What we do
About our project
Background
Radiotherapy departments are facing increasing pressure as cancer incidence continues to rise while staff relatively declines. Efficiency can be improved by both automation and reducing the number of treatment fractions per patient (hypofractionation). However, hypofractionation schedules are typically population-based (“one-size-fits-all”) and depend on long waiting times for trials to be performed. There is no established method for personalized hypofractionation based on patient-specific expected toxicity.
Aim
The main aim of the project is to develop and evaluate an AI-assisted, personalized approach to reduce the number of radiation treatment fractions with minimal increase in toxicity risks compared to current treatment schedules.
Study population
The intended study population consists of patients with lung, head-and-neck, and esophageal cancer, treated with photon and proton therapy.Our research focus
Feasibility of AI-based personalized hypofractionation
We will investigate if AI-based personalized hypofractionation (automated planning for patient-specific fraction reduction) can be effectively applied in lung, head-and-neck, and esophagus photon and proton therapy.Impact on expected toxicity risk
The relationship between the reduction in fraction number and its impact on organ-at risk doses and normal tissue complication probabilities will be investigated.Clinical assessment
Applicability of AI-based personalized hypofractionation in clinical practice will be explored by creating a dedicated graphical user interface for plan selection by physicians.Funds & Grants
This study is funded by Varian Medical Systems, Inc., a Siemens Healthineers Company.
Collaborations
Our team
Beatriz Fernandes – Project coordinator & PhD candidate
Linda Rossi
Jos Elbers
Sebastiaan Breedveld
Ben Heijmen
