Artificial intelligence enhances the accuracy of portal and hepatic vein extraction in computed tomography for virtual hepatectomy.
artificial intelligence
automatic extraction
deep learning
liver vessels
organ vessels
Journal
Journal of hepato-biliary-pancreatic sciences
ISSN: 1868-6982
Titre abrégé: J Hepatobiliary Pancreat Sci
Pays: Japan
ID NLM: 101528587
Informations de publication
Date de publication:
Mar 2022
Mar 2022
Historique:
revised:
05
10
2021
received:
22
07
2021
accepted:
24
10
2021
pubmed:
16
11
2021
medline:
29
3
2022
entrez:
15
11
2021
Statut:
ppublish
Résumé
Current conventional algorithms used for 3-dimensional simulation in virtual hepatectomy still have difficulties distinguishing the portal vein (PV) and hepatic vein (HV). The accuracy of these algorithms was compared with a new deep-learning based algorithm (DLA) using artificial intelligence. A total of 110 living liver donor candidates until 2017, and 46 donor candidates until 2019 were allocated to the training group and validation groups for the DLA, respectively. All PV or HV branches were labeled based on Couinaud's segment classification and the Brisbane 2000 Terminology by hepato-biliary surgeons. Misclassified and missing branches were compared between a conventional tracking-based algorithm (TA) and DLA in the validation group. The sensitivity, specificity, and Dice coefficient for the PV were 0.58, 0.98, and 0.69 using the TA; and 0.84, 0.97, and 0.90 using the DLA (P < .001, excluding specificity); and for the HV, 0.81, 087, and 0.83 using the TA; and 0.93, 0.94 and 0.94 using the DLA (P < .001 to P = .001). The DLA exhibited greater accuracy than the TA. Compared with the TA, artificial intelligence enhanced the accuracy of extraction of the PV and HVs in computed tomography.
Sections du résumé
BACKGROUND/PURPOSE
OBJECTIVE
Current conventional algorithms used for 3-dimensional simulation in virtual hepatectomy still have difficulties distinguishing the portal vein (PV) and hepatic vein (HV). The accuracy of these algorithms was compared with a new deep-learning based algorithm (DLA) using artificial intelligence.
METHODS
METHODS
A total of 110 living liver donor candidates until 2017, and 46 donor candidates until 2019 were allocated to the training group and validation groups for the DLA, respectively. All PV or HV branches were labeled based on Couinaud's segment classification and the Brisbane 2000 Terminology by hepato-biliary surgeons. Misclassified and missing branches were compared between a conventional tracking-based algorithm (TA) and DLA in the validation group.
RESULTS
RESULTS
The sensitivity, specificity, and Dice coefficient for the PV were 0.58, 0.98, and 0.69 using the TA; and 0.84, 0.97, and 0.90 using the DLA (P < .001, excluding specificity); and for the HV, 0.81, 087, and 0.83 using the TA; and 0.93, 0.94 and 0.94 using the DLA (P < .001 to P = .001). The DLA exhibited greater accuracy than the TA.
CONCLUSION
CONCLUSIONS
Compared with the TA, artificial intelligence enhanced the accuracy of extraction of the PV and HVs in computed tomography.
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
359-368Subventions
Organisme : Ministry of Education, Culture, Sports, Science, and Technology of Japan
ID : 19K09191
Informations de copyright
© 2021 Japanese Society of Hepato-Biliary-Pancreatic Surgery.
Références
Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68(6):394-424.
Cherqui D, Laurent A, Mocellin N, Tayar C, Luciani A, Van Nhieu JT, et al. Liver resection for transplantable hepatocellular carcinoma: long-term survival and role of secondary liver transplantation. Ann Surg. 2009;250(5):738-46.
Mise Y, Hasegawa K, Satou S, Shindoh J, Miki K, Akamatsu N, et al. How has virtual hepatectomy changed the practice of liver surgery?: Experience of 1194 virtual hepatectomy before liver resection and living donor liver transplantation. Ann Surg. 2018;268(1):127-33.
Hashimoto D, Dohi T, Tsuzuki M, Horiuchi T, Ohta Y, Chinzei K, et al. Development of a computer-aided surgery system: three-dimensional graphic reconstruction for treatment of liver cancer. Surgery. 1991;109(5):589-96.
Wigmore SJ, Redhead DN, Yan XJ, Casey J, Madhavan K, Dejong CH, et al. Virtual hepatic resection using three-dimensional reconstruction of helical computed tomography angioportograms. Ann Surg. 2001;233(2):221-6.
Shindoh J, Seyama Y, Umekita N. Three-dimensional staining of liver segments with an ultrasound contrast agent as an aid to anatomic liver resection. J Am Coll Surg. 2012;215(2):e5-10.
