Modelling the indentation force response of non-uniform soft tissue using a recurrent neural network

A scaled recurrent neural network (RNN) model is developed which accurately predicts the force response from the indentation of a non-uniform soft tissue sample. The model consists of two components. The RNN is used to predict the force response of indentation using data from a reference tissue sample. A two-parameter component then scales the neural networks predictions relative to previously determined properties of the test sample. This component is based on a strain inverse model of force, which is used to account for the non-uniformity of the tissue between the test and reference data. Experimental force measurements were performed on a highly non-uniform soft tissue analogue to develop and validate the model. Using the visco-elastic Hunt-Crossley model as a benchmark, the developed model provides significantly better prediction. Future research will investigate applying this model to surgical simulations and verifying its application to different biological tissues.

[1]  Rajni V. Patel,et al.  Needle insertion into soft tissue: a survey. , 2007, Medical engineering & physics.

[2]  Hany El Kadi,et al.  Modeling the mechanical behavior of fiber-reinforced polymeric composite materials using artificial neural networks—A review , 2006 .

[3]  O. Kisi,et al.  Predicting the compressive strength of steel fiber added lightweight concrete using neural network , 2008 .

[4]  Helena Pereira,et al.  Prediction of mechanical strength of cork under compression using machine learning techniques , 2015 .

[5]  R. Ogden.  Large deformation isotropic elasticity – on the correlation of theory and experiment for incompressible rubberlike solids , 1972, Proceedings of the Royal Society of London. A. Mathematical and Physical Sciences.

[6]  D. C. Lin,et al.  Spherical indentation of soft matter beyond the Hertzian regime: numerical and experimental validation of hyperelastic models , 2009, Biomechanics and modeling in mechanobiology.

[7]  Bijan Shirinzadeh,et al.  Motion control analysis of a parallel robot assisted minimally invasive surgery/microsurgery system (PRAMiSS) , 2013 .

[8]  Blake Hannaford,et al.  Biomechanical properties of abdominal organs in vivo and postmortem under compression loads. , 2008, Journal of biomechanical engineering.

[9]  Allan Pinkus,et al.  Multilayer Feedforward Networks with a Non-Polynomial Activation Function Can Approximate Any Function , 1991, Neural Networks.

[10]  Eduardo Sontag,et al.  Turing computability with neural nets , 1991 .

[11]  Bijan Shirinzadeh,et al.  A Cellular Neural Network Methodology for Deformable Object Simulation , 2006, IEEE Transactions on Information Technology in Biomedicine.

[12]  C. Lee Giles,et al.  An experimental comparison of recurrent neural networks , 1994, NIPS.

[13]  Allison M. Okamura,et al.  Force modeling for needle insertion into soft tissue , 2004, IEEE Transactions on Biomedical Engineering.

[14]  Stefano Stramigioli,et al.  Bilateral Telemanipulation With Time Delays: A Two-Layer Approach Combining Passivity and Transparency , 2011, IEEE Transactions on Robotics.

[15]  T.,et al.  Training Feedforward Networks with the Marquardt Algorithm , 2004 .

[16]  Eugen Diaconescu,et al.  The use of NARX neural networks to predict chaotic time series , 2008 .

[17]  Geoffrey E. Hinton,et al.  On the importance of initialization and momentum in deep learning , 2013, ICML.

[18]  Ahmed M. Azmy,et al.  Neural networks for predicting compressive strength of structural light weight concrete , 2009 .

[19]  Kumar Vemaganti,et al.  A Bayesian approach to selecting hyperelastic constitutive models of soft tissue , 2015 .

[20]  Inna Sharf,et al.  Literature survey of contact dynamics modelling , 2002 .

[21]  K. Miller,et al.  Total Lagrangian explicit dynamics finite element algorithm for computing soft tissue deformation , 2006 .

[22]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[23]  K. H. Hunt,et al.  Coefficient of Restitution Interpreted as Damping in Vibroimpact , 1975 .

[24]  Septimiu E. Salcudean,et al.  Needle insertion modeling and simulation , 2003, IEEE Trans. Robotics Autom..

[25]  B. Hannaford,et al.  In-vivo and in-situ compressive properties of porcine abdominal soft tissues. , 2003, Studies in health technology and informatics.

[26]  Bijan Shirinzadeh,et al.  Sliding Mode Control of a Piezoelectric Actuator with Neural Network Compensating Rate-Dependent Hysteresis , 2005, Proceedings of the 2005 IEEE International Conference on Robotics and Automation.

[27]  Hava T. Siegelmann,et al.  Computational capabilities of recurrent NARX neural networks , 1997, IEEE Trans. Syst. Man Cybern. Part B.