A Lightweight Optimization Framework for Estimating 3D Brain Tumor Infiltration

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Weidner, Jonas, Balcerak, Michal, Ezhov, Ivan, Datchev, André, Lux, Laurin, Zimmer, Lucas, Rueckert, Daniel, Menze, Björn, Wiestler, Benedikt
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916873041346560
author Weidner, Jonas
Balcerak, Michal
Ezhov, Ivan
Datchev, André
Lux, Laurin
Zimmer, Lucas
Rueckert, Daniel
Menze, Björn
Wiestler, Benedikt
author_facet Weidner, Jonas
Balcerak, Michal
Ezhov, Ivan
Datchev, André
Lux, Laurin
Zimmer, Lucas
Rueckert, Daniel
Menze, Björn
Wiestler, Benedikt
contents Glioblastoma, the most aggressive primary brain tumor, poses a severe clinical challenge due to its diffuse microscopic infiltration, which remains largely undetected on standard MRI. As a result, current radiotherapy planning employs a uniform 15 mm margin around the resection cavity, failing to capture patient-specific tumor spread. Tumor growth modeling offers a promising approach to reveal this hidden infiltration. However, methods based on partial differential equations or physics-informed neural networks tend to be computationally intensive or overly constrained, limiting their clinical adaptability to individual patients. In this work, we propose a lightweight, rapid, and robust optimization framework that estimates the 3D tumor concentration by fitting it to MRI tumor segmentations while enforcing a smooth concentration landscape. This approach achieves superior tumor recurrence prediction on 192 brain tumor patients across two public datasets, outperforming state-of-the-art baselines while reducing runtime from 30 minutes to less than one minute. Furthermore, we demonstrate the framework's versatility and adaptability by showing its ability to seamlessly integrate additional imaging modalities or physical constraints.
format Preprint
id arxiv_https___arxiv_org_abs_2412_13811
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Lightweight Optimization Framework for Estimating 3D Brain Tumor Infiltration
Weidner, Jonas
Balcerak, Michal
Ezhov, Ivan
Datchev, André
Lux, Laurin
Zimmer, Lucas
Rueckert, Daniel
Menze, Björn
Wiestler, Benedikt
Medical Physics
Computer Vision and Pattern Recognition
Glioblastoma, the most aggressive primary brain tumor, poses a severe clinical challenge due to its diffuse microscopic infiltration, which remains largely undetected on standard MRI. As a result, current radiotherapy planning employs a uniform 15 mm margin around the resection cavity, failing to capture patient-specific tumor spread. Tumor growth modeling offers a promising approach to reveal this hidden infiltration. However, methods based on partial differential equations or physics-informed neural networks tend to be computationally intensive or overly constrained, limiting their clinical adaptability to individual patients. In this work, we propose a lightweight, rapid, and robust optimization framework that estimates the 3D tumor concentration by fitting it to MRI tumor segmentations while enforcing a smooth concentration landscape. This approach achieves superior tumor recurrence prediction on 192 brain tumor patients across two public datasets, outperforming state-of-the-art baselines while reducing runtime from 30 minutes to less than one minute. Furthermore, we demonstrate the framework's versatility and adaptability by showing its ability to seamlessly integrate additional imaging modalities or physical constraints.
title A Lightweight Optimization Framework for Estimating 3D Brain Tumor Infiltration
topic Medical Physics
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.13811