A biplanar fluoroscopic approach for the measurement, modeling, and simulation of needle and soft-tissue interaction
Introduction
There are multiple procedures in surgery, where accurate placement of needles is vital to the success of the procedure. One such procedure is prostate brachytherapy during which the surgeon inserts a needle through the perineum, fatty tissue, muscle and then prostate to place radioactive seeds. The goal for seed placement is to maximize the radiation dose to the tumor while minimizing radiation doses to the surrounding tissue such as the rectum, bladder and urethra. Accurate seed positioning inside the prostate is very important for the success of the procedure. Minor deviations in seed placement caused by gland and tissue compression and needle retraction, gland edema, and needle deflections, can lead to significant areas of over dosage or under dosage to the gland (Vicini et al., 1999). For various needle insertion procedures such as prostate brachytherapy, lumbar puncture and liver biopsy the success rate of the procedure is directly related to the clinician’s level of expertise. Therefore, improvement in the complication rates will be dependent on improving the training tools used by clinicians. Obtaining real world parameters for characterizing needle and soft-tissue interaction is the first step towards developing a model to provide accurate haptic feedback in a training simulator for needle insertion tasks.
To date, a number of researchers have explored techniques to improve surgeons’ skills in procedures where little haptic and visual feedback exists. These mainly take the form of haptic simulators, for procedures such as catheter insertion (Gobbetti, 2000), lumbar puncture (Gorman, 2000), epidural blocks (Hiemenz, 1996, Brett et al., 2000), endoscopic surgeries, and laparoscopic surgeries. From the surgical simulation viewpoint, most tissue interaction models assume mechanical properties and develop methods to efficiently solve the tissue simulation problem for robot-assisted surgery/training. Several simulators have developed very sophisticated virtual environments that allow for plastic deformations of the material and interactions in multiple dimensions (Picinbono et al., 2001, Forest et al., 2002). However, it has been difficult to populate these models with data from real tissues.
“Global” elastic deformations of real and phantom tissues have been studied extensively in previous work through simple poking interactions (d’Aulignac et al., 2000, Brouwer, 2001, Ottensmeyer and Salisbury, 2001, Kennedy et al., 2002, Kennedy et al., 2002). However, these methods are simplistic since they do not consider the complex boundary conditions that are normally present, both internal to the organ and on the exterior surface. For needle insertion tasks, simulation and modeling have been conducted by a number of researchers (Stoianovici et al., 1998, Brett et al., 2000, Smith et al., 2001, DiMaio and Salcudean, 2002, Kataoka et al., 2002, Simone, 2002, Alterovitz et al., 2003, Nienhuys and van der Stappen, 2003, Hong et al., 2004, Magill et al., 2004. Only a few groups have modeled and studied the measurement of forces during needle insertion into soft tissue and the effects of needle geometry on the deflection during needle insertions into homogeneous material (Brett et al., 2000, Kataoka et al., 2002, Simone, 2002, O’Leary et al., 2003, Heverly et al., 2005). In this paper, we offer a different approach to separating the different forces acting on the needle and obtaining the cutting force. Our method requires only one force sensor and we can internally image soft tissue movement during a needle insertion and withdrawal task. Needle deflection is also an important part of our study because it has been observed during surgical procedures such as prostate brachytherapy that the needle can deflect from the initial insertion point as it is being inserted through the body by more than 10 mm (Cormack et al., 2000).
Simulations based on real needle insertion forces into soft tissues have been conducted by Brett et al., 2000, Simone, 2002, Heverly et al., 2005). In the work of Heverly and Simone (Simone, 2002, Heverly et al., 2005), the simulation is only a visual reproduction of the forces during needle insertion. In Brett et al. (2000), a force feedback device is integrated into the simulator but the simulation is only for needle insertion and does not include needle withdrawal. In this paper, we produce a simulation of needle insertion and withdrawal with haptic feedback using the 7 degree of freedom haptic feedback device developed in our laboratory. The simulated forces are based on ex vivo experimental data from needle insertion and withdrawal in a soft tissue specimen.
