Recurrent neural networks for generalization towards the vessel geometry in autonomous endovascular guidewire navigation in the aortic arch.
Autonomous
Deep reinforcement learning
Endovascular intervention
Generalization
Recurrent neural network
Journal
International journal of computer assisted radiology and surgery
ISSN: 1861-6429
Titre abrégé: Int J Comput Assist Radiol Surg
Pays: Germany
ID NLM: 101499225
Informations de publication
Date de publication:
Sep 2023
Sep 2023
Historique:
received:
10
01
2023
accepted:
24
04
2023
medline:
11
9
2023
pubmed:
28
5
2023
entrez:
28
5
2023
Statut:
ppublish
Résumé
Endovascular intervention is the state-of-the-art treatment for common cardiovascular diseases, such as heart attack and stroke. Automation of the procedure may improve the working conditions of physicians and provide high-quality care to patients in remote areas, posing a major impact on overall treatment quality. However, this requires the adaption to individual patient anatomies, which currently poses an unsolved challenge. This work investigates an endovascular guidewire controller architecture based on recurrent neural networks. The controller is evaluated in-silico on its ability to adapt to new vessel geometries when navigating through the aortic arch. The controller's generalization capabilities are examined by reducing the number of variations seen during training. For this purpose, an endovascular simulation environment is introduced, which allows guidewire navigation in a parametrizable aortic arch. The recurrent controller achieves a higher navigation success rate of 75.0% after 29,200 interventions compared to 71.6% after 156,800 interventions for a feedforward controller. Furthermore, the recurrent controller generalizes to previously unseen aortic arches and is robust towards size changes of the aortic arch. Being trained on 2048 aortic arch geometries gives the same results as being trained with full variation when evaluated on 1000 different geometries. For interpolation a gap of 30% of the scaling range and for extrapolation additional 10% of the scaling range can be navigated successfully. Adaption to new vessel geometries is essential in the navigation of endovascular instruments. Therefore, the intrinsic generalization to new vessel geometries poses an essential step towards autonomous endovascular robotics.
Identifiants
pubmed: 37245181
doi: 10.1007/s11548-023-02938-7
pii: 10.1007/s11548-023-02938-7
pmc: PMC10491528
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
1735-1744Informations de copyright
© 2023. The Author(s).
Références
Schneider PA (2009) Endovascular skills: guidewire and catheter skills for endovascular surgery. Third, Informa Healthcare USA Inc, New York
Lanzer P (2018) Textbook of catheter-based cardiovascular interventions: a knowledge-based approach. Springer, Berlin
doi: 10.1007/978-3-319-55994-0
Ho P, Cheng SWK, Wu PM, Ting ACW, Poon JTC, Cheng CKM, Mok JHM, Tsang MS (2007) Ionizing radiation absorption of vascular surgeons during endovascular procedures. J Vasc Surg 46(3):455–459. https://doi.org/10.1016/j.jvs.2007.04.034
doi: 10.1016/j.jvs.2007.04.034
pubmed: 17826233
Böckler D (2020) Praktische Tipps für den persönlichen Strahlenschutz bei endovaskulären Eingriffen im Hybrid-Operationssaal. Gefässchirurgie 25(1):19–30. https://doi.org/10.1007/s00772-020-00620-9
doi: 10.1007/s00772-020-00620-9
Yan Y, Hu K, Alcock S, Ghrooda E, Trivedi A, McEachern J, Kaderali Z, Shankar J (2022) Access to endovascular thrombectomy for stroke in rural versus urban regions. Can J Neurol Sci J Can Sci Neurol 49(1):70–75. https://doi.org/10.1017/cjn.2021.35
doi: 10.1017/cjn.2021.35
Marescaux J, Leroy J, Gagner M, Rubino F, Mutter D, Vix M, Butner SE, Smith MK (2001) Transatlantic robot-assisted telesurgery. Nature 413(6854):379–380. https://doi.org/10.1038/35096636
doi: 10.1038/35096636
pubmed: 11574874
Bailo P, Gibelli F, Blandino A, Piccinini A, Ricci G, Sirignano A, Zoja R (2021) Telemedicine applications in the era of COVID-19: telesurgery issues. Int J Environ Res Public Health 19(1):323. https://doi.org/10.3390/ijerph19010323
doi: 10.3390/ijerph19010323
pubmed: 35010581
pmcid: 8751214
