Dataset: https://www.med.upenn.edu/cbica/brats2020/data.html

369 brains in total, 300 for training, 35 for validation, 34 for testing

2D image slices training: convert pixel-level brain tumor segmentation to image-level brain tumor binary classification

Only use T1 brain slices

rescale image slice pixel values to range [0,1]

if sum(all pixel values in one mask slice) > 0:
  label = 1 # has tumor
else:
  label = 0 # no tumor

Testing performance (tumor is positive class, tumor:normal ~= 1:1):

TP: 2620.0

TN: 2281.0

FP: 660.0

FN: 327.0

precision: 0.7987804634518151

recall: 0.8890396712236284

f_measure: 0.8414468267997854

accuracy: 0.8323709097763092

auc_roc: 0.9338790120417066

Data Usage Agreement citations:

[1] B. H. Menze, A. Jakab, S. Bauer, J. Kalpathy-Cramer, K. Farahani, J. Kirby, et al. "The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS)", IEEE Transactions on Medical Imaging 34(10), 1993-2024 (2015) DOI: 10.1109/TMI.2014.2377694

[2] S. Bakas, H. Akbari, A. Sotiras, M. Bilello, M. Rozycki, J.S. Kirby, et al., "Advancing The Cancer Genome Atlas glioma MRI collections with expert segmentation labels and radiomic features", Nature Scientific Data, 4:170117 (2017) DOI: 10.1038/sdata.2017.117

[3] S. Bakas, M. Reyes, A. Jakab, S. Bauer, M. Rempfler, A. Crimi, et al., "Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge", arXiv preprint arXiv:1811.02629 (2018)

[4] S. Bakas, H. Akbari, A. Sotiras, M. Bilello, M. Rozycki, J. Kirby, et al., "Segmentation Labels and Radiomic Features for the Pre-operative Scans of the TCGA-GBM collection", The Cancer Imaging Archive, 2017. DOI: 10.7937/K9/TCIA.2017.KLXWJJ1Q

[5] S. Bakas, H. Akbari, A. Sotiras, M. Bilello, M. Rozycki, J. Kirby, et al., "Segmentation Labels and Radiomic Features for the Pre-operative Scans of the TCGA-LGG collection", The Cancer Imaging Archive, 2017. DOI: 10.7937/K9/TCIA.2017.GJQ7R0EF