ESTRO 2024 - Abstract Book

S46 ESTRO 2024 microenvironment and may also have different prominent antigens. It is therefore being discussed that RT-induced Invited Speaker

systemic immune effects could be better induced if tumour areas are irradiated at several locations in the body in the

metastatic situation. It has also to be considered the different radiosensitivity of immune cells. While T cells are more

radiosensitive, myeloid cells are quite resistant against RT. Again, novel RT techniques, such as FLASH-RT, may take

advantage of a very short-term irradiation of the blood pool and thereby spare the exposure of immune cells to

irradiation. This has to be further proven in future analyses.

From the immunological point of view a final answer whether volume of irradiated is more important than dose

cannot be given. But for the design of optimized multimodal clinical trials for treatment of solid tumors with RT and

immune therapies, multiple factors such as radiation dose, type of immunotherapy, sparing of lymph nodes, number

of irradiated lesions, individual immunophenotype of the tumour cells, as well as the sequence of therapies and the

tumour microenvironment have to be considered. Efficient local and systemic anti-tumour immune responses can

only be achieved through a combination of many factors.

3356

Impact of biased models in the context of fairness towards patients, and how to avoid or minimise biases in our datasets

Sanmi Koyejo

Stanford University, Computer Science, Stanford, USA

Abstract:

In only a few years, algorithmic fairness has grown from a niche topic to a significant component of medical imaging, machine learning, and artificial intelligence research and practice. As a field, we have had some embarrassing mistakes. Yet, our understanding of the core issues, potential impacts, and mitigation approaches has grown. This talk presents a range of recent findings, discussions, questions, and partial answers in algorithmic fairness in recent years. Further, this talk will explore the connections between algorithmic fairness and other ongoing research efforts. We will tackle some of the hard questions you may have about algorithmic fairness and address some misconceptions that have become pervasive.

3357

Impact of deformable image registration uncertainties for external beam radiotherapy applications

Florian Amstutz 1 , Lena Nenoff 2,3 , Martina Murr 4 , Ben Archibald-Heeren 5 , Marco Fusella 6 , Mohammad Hussein 7 , Wolfgang Lechner 8 , Greg Sharp 9 , Ye Zhang 10 , Eliana Vasquez Osorio 11 1 Inselspital, Bern University Hospital, and University of Bern, Division of Medical Radiation Physics and Department of Radiation Oncology, Bern, Switzerland. 2 OncoRay-National Center for Radiation Research in Oncology, Faculty of