Abdominal artery segmentation method from CT volumes using fully convolutional neural network
- 57 Downloads
The purpose of this paper is to present a fully automated abdominal artery segmentation method from a CT volume. Three-dimensional (3D) blood vessel structure information is important for diagnosis and treatment. Information about blood vessels (including arteries) can be used in patient-specific surgical planning and intra-operative navigation. Since blood vessels have large inter-patient variations in branching patterns and positions, a patient-specific blood vessel segmentation method is necessary. Even though deep learning-based segmentation methods provide good segmentation accuracy among large organs, small organs such as blood vessels are not well segmented. We propose a deep learning-based abdominal artery segmentation method from a CT volume. Because the artery is one of small organs that is difficult to segment, we introduced an original training sample generation method and a three-plane segmentation approach to improve segmentation accuracy.
Our proposed method segments abdominal arteries from an abdominal CT volume with a fully convolutional network (FCN). To segment small arteries, we employ a 2D patch-based segmentation method and an area imbalance reduced training patch generation (AIRTPG) method. AIRTPG adjusts patch number imbalances between patches with artery regions and patches without them. These methods improved the segmentation accuracies of small artery regions. Furthermore, we introduced a three-plane segmentation approach to obtain clear 3D segmentation results from 2D patch-based processes. In the three-plane approach, we performed three segmentation processes using patches generated on axial, coronal, and sagittal planes and combined the results to generate a 3D segmentation result.
The evaluation results of the proposed method using 20 cases of abdominal CT volumes show that the averaged F-measure, precision, and recall rates were 87.1%, 85.8%, and 88.4%, respectively. This result outperformed our previous automated FCN-based segmentation method. Our method offers competitive performance compared to the previous blood vessel segmentation methods from 3D volumes.
We developed an abdominal artery segmentation method using FCN. The 2D patch-based and AIRTPG methods effectively segmented the artery regions. In addition, the three-plane approach generated good 3D segmentation results.
KeywordsAbdominal artery CT image Segmentation Fully convolutional network
Parts of this research were supported by MEXT, JSPS KAKENHI Grant Numbers 26108006, 17H00867, JSPS Bilateral International Collaboration Grants, and JST ACT-I (JPMJPR16U9).
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
For this type of study formal consent is not required.
- 1.Maklad AS, Matsuhiro M, Suzuki H, Kawata Y, Niki N, Shimada M, Iinuma G (2018) Automatic blood vessel based-liver segmentation using the portal phase abdominal CT. Proc SPIE Med Imaging 1057527Google Scholar
- 2.Okada T, Linguraru MG, Hori M, Summers RM, Tomiyama N, Sato Y (2013) Abdominal multi-organ CT segmentation using organ correlation graph and prediction-based shape and location priors. Med Image Comput Comput Assist Interv (MICCAI) 8151:275–282Google Scholar
- 4.Amir-Khalili A, Peyrat JM, Abinahed J, Al-Alao O, Al-Ansari A, Hamarneh G, Abugharbieh R (2014) Auto localization and segmentation of occluded vessels in robot-assisted partial nephrectomy. Med Image Comput Comput Assist Interv (MICCAI) 8673:407–414Google Scholar
- 6.Ieiri S, Uemura M, Konishi K, Souzaki R, Nagao Y, Tsutsumi N, Akahoshi T, Ohuchida K, Ohdaira T, Tomikawa M, Tanoue K, Hashizumie M, Taguchi T (2012) Augmented reality navigation system for laparoscopic splenectomy in children based on preoperative CT image using optical tracking device. Pediatr Surg Int 28(4):341–346CrossRefGoogle Scholar
- 7.Lee SW, Shinohara H, Matsuki M, Okuda J, Nomura E, Mabuchi H, Nishiguchi K, Takaori K, Narabayashi I, Tanigawa N (2003) Preoperative simulation of vascular anatomy by three-dimensional computed tomography imaging in laparoscopic gastric cancer surgery. J Am Coll Surg 197(6):927–936CrossRefGoogle Scholar
- 8.Roth HR, Oda M, Shimizu N, Oda H, Hayashi Y, Kitasaka T, Fujiwara M, Misawa K, Mori K (2018) Towards dense volumetric pancreas segmentation in CT using 3D fully convolutional network. Proc SPIE Med Imaging 10574:105740B-1–105740B-6Google Scholar
