the tumor shrinks or swells, and ultimately even if the breathing or movement of the tumor changes.

In a sub-project, work is being carried out together with the CASUS - Center for Advanced Systems Understanding on a fully automated, AI-based clinical decision support system. It should be able to connect to the treatment planning system, the treatment control system and the oncology information system. The clinical implementation of such an online-customized proton therapy requires efficient and secure solutions for the various tasks in the feedback loop. There should be direct and fast feedback to the clinical staff as well as a possibility for efficient and convenient retrospective review of the automated decisions.

Contact:
Prof. Christian Richter
Head Medical Radiation Physics at the HZDR Institute of Radiooncology – OncoRay

More information:
►Press release: ProtOnART – a new consortium for proton online adaptive radiation therapy

►Website PT2030 – Next generation proton therapy: online adaptive


Artificial Intelligence Improves the Diagnosis and Treatment of Cancer

Foto: Logo CASUS ©Copyright: CASUS

X-rays, blood samples, DNA tests - clinics and medical practices collect enormous amounts of data every day. Artificial intelligence (AI) can link this data together and detect hidden patterns. The HZDR Institute CASUS - Center for Advanced Systems Understanding is creating important foundations for AI algorithms that will significantly improve the diagnosis and treatment of tumors.

CASUS is significantly involved in two major European projects: The PIONEER research consortium is bringing together 100 million data sets on prostate cancer research, thereby creating an essential prerequisite for the development of new algorithms. Building on this, the OPTIMA (Optimal Treatment for Patients with Solid Tumors in Europe Through Artificial intelligence) project aims to design AI systems that improve the diagnosis and treatment of prostate, breast and lung cancer. Both PIONEER and OPTIMA are funded by the Innovative Medicines Initiative (IMI) program of the European Union and the European Federation of Pharmaceutical Industries and Associations (EFPIA).

Contact:
Dr. Michael Bussmann
Deputy Director CASUS – Center for Advanced Systems Understanding

More information: 
►Press release: Paving the way for machine learning with medical data