Takamoto T, Hashimoto T, Ogata S, Inoue K, Maruyama Y, Miyazaki A, et al. Planning of anatomical liver segmentectomy and subsegmentectomy with 3-dimensional simulation software. Am J Surg. 2013;206(4):530-8.
Oshiro Y, Yano H, Mitani J, Kim S, Kim J, Fukunaga K, et al. Novel 3-dimensional virtual hepatectomy simulation combined with real-time deformation. World J Gastroenterol. 2015;21(34):9982-92.
Lodewick TM, Arnoldussen CW, Lahaye MJ, van Mierlo KM, Neumann UP, Beets-Tan RG, et al. Fast and accurate liver volumetry prior to hepatectomy. HPB (Oxford). 2016;18(9):764-72.
Bekheit M, Bucur PO, Wartenberg M, Vibert E. Computerized tomography-based anatomic description of the porcine liver. J Surg Res. 2017;210:223-30.
Furusato Hunt OM, Lubner MG, Ziemlewicz TJ, Muñoz Del Rio A, Pickhardt PJ. The liver segmental volume ratio for noninvasive detection of cirrhosis: comparison with established linear and volumetric measures. J Comput Assist Tomogr. 2016;40(3):478-84.
Mise Y, Tani K, Aoki T, Sakamoto Y, Hasegawa K, Sugawara Y, et al. Virtual liver resection: computer-assisted operation planning using a three-dimensional liver representation. J Hepatobiliary Pancreat Sci. 2013;20(2):157-64.
Ohshima S. Volume analyzer SYNAPSE VINCENT for liver analysis. J Hepatobiliary Pancreat Sci. 2014;21(4):235-8.
Watanabe J, Endo K, Tanaka Y, Goto K, Urushibara S, Osaki T, et al. Investigation of the utility and safety of dynamic computed tomography with vasodilators. Yonago Acta Med. 2020;63(1):47-54.
Pang YY. The Brisbane 2000 terminology of liver anatomy and resections. HPB. 2000;2:333-39. HPB (Oxford). 2002;4(2):99; author reply -100.
Couinaud C. Liver anatomy: portal (and suprahepatic) or biliary segmentation. Dig Surg. 1999;16(6):459-67.
Ihara S, Kitamura Y, Li Y, Ito W. Liver Vessel Extraction and Separating Using Graph Analysis. IEICE Technical Report; 2011.
Salem NM, Salem SA, Nandi AK, editors. Segmentation of retinal blood vessels based on analysis of the hessian matrix and Clustering Algorithm. 2007 15th European Signal Processing Conference; 2007 3-7 Sept. 2007.
Keshwani D, Kitamura Y, Ihara S, Iizuka S, Simo-Serra E, editors. TopNet: Topology Preserving Metric Learning for Vessel Tree Reconstruction and Labelling. International Conference on Medical Image Computing and Computer-Assisted Intervention; 2020: Springer.
Dice LR. Measures of the amount of ecologic between species. Ecology. 1945;26(3):297-302.
Klyuzhin I, Chausse G, Bloise I, Lavista Ferres J, Uribe C, Rahmim A. Automated deep segmentation of healthy organs in PSMA PET/CT images. J Nucl Med. 2021;62(Suppl 1):1410.
Takayasu K, Moriyama N, Muramatsu Y, Shima Y, Goto H, Yamada T. Intrahepatic portal vein branches studied by percutaneous transhepatic portography. Radiology. 1985;154(1):31-6.
Huang Q, Sun J, Ding H, Wang X, Wang G. Robust liver vessel extraction using 3D U-Net with variant dice loss function. Comput Biol Med. 2018;101:153-62.
Moccia S, De Momi E, El Hadji S, Mattos LS. Blood vessel segmentation algorithms - Review of methods, datasets and evaluation metrics. Comput Methods Programs Biomed. 2018;158:71-91.
Castiglioni I, Rundo L, Codari M, Di Leo G, Salvatore C, Interlenghi M, et al. AI applications to medical images: from machine learning to deep learning. Phys Med. 2021;83:9-24.
Ibragimov B, Toesca D, Chang D, Koong A, Xing L. Combining deep learning with anatomical analysis for segmentation of the portal vein for liver SBRT planning. Phys Med Biol. 2017;62(23):8943-58.
Kitrungrotsakul T, Han XH, Iwamoto Y, Lin L, Foruzan AH, Xiong W, et al. VesselNet: a deep convolutional neural network with multi pathways for robust hepatic vessel segmentation. Comput Med Imaging Graph. 2019;75:74-83.
Kitrungrotsakul T, Han X-H, Iwamoto Y, Foruzan AH, Lin L , Chen Y-W. Robust hepatic vessel segmentation using multi deep convolution network. Medical Imaging 2017: Biomedical Applications in Molecular, Structural, and Functional Imaging. 2017.