- 11.Fu H, Xu Y, Lin S, Wong DWK, Liu J (2016) Deep vessel: retinal vessel segmentation via deep learning and conditional random field. Med Image Comput Comput Assist Interv (MICCAI) 9901:132–139Google Scholar
- 12.Fu A, Xu Z, Gao M, Buty M, Mollura DJ (2016) Deep vessel tracking: A generalized probabilistic approach via deep learning. In: IEEE 13th international symposium on biomedical imaging (ISBI)Google Scholar
- 14.Prentas̆ic P, Heisler M, Mammo Z, Lee S, Merkur A, Navajas E, Beg MF, S̆arunic M, Loncaric S, (2016) Segmentation of the foveal microvasculature using deep learning networks. Med Image Comput Comput Assist Interv (MICCAI) 9901:132–139Google Scholar
- 15.Dasgupta A, Singh S (2017) A fully convolutional neural network based structured prediction approach towards the retinal vessel segmentation. IEEE 14th international symposium on biomedical imaging (ISBI)Google Scholar
- 16.Zhang Y, Chung ACS (2018) Deep supervision with additional labels for retinal vessel segmentation task. Med Image Comput Comput Assist Interv (MICCAI) 11071:83–91Google Scholar
- 17.Wu Y, Xia Y, Song Y, Zhang Y, Cai W (2018) Multiscale network followed network model for retinal vessel segmentation. Med Image Comput Comput Assist Interv (MICCAI) 11071:119–126Google Scholar
- 22.Frangi AF, Niessen WJ, Vincken KL, Viergever MA (1998) Multiscale vessel enhancement filtering. Med Image Comput Comput Assist Interv (MICCAI) 1496:130–137Google Scholar
- 23.Oda M, Yamamoto T, Yoshino Y, Mori K (2016) Segmentation method of abdominal arteries from CT volumes utilizing intensity transition along arteries. Int J Comput Assist Radiol Surg 11(1):S46–S47Google Scholar
- 25.Chen L, Xie Y, Sun J, Balu N, Mossa-Basha M, Pimentel K, Hatsukami TS, Hwang J-N, Yuan C (2017) Y-net: 3D intracranial artery segmentation using a convolutional autoencoder. arXiv:1712.07194
- 26.Tetteh G, Efremov V, Forkert ND, Schneider M, Kirschke J, Weber B, Zimmer C, Piraud M, Menze BH (2018) DeepVesselNet: vessel segmentation, centerline prediction, and bifurcation detection in 3-D angiographic volumes. arXiv:1803.09340v2
- 28.Kitrungrotsakul T, Han X-H, Iwamoto Y, Foruzan AH, Lin L, Chen W-Y (2017) Robust hepatic vessel segmentation using multi deep convolution network. Proc SPIE Med Imaging 1013711Google Scholar
- 29.Kitrungrotsakul T, Han X-H, Wei X, Chen W-Y (2018) Multi-pathways CNN for robust vascular segmentation. Proc SPIE Med Imaging 105781SGoogle Scholar
- 30.Çiçek Ö, Abdulkadir A, Lienkamp SS, Brox T, Ronneberger O (2016) 3D U-net: learning dense volumetric segmentation from sparse annotation. MedMed Image Comput Comput Assist Interv (MICCAI) 9901:424–432Google Scholar
- 31.Ronneberger O, Fischer P, Brox T (2015) U-Net: convolutional networks for biomedical image segmentation. Med Image Comput Comput Assist Interv (MICCAI) 9351:234–241Google Scholar
- 32.Milletari F, Navab N, Ahmadi SA (2016) V-net: Fully convolutional neural networks for volumetric medical image segmentation. In: Fourth international conference on 3D Vision (3DV), pp 565–571Google Scholar
- 33.Sudre CH, Li W, Vercauteren T, Ourselin S, Cardoso MJ (2017) Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations. Int Workshop Deep Learn Med Image Anal (DLMIA) 10553:240–248Google Scholar
- 34.Oda M, Kitasaka T, Misawa K, Fujiwara M, Mori K (2018) Abdominal artery segmentation from CT volumes using fully convolutional network for small artery segmentation. Int J Comput Assist Radiol Surg 13(1):S20–21Google Scholar
- 35.Diederik PK, Jimmy B (2014) Adam: a method for stochastic optimization. arXiv:1412.6980
- 37.Jang Y, Hong Y, Ha S, Kim S, Chang HJ (2018) Automatic segmentation of LV and RV in cardiac MRI. Statistical Atlases and Computational Models of the Heart, ACDC and MMWHS Challenges, STACOM 10663:161–169Google Scholar
- 38.Patravali J, Jain S, Chilamkurthy S (2018) 2D–3D fully convolutional neural networks for cardiac MR segmentation. Statistical atlases and computational models of the heart, ACDC and MMWHS challenges, STACOM 2017 10663:130–139Google Scholar
- 39.Kurmann T (2017) Simultaneous recognition and pose estimation of instruments in minimally invasive surgery. Med Image Comput Comput Assist Interv (MICCAI) 10434:505–513Google Scholar