TY  - JOUR
AU  - Buchner, Josef A
AU  - Peeken, Jan C
AU  - Etzel, Lucas
AU  - Ezhov, Ivan
AU  - Mayinger, Michael
AU  - Christ, Sebastian M
AU  - Brunner, Thomas B
AU  - Wittig, Andrea
AU  - Menze, Björn
AU  - Zimmer, Claus
AU  - Meyer, Bernhard
AU  - Guckenberger, Matthias
AU  - Andratschke, Nicolaus
AU  - El Shafie, Rami A
AU  - Debus, Jürgen
AU  - Rogers, Susanne
AU  - Riesterer, Oliver
AU  - Schulze, Katrin
AU  - Feldmann, Horst J
AU  - Blanck, Oliver
AU  - Zamboglou, Constantinos
AU  - Ferentinos, Konstantinos
AU  - Bilger, Angelika
AU  - Grosu, Anca L
AU  - Wolff, Robert
AU  - Kirschke, Jan S
AU  - Eitz, Kerstin A
AU  - Combs, Stephanie E
AU  - Bernhardt, Denise
AU  - Rueckert, Daniel
AU  - Piraud, Marie
AU  - Wiestler, Benedikt
AU  - Kofler, Florian
TI  - Identifying core MRI sequences for reliable automatic brain metastasis segmentation.
JO  - Radiotherapy and oncology
VL  - 188
SN  - 0167-8140
CY  - Amsterdam [u.a.]
PB  - Elsevier Science
M1  - DKFZ-2023-01828
SP  - 109901
PY  - 2023
N1  - Volume 188, November 2023, 109901
AB  - Many automatic approaches to brain tumor segmentation employ multiple magnetic resonance imaging (MRI) sequences. The goal of this project was to compare different combinations of input sequences to determine which MRI sequences are needed for effective automated brain metastasis (BM) segmentation.We analyzed preoperative imaging (T1-weighted sequence ± contrast-enhancement (T1/T1-CE), T2-weighted sequence (T2), and T2 fluid-attenuated inversion recovery (T2-FLAIR) sequence) from 339 patients with BMs from seven centers. A baseline 3D U-Net with all four sequences and six U-Nets with plausible sequence combinations (T1-CE, T1, T2-FLAIR, T1-CE+T2-FLAIR, T1-CE+T1+T2-FLAIR, T1-CE+T1) were trained on 239 patients from two centers and subsequently tested on an external cohort of 100 patients from five centers.The model based on T1-CE alone achieved the best segmentation performance for BM segmentation with a median Dice similarity coefficient (DSC) of 0.96. Models trained without T1-CE performed worse (T1-only: DSC = 0.70 and T2-FLAIR-only: DSC = 0.73). For edema segmentation, models that included both T1-CE and T2-FLAIR performed best (DSC = 0.93), while the remaining four models without simultaneous inclusion of these both sequences reached a median DSC of 0.81-0.89.A T1-CE-only protocol suffices for the segmentation of BMs. The combination of T1-CE and T2-FLAIR is important for edema segmentation. Missing either T1-CE or T2-FLAIR decreases performance. These findings may improve imaging routines by omitting unnecessary sequences, thus allowing for faster procedures in daily clinical practice while enabling optimal neural network-based target definitions.
KW  - CNN (Other)
KW  - FLAIR (Other)
KW  - MRI sequences (Other)
KW  - T1 (Other)
KW  - U-net (Other)
KW  - brain metastases (Other)
KW  - deep learning (Other)
KW  - segmentation (Other)
LB  - PUB:(DE-HGF)16
C6  - pmid:37678623
DO  - DOI:10.1016/j.radonc.2023.109901
UR  - https://inrepo02.dkfz.de/record/282682
ER  -