Andreas Max Weng1, Christian Kestler1, Andreas Steven Kunz1, Simon Veldhoen1, Thorsten Alexander Bley1, Herbert Köstler1, and Tobias Wech1
1Department of Diagnostic and Interventional Radiology, University Hospital Würzburg, Würzburg, Germany
Synopsis
Functional lung MRI
still suffers from a time consuming post-processing with manual image
segmentation being its most time consuming part. We introduce and evaluate a
deep learning based semantic image segmentation technique to enable fully
automated post-processing in SENCEFUL-MRI.
Obtained
segmentations were compared to manual segmentations using the DICE similarity
coefficient (DSC). Furthermore, quantitative ventilation values were obtained
after manual and automatic segmentation.
Mean
DSC of the binary segmentation masks was 0.83 ± 0.09 and no significant difference in quantitative ventilation values was
observed. Obtained results show that the time consuming manual post-processing in functional lung MRI can be automated by the proposed neural network.
Introduction
MRI-based
quantitative assessment of lung ventilation has been investigated recently (1–3). While techniques like Fourier
decomposition (4) or SElf-gated Non-Contrast Enhanced
FUnctional Lung imaging (SENCEFUL) (5) are promising for comprehensive
lung investigations, post-processing is still cumbersome, with manual image
segmentation being its most time consuming part.
Therefore, we now introduce
and evaluate a deep learning based semantic image segmentation technique to
enable fully automated post-processing in SENCEFUL-MRI.Methods
Datasets from 12
healthy volunteers were acquired using the previously proposed SENCEFUL
approach (5) covering the whole lung via 10 ± 2 2D slices without gaps. Data was
acquired on a 3T system (Magnetom PRISMA, Siemens Healthcare, Erlangen,
Germany) applying the following parameters: TR: 2.5ms; TE: 0.7ms; flip angle:
8°; FOV: 450x450mm2; matrix size: 128x128; slice thickness: 10 mm. Until
now, after registration of the images one image of the motion corrected dataset
had to be segmented manually to evaluate regional quantitative ventilation (QV)(1). In the present study, this step was
additionally performed by a fully convolutional artificial neural network that
has been trained with 1054 single slices and according manual labels for lung
segmentation from our institutional database. The network has a similar
architecture as the VGG-16 network (6), however, the second part of the
implemented network is represented by a decoder pattern and a final pixel
classification-layer, equivalent to SegNet (7).
Manual
and deep learning based segmentation results were compared using the Dice similarity
coefficient (DSC, (8)). Additionally,
quantitative ventilation was evaluated for both manual and automatic
segmentations, and compared to each other via Spearman’s rho analysis.Results
The mean DSC of the binary segmentation masks was 0.83 ± 0.09. Figure 1 presents a direct comparison of the different segmentation
techniques of a representative dataset. Mean quantitative ventilation over all
2D slices was 0.10 ± 0.04 ml gas/ml lung tissue for both the manual and the
deep learning approach and a significant correlation was found (rho = 0.94, p
< 0.01). Figure 2 shows a scatterplot of the QV values averaged across
single slices for both approaches.
Discussion
Our study shows that the
time consuming manual post-processing in SENCEFUL-MRI can be adequately automated
by the proposed neural network. This lowers the workload of the investigator
and drastically reduces the processing time without impairing quantitative
ventilation results.
The main
metric used for evaluation of the automatic segmentation was the Dice
similarity coefficient, which provided good results. Nevertheless, the pool of
training data was rather small and contained labels from different operators,
with varying manual segmentation style. More training data or transfer learning
based on a network already trained for a similar task (e.g. semantic
segmentation in cardiac MRI) would certainly further improve the overall
robustness.Conclusion
Post-processing in
SENCEFUL can be fully automated by introducing a convolutional neural network for
semantic lung segmentation. Segmentation accuracy and the quality of further
evaluations were not significantly impaired compared to time-consuming
manual processing.Acknowledgements
No acknowledgement found.References
1.
Veldhoen S et al.,Radiology 2017;283:242–251.
2. Voskrebenzev A et al.,
Magn. Reson. Med.
2015;76:1542–1550.
3. Capaldi DPI et al., Radiology 2018;287:693–704.
4. Bauman G et al., Magn. Reson. Med. 2013;69:229–237.
5. Fischer A et al., NMR Biomed.
2014;27:907–17.
6. Simonyan K et al., CoRR
2014;1409.1556.
7. Badrinarayanan V et al.,
IEEE Trans. Pattern Anal. Mach. Intell.
2017;39:2481–2495.
8. Dice LR, Ecology 1945;26:297–302.