Zeike A. Taylor

Biomechanical modelling of soft tissue provides a method for constraining medical image registration, such that the estimated spatial transformation is considered biophysically plausible. Existing methods either directly optimize the loss function containing the biomechanical-constrained regularization term over deformations, which takes much computational time, or are trained using biomechanically plausible data generated via finite element simulation, which is cumbersome. This work first instantiates the recently-proposed physics-informed neural networks (PINNs) to 3D elastic models that are used to establish the partial differential equations (PDEs) representing physics laws of biomechanical constraints to be satisfied. The registration algorithm that aligns point sets considering PINN-imposed biomechanics (i.e., the forward problem) is then formulated. In addition, the inverse problem and its algorithm of physical parameter (i.e., material property) estimation along with the registration are also formulated and developed. We carefully compare linear and nonlinear elasticity theories’ capabilities in solving both tasks of forward registration and inverse physical parameter identification under PINNs respectively. Furthermore, two specific network configurations that leverage one common branch or two individual branches to predict deformation vectors and biomechanical states are also constructed and compared. The proposed PINNs-based registration approaches have been extensively evaluated with three experiments, that is single and multiple patient MRI-US registration using clinical MRI-US pairs, and registration using pairs of undeformed MR images from clinical cases of prostate cancer biopsy and deformed counterparts with finite-element-computed ground-truth deformation. Results demonstrate that the proposed methods achieve state-of-the-art performances compared to biomechanical-model-based and learning-based registration approaches, and the biomechanical constraints of soft tissues have been successfully warranted after registration. The codes are available at https://github.com/ZheMin-1992/Registration_PINNs.

Thrombosis is a biological response closely related to intracranial aneurysms, and the formation of thrombi inside the aneurysm is an important determinant of outcome after endovascular therapy. As the regulation of thrombosis is immensely complicated and the mechanisms governing thrombus formation are not fully understood, mathematical and computational modeling has been increasingly used to gain insight into thrombosis over the last 30 years. To have a robust computational thrombosis model for possible clinical use in the future, it is essential to assess the model's reliability through comprehensive sensitivity analysis of model parameters and validation studies based on clinical information of real patients. Here, we conduct a global sensitivity analysis on a previously developed thrombosis model, utilizing thrombus composition, the flow-induced platelet index, and the bound platelet concentration as output metrics. These metrics are selected for their relevance to thrombus stability. The flow-induced platelet index quantifies the effect of blood flow on the transport of platelets to and from the site of thrombus formation and thus on the final platelet content of the formed thrombus. The sensitivity analysis of the thrombus composition indicates that the concentration of resting platelets most influences the final thrombus composition. Then, for the first time, we validate the thrombosis model based on a real patient case using patient-specific resting platelet concentration and two previously calibrated trigger thresholds for thrombosis initiation. We show that our thrombosis model is capable of predicting thrombus formation both before and after endovascular treatment.

Background: The posterior communicating artery (PComA) is among the most common intracranial aneurysm locations, but flow diverter (FD) treatment with the widely used pipeline embolization device (PED) remains an off-label treatment that is not well understood. PComA aneurysm flow diversion is complicated by the presence of fetal posterior circulation (FPC), which has an estimated prevalence of 4–29% and is more common in people of black (11.5%) than white (4.9%) race. We present the FD-PComA in-silico trial (IST) into FD treatment performance in PComA aneurysms. ISTs use computational modeling and simulation in cohorts of virtual patients to evaluate medical device performance. Methods: We modeled FD treatment in 118 virtual patients with 59 distinct PComA aneurysm anatomies, using computational fluid dynamics to assess post-treatment outcome. Boundary conditions were prescribed to model the effects of non-fetal and FPC, allowing for comparison between these subgroups. Results: FD-PComA predicted reduced treatment success in FPC patients, with an average aneurysm space and time-averaged velocity reduction of 67.8% for non-fetal patients and 46.5% for fetal patients (P < 0.001). Space and time-averaged wall shear stress on the device surface was 29.2 Pa averaged across fetal patients and 23.5 Pa across non-fetal (P < 0.05) patients, suggesting FD endothelialization may be hindered in FPC patients. Morphological variables, such as the size and shape of the aneurysm and PComA size, did not affect the treatment outcome. Conclusions: FD-PComA had significantly lower treatment success rates in PComA aneurysm patients with FPC. We suggest that FPC patients should be treated with an alternative to single PED flow diversion.

