PhD defence R. (Ronald) Booij

A Computed Tomography (CT) scanner is a machine that provides cross-sectional images by the use of x-rays. With technological innovations, diagnostic capabilities improved in both adult and pediatric patients. However, optimization of a CT scan procedure became more and more challenging because of several factors: interrelation of numerous scan and reconstruction parameters, increased workload and human error. In addition, the variety of patients and referral questions necessitates different acquisition strategies. Therefore, artificial intelligence (AI) is integrated in most of the steps of the imaging chain to make procedures more accurate, standardized and supports the users. Even though AI is able to “learn” from analyzing more and more cases, it remains important to know and monitor their performance, as well as to determine their accuracy. This thesis focuses on evaluation and implementation of technological solutions to automate and / or improve distinct aspects of the CT imaging chain. We conclude that knowledgeable solutions like AI should be embraced, as they can be an important factor in assisting medical healthcare workers and may also lighten the workload. In addition, a high knowledge base among the users of intelligent machines is essential for the development and correct application of AI. The human side remains necessary for further optimization in clinical practice and for future developments. Therefore, optimization is achieved by deploying smart technologies and by merging of knowledge obtained by man and machine, creating opportunities for a “knowledgeable” CT scanner. Ultimately, this may lead to a symbiosis between man and machine.

Due to corona, the PhD defences do not take place publicly in the usual way in the Senate Hall or in the Professor Andries Querido Room. The candidates will defend their dissertation either in a small group or online.