This repository includes the source codes to evaluate the survival model described in our paper "Novel ML-based Prognostic Subgrouping of Glioblastoma: A Multi-center Study"
-
src/FeatEx.m : function to extract features
-
src/GBMSurvival_predict.m : matlab fuction to run predictions
-
src/run_GBMsurvival_predict.sh: Bash wrapper to extract features from images, apply the pretrained model, and output a report. Inputs are patient age, preprocessed MRI images (t1,t1gd,t2,t2-flair), and tumor segmentation mask.
-
src/data: model and atlases
- Atlas_Sur_NatMed.nii.gz: The 4th 3D channel includes the Overall Survival Map (OSM) atlas described in the manuscript.
- jakob_stripped_with_cere_lps_256256128.nii.gz: common atlas where all images are deformed to
- templateallregions.nii.gz: segmentation labels of the common atlas
-
src/libs: matlab libraries
-
MATLAB version 9.4 (R2018a)
-
greedy: https://github.com/pyushkevich/greedy [1]
-
python3: For pdf report creation. Dependencies in src/python_dependencies
In addition to making the source codes available here, we have created an online platform where users can preprocess and apply our model on their own images to obtain survival prediction without the need for installation.
-
Sample input data from the public dataset UPENN-GBM (https://doi.org/10.7937/TCIA.709X-DN49) and output report are available in the folder sample_data
-
Step-by-step instructions can be found here.
Codes for the image preprocessing are available separately through the following GitHub repositories
-
Preprocessing pipeline (dicom to nifti conversion, image co-registration to the SRI atlas, and ptional brain extraction and tumor segmentation): https://github.com/CBICA/CaPTk [2]
-
Brain extraction: https://github.com/CBICA/BrainMaGe [3]
-
Tumor segmentation: https://github.com/FETS-AI/Front-End [4].
[1] Yushkevich, P.A., Pluta, J., Wang, H., Wisse, L.E., Das, S. and Wolk, D., 2016. Fast Automatic Segmentation of Hippocampal Subfields and Medial Temporal Lobe Subregions in 3 Tesla and 7 Tesla MRI. Alzheimer's & Dementia: The Journal of the Alzheimer's Association, 12(7), pp.P126-P127.
[2] Davatzikos et al. Cancer imaging phenomics toolkit: quantitative imaging analytics for precision diagnostics and predictive modeling of clinical outcome, J Med Imaging, 5(1):011018, 2018
[3] Thakur, S., et al. Brain extraction on MRI scans in presence of diffuse glioma: Multi-institutional performance evaluation of deep learning methods and robust modality-agnostic training. NeuroImage 220, 117081 (2020).
[4] Pati, S., et al. The federated tumor segmentation (FeTS) tool: an open-source solution to further solid tumor research. Physics in Medicine & Biology (2022).
