Influence of the Antigen Pattern Vector on the Dynamics in a Perceptron-Based Artificial Immune - Tumour- Ecosystem During and After Radiation Therapy

Scheidegger, S; Barba, SM; Fellermann, H; Gaipl, U

doi:10.1007/978-3-031-23929-8_19

Influence of the Antigen Pattern Vector on the Dynamics in a Perceptron-Based Artificial Immune - Tumour- Ecosystem During and After Radiation Therapy

Lookup NU author(s): Dr Harold Fellermann

Downloads

Full text for this publication is not currently held within this repository. Alternative links are provided below where available.

Abstract

Artificial immune-tumor ecosystems can serve as models to explore the complex tumor-host-immune – interactions in silico. This may contribute to a better understanding of the conditions leading to anti-cancer immune response in patients during anti-cancer therapy. For model development, it is important to identify an appropriate model structure which is suitable to mimic the behavior of real biological systems. In this study, the influence of the number of antigens in an artificial adaptive immune system onto an immune-tumor ecosystem during and after radiation therapy (RT) is investigated. For antigen pattern recognition, a perceptron is used. The simulated scenarios with 4, 9 and 12 antigens exhibit differences in the immune response, but in all cases, perceptron weights for host tissue evolve after RT into negative values, leading to an immune-suppressive effect. This effect results from the evolution of the populations in the ecosystem and the training of the perceptron. In conclusion, the response of the proposed artificial immune system is strongly dependent on the ecosystem dynamics, which seems to be the case for the real biological systems as well.