Accurate and efficient modeling of soft-tissue interactions is fundamental for advancing surgical simulation, surgical robotics, and model-based surgical automation. To achieve real-time latency, classical Finite Element Method (FEM) solvers are often replaced with neural approximations; however, naively training such models in a fully data-driven manner without incorporating physical priors frequently leads to poor generalization and physically implausible predictions. We present a novel physics-informed neural simulation framework that enables real-time prediction of soft-tissue deformations under complex single- and multi-grasper interactions. Our approach integrates Kelvinlet-based analytical priors with large-scale FEM data, capturing both linear and nonlinear tissue responses. This hybrid design improves predictive accuracy and physical plausibility across diverse neural architectures while maintaining the low-latency performance required for interactive applications. We validate our method on challenging surgical manipulation tasks involving standard laparoscopic grasping tools, demonstrating substantial improvements in deformation fidelity and temporal stability over existing baselines. These results establish Kelvinlet-augmented learning as a principled and computationally efficient paradigm for real-time, physics-aware soft-tissue simulation in surgical AI.

Image-guided surgery collocates patient-specific data with the physical environment to facilitate surgical decision making. Unfortunately, these guidance systems commonly become compromised by intraoperative soft-tissue deformations. Nonrigid image-to-physical registration methods have been proposed to compensate for deformations, but clinical utility requires compatibility of these techniques with data sparsity and temporal constraints in the operating room. While finite element models can be effective in sparse data scenarios, computation time remains a limitation to widespread deployment. This paper proposes a registration algorithm that uses regularized Kelvinlets, which are analytical solutions to linear elasticity in an infinite domain, to overcome these barriers. This algorithm is demonstrated and compared to finite element-based registration on two datasets: a phantom liver deformation dataset and an in vivo breast deformation dataset. The regularized Kelvinlets algorithm resulted in a significant reduction in computation time compared to the finite element method. Accuracy as evaluated by target registration error was comparable between methods. Average target registration errors were 4.6 ± 1.0 and 3.2 ± 0.8 mm on the liver dataset and 5.4 ± 1.4 and 6.4 ± 1.5 mm on the breast dataset for the regularized Kelvinlets and finite element method, respectively. Limitations of regularized Kelvinlets include the lack of organ-specific geometry and the assumptions of linear elasticity and infinitesimal strain. Despite limitations, this work demonstrates the generalizability of regularized Kelvinlets registration on two soft-tissue elastic organs. This method may improve and accelerate registration for image-guided surgery, and it shows the potential of using regularized Kelvinlets on medical imaging data.

Image-guided intervention for soft tissue organs depends on the accuracy of deformable registration methods to achieve effective results. While registration techniques based on elastic theory are prevalent, no methods yet exist that can prospectively estimate registration uncertainty to regulate sources and mitigate consequences of localization error in deforming organs. This paper introduces registration uncertainty metrics based on dispersion of strain energy from boundary constraints to predict the proportion of target registration error (TRE) remaining after nonrigid elastic registration. These uncertainty metrics depend on the spatial distribution of intraoperative constraints provided to registration with relation to patient-specific organ geometry. Predictive linear and bivariate gamma models are fit and cross-validated using an existing dataset of 6291 simulated registration examples, plus 699 novel simulated registrations withheld for independent validation. Average uncertainty and average proportion of TRE remaining after elastic registration are strongly correlated ( r = 0.78 ), with mean absolute difference in predicted TRE equivalent to 0.9 ± 0.6 mm (cross-validation) and 0.9 ± 0.5 mm (independent validation). Spatial uncertainty maps also permit localized TRE estimates accurate to an equivalent of 3.0 ± 3.1 mm (cross-validation) and 1.6 ± 1.2 mm (independent validation). Additional clinical evaluation of vascular features yields localized TRE estimates accurate to 3.4 ± 3.2 mm. This work formalizes a lower bound for the inherent uncertainty of nonrigid elastic registrations given coverage of intraoperative data constraints, and demonstrates a relation to TRE that can be predictively leveraged to inform data collection and provide a measure of registration confidence for elastic methods.

During image guided liver surgery, soft tissue deformation can cause considerable error when attempting to achieve accurate localization of the surgical anatomy through image-to-physical registration. In this paper, a linearized iterative boundary reconstruction technique is proposed to account for these deformations. The approach leverages a superposed formulation of boundary conditions to rapidly and accurately estimate the deformation applied to a preoperative model of the organ given sparse intraoperative data of surface and subsurface features. With this method, tracked intraoperative ultrasound (iUS) is investigated as a potential data source for augmenting registration accuracy beyond the capacity of conventional organ surface registration. In an expansive simulated dataset, features including vessel contours, vessel centerlines, and the posterior liver surface are extracted from iUS planes. Registration accuracy is compared across increasing data density to establish how iUS can be best employed to improve target registration error (TRE). From a baseline average TRE of 11.4 ± 2.2 mm using sparse surface data only, incorporating additional sparse features from three iUS planes improved average TRE to 6.4 ± 1.0 mm. Furthermore, increasing the sparse coverage to 16 tracked iUS planes improved average TRE to 3.9 ± 0.7 mm, exceeding the accuracy of registration based on complete surface data available with more cumbersome intraoperative CT without contrast. Additionally, the approach was applied to three clinical cases where on average error improved 67% over rigid registration and 56% over deformable surface registration when incorporating additional features from one independent tracked iUS plane.

