Force classification during robotic interventions through simulation-trained neural networks
Intravitreal injection is among the most frequent treatment strategies for chronic ophthalmic diseases. The last decade has seen a serious increase in the number of intravitreal injections, and with it, adverse effects and drawbacks. To tackle these problems, medical assistive devices for robotized injections have been suggested and are projected to enhance delivery mechanisms for a new generation of pharmacological solutions. In this paper, we present a method aimed at improving the safety characteristics of upcoming robotic systems. Our vision-based method uses a combination of 2D OCT data, numerical simulation and machine learning to classify the range of the force applied by an injection needle on the sclera.
We design a neural network to classify force ranges from optical coherence tomography (OCT) images of the sclera directly. To avoid the need for large real data sets, the network is trained on images of simulated deformed sclera. This simulation is based on a finite element method, and the model is parameterized using a Bayesian filter applied to observations of the deformation in OCT images.
We validate our approach on real OCT data collected on five ex vivo porcine eyes using a robotically guided needle. The thorough parameterization of the simulations leads to a very good agreement between the virtually generated samples used to train the network and the real OCT acquisitions. Results show that the applied force range on real data can be predicted with 93% accuracy.
Through a simulation-trained neural network, our approach estimates the force range applied by a robotically guided needle on the sclera based solely on a single OCT slice of the deformed sclera. Being real-time, this solution can be integrated in the control loop of the system, permitting the prompt withdrawal of the needle for safety reasons.
KeywordsFinite element modeling Bayesian inference Artificial neural networks Force estimation in robotics
The authors would like to thank Jan Hermann and Tatiana Fountoukidou for the help in the experimental data gathering and Igor Peterlik for his advice and code on Bayesian filtering.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Human and animal rights
This article does not contain any studies with animals performed by any of the authors.
This articles does not contain patient data.
- 3.Jagtap AD, Riviere CN (2004) Applied force during vitreoretinal microsurgery with handheld instruments. In: The 26th annual international conference of the IEEE engineering in medicine and biology society, vol 1, pp 2771–2773. IEEEGoogle Scholar
- 4.Weber S, Gavaghan K, Wimmer W, Williamson T, Gerber N, Anso J, Bell B, Feldmann A, Rathgeb C, Matulic M, Stebinger M, Schneider D, Mantokoudis G, Scheidegger O, Wagner F, Kompis M, Caversaccio M (2017) Instrument flight to the inner ear. Sci Robot 2(4):eaal4916 CrossRefPubMedPubMedCentralGoogle Scholar
- 5.Haidegger T, Benyó B, Kovács L, Benyó Z (2009). Force sensing and force control for surgical robots. In: 7th IFAC symposium on modeling and control in biomedical systems, vol, 7, no 1, pp 413–418Google Scholar
- 8.Aviles AI, Marban A, Sobrevilla P, Fernandez J, Casals A (2014) A recurrent neural network approach for 3D vision-based force estimation. In: International conference on image processing theory, tools and applications (IPTA). pp 1–6. IEEEGoogle Scholar
- 9.Pakhomov D, Premachandran V, Allan M, Azizian M, Navab N (2017) Deep residual learning for instrument segmentation in robotic surgery. arXiv preprint arXiv:1703.08580
- 10.Aviles AI, Alsaleh S, Sobrevilla P, Casals A (2015) Sensorless force estimation using a neuro-vision-based approach for robotic-assisted surgery. In: 7th International IEEE/EMBS conference on neural engineering (NER), 2015. pp 86–89. IEEEGoogle Scholar
- 11.Aggarwal V, Asadi H, Gupta M, Lee JJ, Yu D (2018) Covfefe: a computer vision approach for estimating force exertion. arXiv preprint arXiv:1809.09293
- 12.Mendizabal A, Fountoukidou T, Hermann J, Sznitman R, Cotin S (2018) A combined simulation and machine learning approach for image-based force classification during robotized intravitreal injections. In: International conference on medical image computing and computer-assisted intervention, pp 12–20Google Scholar
- 13.Moireau P, Chapelle D (2011) Reduced-order Unscented Kalman Filtering with application to parameter identification in large-dimensional systems. ESAIM: Control Optim Calc Var 17(2):380–405Google Scholar
- 15.Bro-Nielsen M, Cotin S (1996) Real-time volumetric deformable models for surgery simulation using finite elements and condensation. In: Computer graphics forum. Blackwell Science Ltd, Edinburgh, vol 15, no 3, pp 57–66Google Scholar
- 16.Olsen TW, Sanderson S, Feng X, Hubbard WC (2002) Porcine sclera: thickness and surface area. Investig Ophthalmol Vis Sci 43(8):2529–2532Google Scholar