6 PhD positions on Computer Vision and Machine Learning for Medical

In the past decade Deep Learning, Machine Learning, and Computer Vision has progressed with leaps and strides, to the extent that these technologies can be readily used in practice in many applications. One of the most prominent applications is health and oncology. Cancer is the second most frequent cause of death worldwide. The time is ripe to combine the power of Deep Machine Learning, Computer Vision, and pristine Oncological knowledge and data for an application with tremendous impact: battling cancer with adaptive radiotherapy. Adaptive radiotherapy is state-of-the-art medical technology for killing the tumor by radiation. While initially the radiotherapy equipment mostly relies on pre-recorded treatment plans, the idea is to use Artificial Intelligence (AI) for dynamic and patient-specific treatment plans, using CT-based imaging to recognize, analyze, and model the behavior of the tumor`real time'.To make this vision a reality, we have formed a Partnership for Online Personalized and AI-enabled Adaptive Radiotherapy project, between the University of Amsterdam (UvA), the Netherlands Cancer Institute (NKI), and Elekta, or POP-AART for short.

The ambitious POP-AART project combines the cutting-edge expertise in AI by the UvA, in Oncology by the NKI, and in manufacturing state-of-the-art adaptive radiotherapy equipment by Elekta. The UvA has a strong track record in AI and machine learning. The NKI is a global leader and internationally recognized center for cancer research and treatment of cancer patients. Elekta Oncology Systems produces state of the art equipment and software for radiation oncology. Together they collaborate in the POP-AART lab. The lab will focus on the use of AI to improve precision radiation in oncology. The UvA and the NKI have identified 6 relevant research directions for this project and are hiring PhD students in computer vision and machine learning for POP-AART.

In particular, we are looking for PhD candidates on the following projects:

Project 1: Deep Generative Learning for Cone Beam Computed Tomography;
Project 2: Learning Inverse Models for Cone Beam Computed Tomography Reconstruction;
Project 3: Deep Learning Geometry for 3D Medical Image Registration';
Project 4: Interactive Deep Learning for Medical Segmentation;
Project 5: Learning to Forecast for Adaptive Radiation Therapy;
Project 6: Reinforcement Learning for Radiation Treatment Plan optimization.