Imaging modalities used during needle insertion into soft tissue have mainly been utilized for needle guidance. However, some researchers have used the imaging modalities to track fiducials during needle insertion into clear homogeneous phantom tissues to obtain properties about the needle and phantom tissue interactions (DiMaio and Salcudean, 2002, Crouch et al., 2005). However, these experiments required knowledge of the properties of the phantom tissues prior to experiments and also required the phantom to have homogenous material properties. Kerdok et al. (2003) developed the truth cube which consists of fiducials implanted inside a homogeneous phantom tissue. The 3D motion of the markers during compression of the phantom was used to validate a finite element model. This work was one of the first approaches to quantify the internal local tissue movement. However, it did not include needle insertion nor was it conducted on biological tissues. Our approach involves tracking fiducials using two C-arm fluoroscopes. The fiducials are placed “semi-randomly” in a soft tissue. With the data collected from our experimental setup, we can extract the necessary parameters such as puncture force, cutting forces, and local effective modulus, required for accurate modeling of needle and soft-tissue interaction. We were also able to validate the finite element model for needle insertion into soft tissue for an arbitrary needle insertion task. To our knowledge, there has been no prior work on measuring in real-time the 3-D movement of fiducials (beads) in soft tissue during needle insertion and we believe that our approach using two C-arms is novel. This type of reality based modeling is critical for providing accurate haptic feedback in surgical simulation. Fig. 1 shows a schematic of the proposed reality based modeling approach.
This paper is divided into five sections. In Section 2, we discuss the materials and methods used for tracking the needle and implanted markers in the soft tissue during an insertion and withdrawal task. We also introduce our 3D finite element model and the 7 degree-of-freedom (DOF) haptic device used for simulation. The results are shown in Section 3, which also includes the results from the needle insertion simulation. Section 4 is our discussion of the results and in Section 5 we present our conclusions.
Section snippets
Materials and methods
Table 1 demonstrates our proposed approach for modeling needle and soft-tissue interaction during a needle insertion and withdrawal task. This paper presents the following: (a) the computational model for needle puncture to estimate the local tissue stiffness prior to puncture and (b) reality-based estimation of the friction force during needle insertion and withdrawal, and (c) the quantification of the force versus displacement plot during tissue relaxation.
Needle deflection and marker movement
Fig. 12 shows the trajectory of the needle through the soft tissue during needle insertion and withdrawal. Deflection of the needle tip from the straight line trajectory was observed during every needle experiment. Table 2 shows the average deflection for the insertion speeds of 1.016 mm/s, 12.7 mm/s, and 25.4 mm/s. Although within one standard deviation of each other, deflection was found to increase slightly with an increased needle velocity. The maximum deflection from the straight line needle
Needle movement
Needle deflection is important to measure and predict, especially for training radiation oncologists to accurately place seeds in the prostate during a prostate brachytherapy procedure. In a typical prostate brachytherapy task the needle can deflect as much as 10 mm from the initial insertion point as it travels through the perineum, fat, muscle, and then prostate (Cormack et al., 2000). This deflection of the needle sometimes requires recomputation of the dosage information for radioactive seed
Conclusion
This study demonstrates a unique approach for estimating needle and soft tissue interaction. Using a dual C-arm fluoroscope setup and implanted radio-opaque markers, we can obtain in real time, 3D needle trajectory and internal global and local tissue deformation during needle insertion into soft tissue. Information such as the time of tissue puncture, various internal puncture events, tissue relaxation, and the start of kinetic friction during withdrawal can be extracted from the X-ray images
Acknowledgements
This work was supported in part by National Science Foundation CAREER Award IIS 0133471, NSF ITR Award 0312709 and the Beukenkamp Endowment for Prostate Cancer in Drexel University College of Medicine, Department of Surgery. We thank Dr. Waqus Anjum, Dr. Justin Chandler, and Jeff Justin for their valuable help in conducting dual C-ARM experiments and C-arm calibration. We thank Gregory Tholey for allowing us to use the haptic feedback device for our needle simulation trials.
