PhD-position: Implementation of AI-based contouring and treatment planning for daily personalized RT treatments for lung cancer

Since several years we investigate the use of convolutional neural networks (CNNs) to improve the quality of the preparation of a radiotherapy treatment plan, e.g. for head and neck cancer. For other indications investigations are ongoing by dedicated physicists of the RT department and Engineers of the Medical Imaging Research Center.  Besides an improvement in quality, it is seen that AI also decreases the time needed to prepare a treatment plan, making it suitable to use it in adaptive radiotherapy, where the plan is renewed during the course of the treatment. The current project aims at studying the advantages and challenges of the clinical implementation of AI in the daily routine of the treatment of Non-small-cell lung cancer (NSCLC) in the radiotherapy department of the University Hospitals of Leuven. The general research objective is to investigate the potential of a CBCT-only workflow for online adaptive RT for NSCLC.

You will first perform a retrospective analysis and do the needed data farming to investigate the use of in-house and commercial AI solutions for both auto contouring and treatment planning on CBCT and/or synthetic CT and thus the possibility to use it in an adaptive workflow. Then you will use this knowledge to setup and execute a clinical study to investigate the use of these AI based solutions in a clinical adaptive workflow and to evaluate the impact of a CBCT only workflow. You will collaborate closely with the clinical team and researchers investigating new AI-models.

The project will be executed by an inspiring collaboration with the Medical Imaging Research Center and partly sponsored by Varian Medical system in the project “Optimization free treatment planning to boost the speed of adaptive radiotherapy within clinical viable time slots”.

You will investigate the clinical implementation of AI in daily routine for patients with Lung cancer, treated with radiotherapy at the radiotherapy department of the University Hospitals of Leuven. The general research objective is to investigate the potential of a CBCT-only workflow for online adaptive RT for NSCLC. The research tasks include: retrospective analysis and data farming for AI-based contouring and treatment planning on CBCT and synthetic CT and setting up and executing a clinical study implementing the previously designed tools on repeat 4DCT. You will collaborate closely with the clinical team and other researchers investigating new AI-models. The end goal is a clinical study evaluating the impact of a CBCT only workflow.