GBMbench

Glioblastoma Model Benchmark (working title) is a work in progress with the goal of assessing the possible benefit of Glioblastoma models for better radiotherapy planning.

Features

• Standardized processing pipeline for Brain MRIs
• Extensible benchmark framework for dockered Glioblastoma models
• Easy-to-use minimal API

Setting up

Prerequisites

• Docker: Required for running BRATS and custom Glioblastoma models. See Docker and NVIDIA Container Toolkit Setup.
• dicom2niix: Required if you plan to process raw DICOM data.

Installation

git clone https://github.com/LMZimmer/gbm_bench.git
cd gbm_bench
pip install .

Docker and NVIDIA Container Toolkit Setup

• Docker: Installation instructions on the official website
• NVIDIA Container Toolkit: Refer to the NVIDIA install guide and the official GitHub page

Adding new growth models

This repository can be used to perform inference or benchmark with your own tumor growth model. To this end, you need to create a docker image of your growth model. The following sections serve as guideline on how the image should be created.

Directory structure

Input and output data are passed to/from the container using mounted directories:

Input:

/mlcube_io0
   ┗ Patient-00000
      ┣ 00000-t1c.nii.gz
      ┣ 00000-gm.nii.gz
      ┣ 00000-wm.nii.gz
      ┣ 00000-csf.nii.gz
      ┣ 00000-tumorseg.nii.gz
      ┗ 00000-pet.nii.gz

Output:

/mlcube_io1
   ┗ 00000.nii.gz

Dockerfile Example

Ensure the container adheres to the above I/O structure. An example Dockerfile could be:

# Image and environment variables
FROM nvidia/cuda:11.8.0-cudnn8-runtime-ubuntu22.04

ENV DEBIAN_FRONTEND=noninteractive
ENV NVIDIA_VISIBLE_DEVICES=all
ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility

# Install python
RUN apt-get update && apt-get install -y --no-install-recommends \
    python3 python3-pip python3-dev git && \
    apt-get clean && rm -rf /var/lib/apt/lists/*
RUN python3 -m pip install --no-cache-dir --upgrade pip

WORKDIR /app

# Install requirements
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

# Copy your code to workdir
COPY . .
ENTRYPOINT ["python3", "inference.py"]

Adding new datasets

Datasets are handled by the LongitudinalDataset class in utils.parsing. This class supports:

• Parsing from a specific directory structure
• Reading/writing dataset paths via JSON

The recommended approach is to provide a compatible .json file. Example: data/datasets/rhuh.json. Automatic parsing is also supported, though naming consistency (e.g., for preop, postop, followup) can be a challenge.