References (33)
- et al.
Optimizing target coverage by dosimetric feedback during prostate brachytherapy
International Journal of Radiation Oncology Biology and Physics
(2000) - et al.
Truth cube: establishing physical standards for real time soft tissue simulation
Medical Image Analysis
(2003) - et al.
Three dimensional ultrasound-guided core needle breast biopsy
Ultrasound in Medicine and Biology
(2001) - et al.
A comprehensive review of prostate cancer brachytherapy: defining an optimal technique
International Journal of Radiation Oncology Biology and Physics
(1999) - Alterovitz, R., Lim, A. et al. 2005. Steering flexible needles under markov motion uncertainty. In: Proceedings of the...
- et al.
Simulating needle insertion and radioactive seed implantation for prostate brachytherapy
Medicine Meets Virtual Reality
(2003) - et al.
Schemes for the identification of tissue types and boundaries at the tool point for surgical needles
IEEE Transactions on Information Technology in Biomedicine
(2000) Measuring in vivo animal soft tissue properties for haptic modeling in surgical simulation
Medicine Meets Virtual Reality
(2001)- et al.
Computer-aided radiation therapy simulation: image intensifier spatial distortion correction for large field of view digital flouroscopy
Physics in Medicine and Biology
(1998) - et al.
A velocity dependent model for needle insertion in soft tissue
Mathematical modeling for keyhole surgery simulations: a biomechanical model for spleen tissue
IMA Journal of Applied Mathematics
Three-dimensional computed tomographic reconstruction using a C-arm mounted XRII: correction of image intensifier distortion
Medical Physics
Catheter insertion simulation with co-registered direct volume rendering and haptic feedback
Medicine Meets Virtual Reality
Cited by (58)
Cutting characteristics of viscoelastic membranes under hypodermic needle insertion
2024, International Journal of Mechanical SciencesIssues in closed-loop needle steering
2017, Control Engineering PracticeA generic three-dimensional static force distribution basis for a medical needle inserted into soft tissue
2013, Journal of the Mechanical Behavior of Biomedical MaterialsCitation Excerpt :These reasons include the different characteristics of the various layers of tissues encountered by the needle (fat, muscle, etc.), the dependence of their biomechanical properties on the patient (age, disease, etc.) and the properties of the needle itself. Several authors have considered the forces involved during the dynamic insertion of a needle (Kataoka et al., 2002; DiMaio and Salcudean, 2002; Okamura et al., 2004; Washio and Chinzei, 2004; Crouch et al., 2005; Podder et al., 2006a; Hing et al., 2007; Misra et al., 2008; Roesthuis et al., 2011) as shown in Fig. 1(a), and as a function of insertion depth. Furthermore, for most of the forces considered (Simone and Okamura, 2002; Dehghan et al., 2008; Abayazid et al., 2011; Asadian et al., 2012), the associated parameters of the model have to be fitted using experimental data from previous insertions.
3D navigation and monitoring for spinal milling operation based on registration between multiplanar fluoroscopy and CT images
2012, Computer Methods and Programs in BiomedicineCitation Excerpt :One way is to adopt orthogonal imaging devices in radiotherapy [17,18]. Two C-Arms are required for intraoperative bi-planar navigation [19], but the configuration is quite cumbersome. Another category is more suitable for CAOS systems by using just one conventional imaging device.
Needle-tissue interaction forces - A survey of experimental data
2012, Medical Engineering and PhysicsCitation Excerpt :Different approaches can be adopted for measuring the friction force acting on a needle. For example, Hing et al. [19,20] assume that the force during removal of the needle is due to friction only, thereby yielding a direct friction measurement. The resulting friction curve shown in Fig. 4, for porcine liver, can be interpreted as approximately linear.