For the past month, we’ve had the pleasure of getting to know Florian van Leeuwen, a PhD Candidate in Statistics at Utrecht University. Florian’s PhD focuses on prediction models for many different types of complex datasets, but he has specific interests in applying these models to cases of brain injury. We’ve been hosting him in Philadelphia at Moberg Analytics, where he’s been collaborating on assessing the performance of ICP forecasting models, focusing on the use of a pre-trained transformer model for time series data.
In this interview, Florian sits down with Dick to discuss what he believes should be the future of neuroscience with specific regards to brain injury and neurocritical care. The two speak on how prediction models can assist clinicians in making difficult decisions regarding individual TBI patients, and how we can make models output more reliable, precise predictions from complex multimodal data sources.