Evaluating glioma growth predictions as a forward ranking problem
This work addresses the challenge of accurately predicting tumor growth for medical applications, but it is incremental as it focuses on evaluation methods rather than new prediction models.
The paper tackles the problem of evaluating glioma growth predictions by framing it as a ranking problem rather than segmentation, using average precision as a metric to assess spatial infiltration patterns, and shows that better model fit does not always lead to better predictive performance.
The problem of tumor growth prediction is challenging, but promising results have been achieved with both model-driven and statistical methods. In this work, we present a framework for the evaluation of growth predictions that focuses on the spatial infiltration patterns, and specifically evaluating a prediction of future growth. We propose to frame the problem as a ranking problem rather than a segmentation problem. Using the average precision as a metric, we can evaluate the results with segmentations while using the full spatiotemporal prediction. Furthermore, by separating the model goodness-of-fit from future predictive performance, we show that in some cases, a better fit of model parameters does not guarantee a better the predictive power.