Liver segmentation using Turbolift learning for CT and cone-beam C-arm perfusion imaging

Model-based reconstruction employing the time separation technique (TST) was found to improve dynamic perfusion imaging of the liver using C-arm cone-beam computed tomography (CBCT). To apply TST using prior knowledge extracted from CT perfusion data, the liver should be accurately segmented from th...

Full description

Saved in:
Bibliographic Details
Published in:Computers in biology and medicine 2023-03, Vol.154, p.106539-106539, Article 106539
Main Authors: Haseljić, Hana, Chatterjee, Soumick, Frysch, Robert, Kulvait, Vojtěch, Semshchikov, Vladimir, Hensen, Bennet, Wacker, Frank, Brüsch, Inga, Werncke, Thomas, Speck, Oliver, Nürnberger, Andreas, Rose, Georg
Format: Article
Language:English
Subjects:
CBCT
Computed tomography
Cone-Beam Computed Tomography - methods
Dynamic perfusion imaging
Image Processing, Computer-Assisted - methods
Image segmentation
Learning
Liver
Liver - diagnostic imaging
Liver diseases
Liver segmentation
Medical imaging
Perfusion
Statistical analysis
Tomography, X-Ray Computed
Training
TST
Visualization
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Model-based reconstruction employing the time separation technique (TST) was found to improve dynamic perfusion imaging of the liver using C-arm cone-beam computed tomography (CBCT). To apply TST using prior knowledge extracted from CT perfusion data, the liver should be accurately segmented from the CT scans. Reconstructions of primary and model-based CBCT data need to be segmented for proper visualisation and interpretation of perfusion maps. This research proposes Turbolift learning, which trains a modified version of the multi-scale Attention UNet on different liver segmentation tasks serially, following the order of the trainings CT, CBCT, CBCT TST – making the previous trainings act as pre-training stages for the subsequent ones – addressing the problem of limited number of datasets for training. For the final task of liver segmentation from CBCT TST, the proposed method achieved an overall Dice scores of 0.874±0.031 and 0.905±0.007 in 6-fold and 4-fold cross-validation experiments, respectively — securing statistically significant improvements over the model, which was trained only for that task. Experiments revealed that Turbolift not only improves the overall performance of the model but also makes it robust against artefacts originating from the embolisation materials and truncation artefacts. Additionally, in-depth analyses confirmed the order of the segmentation tasks. This paper shows the potential of segmenting the liver from CT, CBCT, and CBCT TST, learning from the available limited training data, which can possibly be used in the future for the visualisation and evaluation of the perfusion maps for the treatment evaluation of liver diseases. •This research proposes Turbolift learning which trains a modified version of the multi-scale Attention UNet on different liver segmentation tasks serially.•Making the previous trainings act as pre-training stages for the subsequent ones — CT, CBCT, CBCT TST.•Addressing the problem of limited number of datasets for training using the stage-wise training strategy.•For the final task of liver segmentation from CBCT TST, the proposed method achieved an overall Dice scores of 0.874±0.031 and 0.905±0.007 in 6-fold and 4-fold cross-validation experiments.•Turbolift not only improves the overall performance of the model but also makes it robust against artefacts originating from the embolisation materials and truncation artefacts.
ISSN:0010-4825
1879-0534
DOI:10.1016/j.compbiomed.2023.106539