On Thursday 12 June at 15:00, Kim Møller Hochreuter defends his PhD dissertation entitled "Combined Deep Learning and Advanced Imaging for Improved Individualized Radiotherapy Target Definition in Glioblastoma Patients. A National DNOG Study".

Glioblastoma is the most aggressive form of brain cancer, often spreading in ways that standard scans cannot detect. Radiation therapy is a key part of treatment, but defining where to target the radiation most effectively remains a challenge. Traditionally, doctors expand the visible tumor area evenly in all directions, even though the cancer tends to spread along specific pathways in the brain’s white matter.

This PhD project explored whether combining artificial intelligence (AI) with advanced MRI scans can improve how radiation targets are defined. AI was used to automatically outline tumors on brain scans with high precision. At the same time, MRI scans that map brain structure were used to guide more realistic models of tumor spread.

The results show that careful manual edits to training data improve AI accuracy, and that tumor spread can be modeled in a way that reflects the brain’s structure. These methods could lead to more personalized and precise radiation therapy for patients with glioblastoma.

The summary is written by the PhD student.

The defence is public and takes place in auditorium C114-101, Aarhus University Hospital. Please see the press release for more information. 

Contact

PhD student Kim Møller Hochreuter
Mail: kim.hochreuter@ph.au.dk    
Phone: 29732718

Read full press release