# Deep Probabilistic Modeling of Glioma Growth

## TL;DR

This paper introduces a data-driven probabilistic approach for modeling glioma growth, learning tumor dynamics directly from imaging data without relying on explicit biological models, enabling prediction of future tumor appearances.

## Contribution

It presents a novel probabilistic segmentation and representation learning method that implicitly captures tumor growth dynamics from data, bypassing traditional biologically inspired models.

## Key findings

- Successfully learns a distribution of plausible future tumor appearances.
- Demonstrates effectiveness in predicting tumor growth from imaging data.
- Provides a flexible framework for glioma growth modeling.

## Abstract

Existing approaches to modeling the dynamics of brain tumor growth, specifically glioma, employ biologically inspired models of cell diffusion, using image data to estimate the associated parameters. In this work, we propose an alternative approach based on recent advances in probabilistic segmentation and representation learning that implicitly learns growth dynamics directly from data without an underlying explicit model. We present evidence that our approach is able to learn a distribution of plausible future tumor appearances conditioned on past observations of the same tumor.