Advertisement

Robot-assisted flexible needle insertion using universal distributional deep reinforcement learning

  • Xiaoyu TanEmail author
  • Yonggu Lee
  • Chin-Boon Chng
  • Kah-Bin Lim
  • Chee-Kong Chui
Original Article
  • 81 Downloads

Abstract

Purpose

Flexible needle insertion is an important minimally invasive surgery approach for biopsy and radio-frequency ablation. This approach can minimize intraoperative trauma and improve postoperative recovery. We propose a new path planning framework using multi-goal deep reinforcement learning to overcome the difficulties in uncertain needle–tissue interactions and enhance the robustness of robot-assisted insertion process.

Methods

This framework utilizes a new algorithm called universal distributional Q-learning (UDQL) to learn a stable steering policy and perform risk management by visualizing the learned Q-value distribution. To further improve the robustness, universal value function approximation is leveraged in the training process of UDQL to maximize generalization and connect to diagnosis by adapting fast re-planning and transfer learning.

Results

Computer simulation and phantom experimental results show our proposed framework can securely steer flexible needles with high insertion accuracy and robustness. The framework also improves robustness by providing distribution information to clinicians for diagnosis and decision making during surgery.

Conclusions

Compared with previous methods, the proposed framework can perform multi-target needle insertion through single insertion point qunder continuous state space model with higher accuracy and robustness.

Keywords

Deep learning Deep reinforcement learning Needle steering Tool–tissue interaction Uncertainty 

Notes

Acknowledgements

The last author would like to acknowledge the contribution of A/Prof Stephen Chang of Mount Elizabeth Hospital, Singapore for his input on surgeries and medical education.

Funding

The research and development of the prototype Image-guide Radio-frequency Ablation Surgical System was supported in parts by Research Grants from Singapore Agency of Science and Technology (A*Star) and Ministry of Education, Singapore respectively.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Informed consent

Informed consent was obtained from all individual participants included in the study.