In open image-guided liver surgery (IGLS), a sparse representation of the intraoperative organ surface can be acquired to drive image-to-physical registration. We hypothesize that uncharacterized error induced by variation in the collection patterns of organ surface data limits the accuracy and robustness of an IGLS registration. Clinical validation of such registration methods is challenged due to the difficulty in obtaining data representative of the true state of organ deformation. We propose a novel human-to-phantom validation framework that transforms surface collection patterns from in vivo IGLS procedures (n = 13) onto a well-characterized hepatic deformation phantom for the purpose of validating surface-driven, volumetric nonrigid registration methods. An important feature of the approach is that it centers on combining workflow-realistic data acquisition and surgical deformations that are appropriate in behavior and magnitude. Using the approach, we investigate volumetric target registration error (TRE) with both current rigid IGLS and our improved nonrigid registration methods. Additionally, we introduce a spatial data resampling approach to mitigate the workflow-sensitive sampling problem. Using our human-to-phantom approach, TRE after routine rigid registration was 10.9 ± 0.6 mm with a signed closest point distance associated with residual surface fit in the range of ±10 mm, highly representative of open liver resections. After applying our novel resampling strategy and improved deformation correction method, TRE was reduced by 51%, i.e., a TRE of 5.3 ± 0.5 mm. This paper reported herein realizes a novel tractable approach for the validation of image-to-physical registration methods and demonstrates promising results for our correction method.

In open abdominal image-guided liver surgery, sparse measurements of the organ surface can be taken intraoperatively via a laser-range scanning device or a tracked stylus with relatively little impact on surgical workflow. We propose a novel nonrigid registration method which uses sparse surface data to reconstruct a mapping between the preoperative CT volume and the intraoperative patient space. The mapping is generated using a tissue mechanics model subject to boundary conditions consistent with surgical supportive packing during liver resection therapy. Our approach iteratively chooses parameters which define these boundary conditions such that the deformed tissue model best fits the intraoperative surface data. Using two liver phantoms, we gathered a total of five deformation datasets with conditions comparable to open surgery. The proposed nonrigid method achieved a mean target registration error (TRE) of 3.3 mm for targets dispersed throughout the phantom volume, using a limited region of surface data to drive the nonrigid registration algorithm, while rigid registration resulted in a mean TRE of 9.5 mm. In addition, we studied the effect of surface data extent, the inclusion of subsurface data, the trade-offs of using a nonlinear tissue model, robustness to rigid misalignments, and the feasibility in five clinical datasets.

Steerable needles can potentially increase the accuracy of needle-based diagnosis and therapy delivery, provided they can be adequately controlled based on medical image information. We propose a novel sliding mode control law that can be used to deliver the tip of a flexible asymmetric-tipped needle to a desired point, or to track a desired trajectory within tissue. The proposed control strategy requires no a priori knowledge of model parameters, has bounded input speeds, and requires little computational resources. We show that if the standard nonholonomic model for tip-steered needles holds, then the control law will converge to desired targets in a reachable workspace, within a tolerance that can be defined by the control parameters. Experimental results validate the control law for target points and trajectory following in phantom tissue and ex vivo liver. Experiments with targets that move during insertion illustrate robustness to disturbances caused by tissue deformation.

Image-guided liver surgery requires the ability to identify and compensate for soft tissue deformation in the organ. The predeformed state is represented as a complete three-dimensional surface of the organ, while the intraoperative data is a range scan point cloud acquired from the exposed liver surface. The first step is to rigidly align the coordinate systems of the intraoperative and preoperative data. Most traditional rigid registration methods minimize an error metric over the entire data set. In this paper, a new deformation-identifying rigid registration (DIRR) is reported that identifies and aligns minimally deformed regions of the data using a modified closest point distance cost function. Once a rigid alignment has been established, deformation is accounted for using a linearly elastic finite element model (FEM) and implemented using an incremental framework to resolve geometric nonlinearities. Boundary conditions for the incremental formulation are generated from intraoperatively acquired range scan surfaces of the exposed liver surface. A series of phantom experiments is presented to assess the fidelity of the DIRR and the combined DIRR/FEM approaches separately. The DIRR approach identified deforming regions in 90% of cases under conditions of realistic surgical exposure. With respect to the DIRR/FEM algorithm, subsurface target errors were correctly located to within 4 mm in phantom experiments.

