12-02, 15:10–15:40 (Europe/Amsterdam), Auditorium
Cognitive impairment is common amongst patients with primary brain tumors (PBT). The exact mechanism by which primary brain tumors affect different cognitive functions, however, is not well understood. Cognitive impairment in PBT patients is likely the result of local effects of the tumor, global effects of the tumor, and patient characteristics. Finding predictors, or the potentially complex interactions between them, may improve our understanding of how different variables influence cognitive function. Moreover, this may facilitate personalized prediction of cognitive function aid personalized treatment decisions. Several big challenges arise when aiming to make personalized predictions of cognitive functioning in PBT patients, many of these problems likely generalize to other applied machine learning tasks.
In many prediction tasks, we encounter similar challenges. Small sample sizes, weak and high dimensional predictors, and lots of noisy and difficult to interpret outcome measures to name a few. Problems we need to solve without hurting explainability. In this talk, I plan to address several problems that we encountered while predicting cognitive functioning in primary brain tumor patient that we believe to generalize well to many applied predictions tasks.
More specifically, I will dive into the following topics:
- Primary brain tumors, cognitive function, and treatment options
- Formulating our modeling problem and the challenges we often face
- Reduce the dimensionality of our noisy and high dimensional output variable. What are our options?
- Current challenges when segmenting brain tumors, the reasons deep learning models are still not good enough in practice.
- Evaluating a large set of models to predict cognitive function without introducing bias using Double Loop Cross Validation
- Using Multidimensional Scaling to obtain a low dimensional representation of tumor location. How representing data based on similarity may help creating more meaningfull embedding space
No previous knowledge expected
Data scientist and deep learning researcher and PhD student at the Elisabeth-Tweesteden hospital, Neurosurgery department and Tilbrurg University, department of cognitive sciences and AI.