Human and animal rights

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

References

  1. 1.
    Hamad GG, Curet M (2010) Minimally invasive surgery. Am J Surg 199(2):263–265CrossRefGoogle Scholar
  2. 2.
    Tan X, Chng C-B, Ye S, Lim K-B, Chui C-K (2019) Robot-assisted training in laparoscopy using deep reinforcement learning. IEEE Robot Autom Lett 4(2):485–492CrossRefGoogle Scholar
  3. 3.
    Hiraki T, Kamegawa T, Matsuno T, Sakurai J, Kirita Y, Matsuura R, Yamaguchi T, Sasaki T, Mitsuhashi T, Komaki T (2017) Robotically driven ct-guided needle insertion: preliminary results in phantom and animal experiments. Radiology 285(2):454–461CrossRefGoogle Scholar
  4. 4.
    Abolhassani N, Patel R, Moallem M (2007) Needle insertion into soft tissue: a survey. Med Eng Phys 29(4):413–431CrossRefGoogle Scholar
  5. 5.
    Schaul T, Horgan D, Gregor K, Silver D (2015a) Universal value function approximators. In: International conference on machine learning. pp 1312–1320Google Scholar
  6. 6.
    Tan X, Yu P, Lim K-B, Chui C-K (2018) Robust path planning for flexible needle insertion using Markov decision processes. Int J Comput Assist Radiol Surg 13(9):1439–1451CrossRefGoogle Scholar
  7. 7.
    Duan B, Wen R, Chng C-B, Wang W, Liu P, Qin J, Peneyra JL, Chang SK-Y, Heng P-A, Chui C-K (2015) Image-guided robotic system for radiofrequency ablation of large liver tumor with single incision. In: 2015 12th International conference on ubiquitous robots and ambient intelligence (URAI). IEEE, pp 284–289Google Scholar
  8. 8.
    DiMaio SP, Salcudean SE (2005) Needle steering and motion planning in soft tissues. IEEE Trans Biomed Eng 52(6):965–974CrossRefGoogle Scholar
  9. 9.
    Taylor RH, Menciassi A, Fichtinger G, Fiorini P, Dario P (2016) Medical robotics and computer-integrated surgery. In: Siciliano B, Khatib O (eds) Springer handbook of robotics. Springer, Cham, pp 1657–1684CrossRefGoogle Scholar
  10. 10.
    Liu P, Qin J, Duan B, Wang Q, Tan X, Zhao B, Jonnathan PL, Chui C-K, Heng P-A (2018) Overlapping radiofrequency ablation planning and robot-assisted needle insertion for large liver tumors. Int J Med Robot Comput Assist Surg 15:e1952CrossRefGoogle Scholar
  11. 11.
    Chatelain P, Krupa A, Navab N (2015) 3d ultrasound-guided robotic steering of a flexible needle via visual servoing. In: IEEE international conference on robotics and automation, ICRA’15Google Scholar
  12. 12.
    Alterovitz R, Siméon T, Goldberg KY (2007) The stochastic motion roadmap: a sampling framework for planning with Markov motion uncertainty. In: Robotics: science and systems, vol 3, pp 233–241Google Scholar
  13. 13.
    Alterovitz R, Branicky M, Goldberg K (2008) Motion planning under uncertainty for image-guided medical needle steering. Int J Robot Res 27(11–12):1361–1374CrossRefGoogle Scholar
  14. 14.
    Morar A, Moldoveanu F, Gröller E (2012) Image segmentation based on active contours without edges. In: 2012 IEEE 8th international conference on intelligent computer communication and processing. IEEE, pp 213–220Google Scholar
  15. 15.
    Chen X, Nguyen BP, Chui C-K, Ong S-H (2016) Automated brain tumor segmentation using kernel dictionary learning and superpixel-level features. In: 2016 IEEE international conference on systems, man, and cybernetics (SMC). IEEE, pp 002547–002552Google Scholar
  16. 16.
    Sutton RS, Barto AG (1998) Introduction to reinforcement learning, vol 135. MIT Press, CambridgeGoogle Scholar
  17. 17.
    Deng L, Yu D (2014) Deep learning: methods and applications. Found Trends® Signal Process 7(3–4):197–387CrossRefGoogle Scholar
  18. 18.
    Silver D, Schrittwieser J, Simonyan K, Antonoglou I, Huang A, Guez A, Hubert T, Baker L, Lai M, Bolton A (2017) Mastering the game of go without human knowledge. Nature 550(7676):354CrossRefGoogle Scholar
  19. 19.
    Schulman J, Wolski F, Dhariwal P, Radford A, Klimov O (2017) Proximal policy optimization algorithms. arXiv:1707.06347
  20. 20.
    Fu YB, Chui CK, Teo CL (2013) Liver tissue characterization from uniaxial stress–strain data using probabilistic and inverse finite element methods. J Mech Behav Biomed Mater 20:105–112CrossRefGoogle Scholar
  21. 21.
    Fu YB, Chui CK (2014) Modelling and simulation of porcine liver tissue indentation using finite element method and uniaxial stress–strain data. J Biomech 47(10):2430–2435CrossRefGoogle Scholar
  22. 22.
    Qu C, Mannor S, Xu H (2018) Nonlinear distributional gradient temporal-difference learning. arXiv:1805.07732
  23. 23.
    Bellemare MG, Dabney W, Munos R (2017) A distributional perspective on reinforcement learning. arXiv:1707.06887
  24. 24.
    Andrychowicz M, Wolski F, Ray A, Schneider J, Fong R, Welinder P, McGrew B, Tobin J, Abbeel OP, Zaremba W (2017) Hindsight experience replay. In: Advances in neural information processing systems, pp 5048–5058Google Scholar
  25. 25.
    Schaul T, Quan J, Antonoglou I, Silver D (2015b) Prioritized experience replay. arXiv:1511.05952
  26. 26.
    Tamar A, Di Castro D, Mannor S (2016) Learning the variance of the reward-to-go. J Mach Learn Res 17(1):361–396Google Scholar
  27. 27.
    Yang L, Wen R, Qin J, Chui C-K, Lim K-B, Chang SK-Y (2010) A robotic system for overlapping radiofrequency ablation in large tumor treatment. IEEE/ASME Trans Mechatron 15(6):887–897CrossRefGoogle Scholar
  28. 28.
    Tan X, Chng C-B, Duan B, Ho Y, Wen R, Chen X, Lim K-B, Chui C-K (2017) Cognitive engine for robot-assisted radio-frequency ablation system. Acta Polytech Hung 14(1):129–145Google Scholar
  29. 29.
    Tan X, Chng C-B, Duan B, Ho Y, Wen R, Chen X, Lim K-B, Chui C-K (2016) Design and implementation of a patient-specific cognitive engine for robotic needle insertion. In: 2016 IEEE international conference on systems, man, and cybernetics (SMC). IEEE, pp 000560–000565Google Scholar
  30. 30.
    Van Hasselt H, Guez A, Silver D (2016) Deep reinforcement learning with double Q-learning. In: AAAI, vol 16, pp 2094–2100Google Scholar
  31. 31.
    Wasserstein RL, Lazar NA (2016) The asa’s statement on p values: context, process, and purpose. Am Stat 70(2):129–133CrossRefGoogle Scholar
  32. 32.
    Leong F, Huang W-H, Chui C-K (2013) Modeling and analysis of coagulated liver tissue and its interaction with a scalpel blade. Med Biol Eng Comput 51(6):687–695CrossRefGoogle Scholar
  33. 33.
    Tokuda J, Song S-E, Fischer GS, Iordachita II, Seifabadi R, Cho NB, Tuncali K, Fichtinger G, Tempany CM, Hata N (2012) Preclinical evaluation of an MRI-compatible pneumatic robot for angulated needle placement in transperineal prostate interventions. Int J Comput Assist Radiol Surg 7(6):949–957CrossRefGoogle Scholar
  34. 34.
    Krieger A, Susil RC, Fichtinger G, Atalar E, Whitcomb LL (2004) Design of a novel MRI compatible manipulator for image guided prostate intervention. In: IEEE international conference on robotics and automation, proceedings on ICRA’04, vol 1. IEEE, pp 377–382Google Scholar
  35. 35.
    Schouten MG, Ansems J, Renema WKJ, Bosboom D, Scheenen TWJ, Fütterer JJ (2010) The accuracy and safety aspects of a novel robotic needle guide manipulator to perform transrectal prostate biopsies. Med Phys 37(9):4744–4750CrossRefGoogle Scholar

Copyright information

© CARS 2019

Authors and Affiliations

  1. 1.Department of Mechanical EngineeringNational University of SingaporeSingaporeSingapore

Personalised recommendations