This paper reports a novel method to track brain shift using a laser-range scanner (LRS) and nonrigid registration techniques. The LRS used in this paper is capable of generating textured point-clouds describing the surface geometry/intensity pattern of the brain as presented during cranial surgery. Using serial LRS acquisitions of the brain's surface and two-dimensional (2-D) nonrigid image registration, we developed a method to track surface motion during neurosurgical procedures. A series of experiments devised to evaluate the performance of the developed shift-tracking protocol are reported. In a controlled, quantitative phantom experiment, the results demonstrate that the surface shift-tracking protocol is capable of resolving shift to an accuracy of approximately 1.6 mm given initial shifts on the order of 15 mm. Furthermore, in a preliminary in vivo case using the tracked LRS and an independent optical measurement system, the automatic protocol was able to reconstruct 50% of the brain shift with an accuracy of 3.7 mm while the manual measurement was able to reconstruct 77% with an accuracy of 2.1 mm. The results suggest that a LRS is an effective tool for tracking brain surface shift during neurosurgery.

The correlation between tissue stiffness and health is an accepted form of organ disease assessment. As a result, there has been a significant amount of interest in developing methods to image elasticity parameters (i.e., elastography). The modality independent elastography (MIE) method combines a nonlinear optimization framework, computer models of soft-tissue deformation, and standard measures of image similarity to reconstruct elastic property distributions within soft tissue. In this paper, simulation results demonstrate successful elasticity image reconstructions in breast cross-sectional images acquired from magnetic resonance (MR) imaging. Results from phantom experiments illustrate its modality independence by reconstructing elasticity images of the same phantom in both MR and computed tomographic imaging units. Additional results regarding the performance of a new multigrid strategy to MIE and the implementation of a parallel architecture are also presented.

In this paper, a method of acquiring intraoperative data using a laser range scanner (LRS) is presented within the context of model-updated image-guided surgery. Registering textured point clouds generated by the LRS to tomographic data is explored using established point-based and surface techniques as well as a novel method that incorporates geometry and intensity information via mutual information (SurfaceMI). Phantom registration studies were performed to examine accuracy and robustness for each framework. In addition, an in vivo registration is performed to demonstrate feasibility of the data acquisition system in the operating room. Results indicate that SurfaceMI performed better in many cases than point-based (PBR) and iterative closest point (ICP) methods for registration of textured point clouds. Mean target registration error (TRE) for simulated deep tissue targets in a phantom were 1.0 +/- 0.2, 2.0 +/- 0.3, and 1.2 +/- 0.3 mm for PBR, ICP, and SurfaceMI, respectively. With regard to in vivo registration, the mean TRE of vessel contour points for each framework was 1.9 +/- 1.0, 0.9 +/- 0.6, and 1.3 +/- 0.5 for PBR, ICP, and SurfaceMI, respectively. The methods discussed in this paper in conjunction with the quantitative data provide impetus for using LRS technology within the model-updated image-guided surgery framework.

Reconstructing images of large high-contrast objects with microwave methods has proved difficult. Successful images have generally been obtained by using a priori information to constrain the image reconstruction to recover the correct electromagnetic property distribution. In these situations, the measured electric field phases as a function of receiver position around the periphery of the imaging field-of-view vary rapidly often undergoing changes of greater than pi radians especially when the object contrast and illumination frequency increase. In this paper, we introduce a modified form of a Maxwell equation model-based image reconstruction algorithm which directly incorporates log-magnitude and phase of the measured electric field data. By doing so, measured phase variation can be unwrapped and distributed over more than one Rieman sheet in the complex plane. Simulation studies and microwave imaging experiments demonstrate that significant image quality enhancements occur with this approach for large high-contrast objects. Simple strategies for visualizing and unwrapping phase values as a function of the transmitter and receiver positions within our microwave imaging array are described. Metrics of the degree of phase variation expressed in terms of the amount and extent of phase wrapping are defined and found to be figures-of-merit which estimate when it is critical to deploy the new image reconstruction approach. In these cases, the new algorithm recovers high-quality images without resorting to the use of a priori information on object contrast and/or size as previously required.

Image-guided neurosurgery relies on accurate registration of the patient, the preoperative image series, and the surgical instruments in the same coordinate space. Recent clinical reports have documented the magnitude of gravity-induced brain deformation in the operating room and suggest these levels of tissue motion may compromise the integrity of such systems. We are investigating a model-based strategy which exploits the wealth of readily-available preoperative information in conjunction with intraoperatively acquired data to construct and drive a three dimensional (3-D) computational model which estimates volumetric displacements in order to update the neuronavigational image set. Using model calculations, the preoperative image database can be deformed to generate a more accurate representation of the surgical focus during an operation. In this paper, we present a preliminary study of four patients that experienced substantial brain deformation from gravity and correlate cortical shift measurements with model predictions. Additionally, we illustrate our image deforming algorithm and demonstrate that preoperative image resolution is maintained. Results over the four cases show that the brain shifted, on average, 5.7 mm in the direction of gravity and that model predictions could reduce this misregistration error to an average of 1.2 mm.