Minimally invasive surgery (MIS) remains technically demanding due to the difficulty of tracking hidden critical structures within the moving anatomy of the patient. In this study, we propose a soft tissue deformation tracking augmented reality (AR) navigation pipeline for laparoscopic surgery of the kidneys. The proposed navigation pipeline addresses two main sub-problems: the initial registration and deformation tracking. Our method utilizes preoperative MR or CT data and binocular laparoscopes without any additional interventional hardware. The initial registration is resolved through a probabilistic rigid registration algorithm and elastic compensation based on dense point cloud reconstruction. For deformation tracking, the sparse feature point displacement vector field continuously provides temporal boundary conditions for the biomechanical model. To enhance the accuracy of the displacement vector field, a novel feature points selection strategy based on deep learning is proposed. Moreover, an ex-vivo experimental method for internal structures error assessment is presented. The ex-vivo experiments indicate an external surface reprojection error of 4.07 ± 2.17mm and a maximum mean absolutely error for internal structures of 2.98mm. In-vivo experiments indicate mean absolutely error of 3.28 ± 0.40mm and 1.90±0.24mm, respectively. The combined qualitative and quantitative findings indicated the potential of our AR-assisted navigation system in improving the clinical application of laparoscopic kidney surgery.

How prevalent is spontaneous thrombosis in a population containing all sizes of intracranial aneurysms? How can we calibrate computational models of thrombosis based on published data? How does spontaneous thrombosis differ in normo- and hypertensive subjects? We address the first question through a thorough analysis of published datasets that provide spontaneous thrombosis rates across different aneurysm characteristics. This analysis provides data for a subgroup of the general population of aneurysms, namely, those of large and giant size (>10 mm). Based on these observed spontaneous thrombosis rates, our computational modeling platform enables the first in silico observational study of spontaneous thrombosis prevalence across a broader set of aneurysm phenotypes. We generate 109 virtual patients and use a novel approach to calibrate two trigger thresholds: residence time and shear rate, thus addressing the second question. We then address the third question by utilizing this calibrated model to provide new insight into the effects of hypertension on spontaneous thrombosis. We demonstrate how a mechanistic thrombosis model calibrated on an intracranial aneurysm cohort can help estimate spontaneous thrombosis prevalence in a broader aneurysm population. This study is enabled through a fully automatic multi-scale modeling pipeline. We use the clinical spontaneous thrombosis data as an indirect population-level validation of a complex computational modeling framework. Furthermore, our framework allows exploration of the influence of hypertension in spontaneous thrombosis. This lays the foundation for in silico clinical trials of cerebrovascular devices in high-risk populations, e.g., assessing the performance of flow diverters in aneurysms for hypertensive patients.

Biomechanical modelling of soft tissue provides a non-data-driven method for constraining medical image registration, such that the estimated spatial transformation is considered biophysically plausible. This has not only been adopted in real-world clinical applications, such as the MR-to-ultrasound registration for prostate intervention of interest in this work, but also provides an explainable means of understanding the organ motion and spatial correspondence establishment. This work instantiates the recently-proposed physics-informed neural networks (PINNs) to a 3D linear elastic model for modelling prostate motion commonly encountered during transrectal ultrasound guided procedures. To overcome a widely-recognised challenge in generalising PINNs to different subjects, we propose to use PointNet as the nodal-permutation-invariant feature extractor, together with a registration algorithm that aligns point sets and simultaneously takes into account the PINN-imposed biomechanics. The proposed method has been both developed and validated in both patient-specific and multi-patient manner.

How prevalent is spontaneous thrombosis (ST) in intracranial aneurysms (IAs) for an all-size pop- ulation? How can we calibrate computational models of thrombosis based on published data from size-specific aneurysms cohorts? How does ST differ in normo- and hypertensive subjects? We ad- dress the first question by a thorough analysis of published that provide ST rates across different patient demographics and aneurysm characteristics. This analysis provides data for a subgroup of the gen- eral population, viz. large and giant aneurysms (>10 mm). Based on these observed ST rates, our novel computational modelling platform enables the first in-silico observational study of ST prevalence across a broader set of IA phenotypes. We generate 109 virtual patients and use a novel approach to calibrate two trigger thresholds: residence time (RT) and shear rate (SR), thus addressing the second question. We then address the third question by utilising this calibrated thrombosis model to provide new insights on the effects of hypertension on ST. We demonstrate how a mechanistic ST model calibrated on a reduced IA cohort can help esti- mate ST prevalence in a broader IA population. This study was enabled through a comprehensive and fully automatic multi-scale modeling pipeline. We use an emerging property, viz. ST, as an indirect population-level validation of a complex computational modeling framework. Furthermore, our frame- work allows exploration of the influence of hypertension in ST. This lays the foundation for in-silico clinical trials of cerebrovascular devices in high-risk populations, e.g. assessing the performance of flow diverters in cerebral aneurysms for hypertensive patients.