On November 30th , Elfi Hofmeijer from the UT will receive the Philips Graduation Award for Data Science and Artificial Intelligence in Health Care (which involves a sum of 3,000 euros). This award is in recognition of her research into artificial intelligence (AI). Her research was part of a project aimed at better integrating AI into hospitals’ radiology departments, by means of training and with the aid of technical tools. This mainly involves enhancing the screening process for three diseases: lung cancer, chronic obstructive pulmonary disease (COPD) and cardiovascular disease.
Elfi Hofmeijer’s research concerned the simulation of CT images of lungs with visible lung nodules. The malignant variant of these lung nodules could be the result of lung cancer. These are simulated medical images, which means that they were not obtained from real patients. However, they do need to look sufficiently realistic that there is no way to distinguish them from the real thing. Hofmeijer explains that “To assess the simulated images, I also explored current methods for determining whether simulated medical images appear genuinely realistic.”
Hofmeijer explains why this research is so important: “If we can realistically simulate medical images and have influence on some of their characteristics, this has two major advantages. Firstly, current AI-based detection algorithms can be improved and, secondly, a wide variety of training material can be created for use in performing practice diagnoses by radiologists. For example, lung nodules can occur in different parts of the lung and can have different shapes. We want to be able to adjust these characteristics. If this is successful, it would give us the ability to generate additional images of relatively uncommon disease types. It could also be used to extend the training of current AI-based detection techniques. In addition, this method has the potential to generate extra training material for educational programmes in Radiology.”
Hofmeijer is currently pursuing her PhD research at the Robotics and Mechatronics (RAM) department at the University of Twente’s Faculty of Electrical Engineering, Mathematics and Computer Science (EEMCS). She is working on the development of a Medical Image Simulator. Her research is part of a work package associated with the B3Care project sponsored by the ZonMw. The project is headed by dr. ir. Ferdi van der Heijden and Prof. dr. ir. C.H. Slump.
This year marks the fifth occasion on which the Philips Graduation Award has been conferred. With this award, Philips is keen to emphasise the importance of innovations in the field of Biomedical Technology. This is just one of the avenues being explored by Philips HealthTech, in keeping with its goal of delivering sustainable healthcare for the future.