Zhao Y, Guo S, Wang Y, Cui J, Ma Y, Zeng Y, Liu X, Jiang Y, Li Y, Shi L, Xiao N (2019) A CNN-based prototype method of unstructured surgical state perception and navigation for an endovascular surgery robot. Med Biol Eng Comput 57(9):1875–1887. https://doi.org/10.1007/s11517-019-02002-0
doi: 10.1007/s11517-019-02002-0
pubmed: 31222531
Kweon J, Kim K, Lee C, Kwon H, Park J, Song K, Kim YI, Park J, Back I, Roh J-H, Moon Y, Choi J, Kim Y-H (2021) Deep reinforcement learning for guidewire navigation in coronary artery phantom. IEEE Access 9:166409–166422. https://doi.org/10.1109/ACCESS.2021.3135277
doi: 10.1109/ACCESS.2021.3135277
Song H-S, Yi B-J, Won JY, Woo J (2022) Learning-based catheter and guidewire-driven autonomous vascular intervention robotic system for reduced repulsive force. J Comput Des Eng 9(5):1549–1564. https://doi.org/10.1093/jcde/qwac074
doi: 10.1093/jcde/qwac074
Meng F, Guo S, Zhou W, Chen Z (2022) Evaluation of an autonomous navigation method for vascular interventional surgery in virtual environment. In: 2022 IEEE international conference on mechatronics and automation (ICMA), pp 1599–1604. https://doi.org/10.1109/ICMA54519.2022.9856107
Karstensen L, Ritter J, Hatzl J, Pätz T, Langejürgen J, Uhl C, Mathis-Ullrich F (2022) Learning-based autonomous vascular guidewire navigation without human demonstration in the venous system of a porcine liver. Int J Comput Assist Radiol Surg. https://doi.org/10.1007/s11548-022-02646-8
doi: 10.1007/s11548-022-02646-8
pubmed: 35604490
pmcid: 9515141
Chi W, Dagnino G, Kwok TMY, Nguyen A, Kundrat D, Abdelaziz MEMK, Riga C, Bicknell C, Yang GZ (2020) Collaborative robot-assisted endovascular catheterization with generative adversarial imitation learning. Proc IEEE Int Conf Robot Autom. https://doi.org/10.1109/ICRA40945.2020.9196912
doi: 10.1109/ICRA40945.2020.9196912
Yang D, Song J, Hu Y (2022) Guidewire feeding method based on deep reinforcement learning for vascular intervention robot. In: 2022 IEEE international conference on mechatronics and automation (ICMA), pp 1287–1293. https://doi.org/10.1109/ICMA54519.2022.9856351
Li H, Zhou X-H, Xie X-L, Liu S-Q, Gui M-J, Xiang T-Y, Wang J-L, Hou Z-G (2023) Discrete soft actor-critic with auto-encoder on vascular robotic system. Robotica 41(4):1115–1126. https://doi.org/10.1017/S0263574722001527
doi: 10.1017/S0263574722001527
Kirk R, Zhang A, Grefenstette E, Rocktäschel T (2022) A survey of generalisation in deep reinforcement learning. ArXiv211109794 Cs. Accessed: Feb 17, 2022. http://arxiv.org/abs/2111.09794
Natsis KI, Tsitouridis IA, Didagelos MV, Fillipidis AA, Vlasis KG, Tsikaras PD (2009) Anatomical variations in the branches of the human aortic arch in 633 angiographies: clinical significance and literature review. Surg Radiol Anat 31(5):319–323. https://doi.org/10.1007/s00276-008-0442-2
doi: 10.1007/s00276-008-0442-2
pubmed: 19034377
Faure F, Duriez C, Delingette H, Allard J, Gilles B, Marchesseau S, Talbot H, Courtecuisse H, Bousquet G, Peterlik I, Cotin S (2012) SOFA: a multi-model framework for interactive physical simulation. In: Payan Y (ed) Studies in mechanobiology, tissue engineering and biomaterials, vol 11. Springer, Berlin, pp 283–321. https://doi.org/10.1007/8415_2012_125
doi: 10.1007/8415_2012_125
Wei Y, Cotin S, Dequidt J, Duriez C, Allard J, Kerrie E (2012) A (near) real-time simulation method of aneurysm coil embolization. In: Murai Y (ed) Aneurysm. InTech, New York. https://doi.org/10.5772/48635
doi: 10.5772/48635
Haarnoja T, Zhou A, Hartikainen K, Tucker G, Ha S, Tan J, Kumar V, Zhu H, Gupta A, Abbeel P, Levine S (2019) Soft actor-critic algorithms and applications. ArXiv181205905 Cs Stat. Accessed: Jan 07, 2022. http://arxiv.org/abs/1812.05905
Meng L, Gorbet R, Kulic D (2021) Memory-based deep reinforcement learning for POMDPs. In: 2021 IEEE/RSJ international conference on intelligent robots and systems (IROS), Prague, Czech Republic, pp 5619–5626. https://doi.org/10.1109/IROS51168.2021.9636140
Ma L, Liu Y, Chen J, Jin D (2019) Learning to navigate in indoor environments: from memorizing to reasoning. CoRR abs/1904.06933. http://arxiv.org/abs/1904.06933
Wilson NM, Ortiz AK, Johnson AB (2013) The vascular model repository: a public resource of medical imaging data and blood flow simulation results. J Med Devices 7(4):040923. https://doi.org/10.1115/1.4025983
doi: 10.1115/1.4025983