TY - JOUR
T1 - Repurposing the Public BraTS Dataset for Postoperative Brain Tumour Treatment Response Monitoring
AU - Sørensen, Peter Jagd
AU - Ladefoged, Claes Nøhr
AU - Larsen, Vibeke Andrée
AU - Andersen, Flemming Littrup
AU - Nielsen, Michael Bachmann
AU - Poulsen, Hans Skovgaard
AU - Carlsen, Jonathan Frederik
AU - Hansen, Adam Espe
N1 - Publisher Copyright:
© 2024 by the authors.
PY - 2024
Y1 - 2024
N2 - The Brain Tumor Segmentation (BraTS) Challenge has been a main driver of the development of deep learning (DL) algorithms and provides by far the largest publicly available expert-annotated brain tumour dataset but contains solely preoperative examinations. The aim of our study was to facilitate the use of the BraTS dataset for training DL brain tumour segmentation algorithms for a postoperative setting. To this end, we introduced an automatic conversion of the three-label BraTS annotation protocol to a two-label annotation protocol suitable for postoperative brain tumour segmentation. To assess the viability of the label conversion, we trained a DL algorithm using both the three-label and the two-label annotation protocols. We assessed the models pre- and postoperatively and compared the performance with a state-of-the-art DL method. The DL algorithm trained using the BraTS three-label annotation misclassified parts of 10 out of 41 fluid-filled resection cavities in 72 postoperative glioblastoma MRIs, whereas the two-label model showed no such inaccuracies. The tumour segmentation performance of the two-label model both pre- and postoperatively was comparable to that of a state-of-the-art algorithm for tumour volumes larger than 1 cm3. Our study enables using the BraTS dataset as a basis for the training of DL algorithms for postoperative tumour segmentation.
AB - The Brain Tumor Segmentation (BraTS) Challenge has been a main driver of the development of deep learning (DL) algorithms and provides by far the largest publicly available expert-annotated brain tumour dataset but contains solely preoperative examinations. The aim of our study was to facilitate the use of the BraTS dataset for training DL brain tumour segmentation algorithms for a postoperative setting. To this end, we introduced an automatic conversion of the three-label BraTS annotation protocol to a two-label annotation protocol suitable for postoperative brain tumour segmentation. To assess the viability of the label conversion, we trained a DL algorithm using both the three-label and the two-label annotation protocols. We assessed the models pre- and postoperatively and compared the performance with a state-of-the-art DL method. The DL algorithm trained using the BraTS three-label annotation misclassified parts of 10 out of 41 fluid-filled resection cavities in 72 postoperative glioblastoma MRIs, whereas the two-label model showed no such inaccuracies. The tumour segmentation performance of the two-label model both pre- and postoperatively was comparable to that of a state-of-the-art algorithm for tumour volumes larger than 1 cm3. Our study enables using the BraTS dataset as a basis for the training of DL algorithms for postoperative tumour segmentation.
KW - annotation protocol
KW - automatic
KW - Brain Tumor Segmentation Challenge
KW - brain tumour segmentation
KW - BraTS
KW - deep learning algorithm
KW - magnetic resonance imaging
KW - MRI
KW - postoperative
KW - treatment monitoring
U2 - 10.3390/tomography10090105
DO - 10.3390/tomography10090105
M3 - Journal article
C2 - 39330751
AN - SCOPUS:85205074796
VL - 10
SP - 1397
EP - 1410
JO - Tomography - A Journal for Imaging Research
JF - Tomography - A Journal for Imaging Research
SN - 2379-1381
IS - 9
ER -