Michael I. Miga - US grants

DESCRIPTION (provided by applicant): With respect to the presenting patient population, approximately 69,720 new cases of primary brain tumors are expected to be diagnosed in 2013 with approximately 35% of those tumors being malignant. Brain tumors are the second leading cause of cancer-related deaths in children under age 20, and males between the ages of 20-30. Interestingly, incidence rates are significantly higher in developed countries, approximately 70% higher. Five year survival rates following diagnosis of a primary brain tumor is approximately 33.8% on average but vary considerably with age (e.g. over the age of 65 is 10%). According to the National Cancer Institute surgical removal is the recommended treatment for most brain tumors with the goal of the most complete resection possible while preserving neurological function. With respect to the mission of the National Institute for Neurological Disorders and Stroke, more complete resection reduces the burden of neurological disease by restoring function, extending life, and improving the quality of that life. With respect to surgical therapy, the deployment of visual displays that relate the patient's exposed brain within the operating room (OR) to the pre- operatively acquired neuroanatomical images has become commonplace. More specifically, surgeons can use a pen-like stylus to point at a specific feature on the patient's brain tissue and see where that tissue resides on the neuroanatomical images as facilitated by an interactive display. One detriment to this process is when the patient's brain deforms due to common surgical manipulations. As a result, the alignment between images and the patient's physical brain becomes compromised and surgical error could ensue. Over years 1-7 of this award (initial submission & first renewal), we have demonstrated that intraoperatively acquired cortical surface geometric data can be used to: improve image-to-patient alignment, measure brain deformations during surgery, and drive a computational approach to brain shift correction during image-guided neurosurgery. In this application, we take the final step of extensive intraoperative validation and independent testing. The hypothesis to be tested is that computer models, cortical surface geometric data, and tracked stylus digitization technology when used to compensate for deformation during image-guided brain tumor surgery can predict the locations of eloquent brain surround pathology such that this approach is an effective surrogate to intraoperative magnetic resonance (iMR) imaging. The specific aims to accomplish this are to: (1) integrate a novel computer-vision based approach to surgical field digitization within the surgical microscope and commercial guidance environment such that acquired data can drive a computer model-based approach to correct displays for deformations, (2) conduct a 20 patient retrospective validation study with independent collaborator Dr. Alex Golby of Brigham and Women's hospital comparing the results from our cortical surface deformation driven compensation pipeline to that acquired by the gold standard iMR imaging approach, (3) conduct two prospective 20 patient studies at Vanderbilt and the Brigham respectively, and evaluate intraoperative workflow, as well as the efficacy of corrected displays at institutions wit and without the presence of the iMR alternative. With respect to the importance of this work, in large part, the use of image-guided surgery for surgical resection in soft-tissue organs has been confined primarily to the cranial environment with the common understanding that 'brain shift' is a problem. With the resolution of deformations by this low-cost minimally encumbered solution, the ability to translate image-guided surgery to other soft- tissue organs could become commonplace. Furthermore, the approaches herein are inexpensive when compared to intra-operative imaging methods (e.g. MR), and scalable, i.e. capable of widespread adoption. This application could be seminal in providing the needed surrogate for the wider populations of patient that could benefit from improved 'deformation corrected' guidance but may not have access to iMR facilities.

DESCRIPTION (provided by applicant): To a degree, the use of soft-tissue modeling for updating image-guided navigational systems has not been embraced by the mainstream scientific community. It has only just recently found application within the neuronavigation community (although no commercial systems are available yet) and is still under investigation. Much of this frustration is not due to the growing number of methodologies but rather to a misunderstanding of the goals of model-updating, and an inability to test and validate. With respect to the former, it is na[unreadable]ve to believe that modeling can account for all fine-scale deformations. However, the question to be answered is, within the confines of surgical margin, can model-updating significantly impact surgical resection? This is a central research hypothesis to be investigated within this application. What sets this application apart is that if the milestones are achieved, the outcome could result in a soft-tissue deformation correction system for image- guided liver surgery systems that could be immediately commercially available for patient care. More specifically, at the conclusion of this work, an image-guided liver surgical system capable of deformation correction will be generated, a preliminary experience with the fidelity of those corrections will be established, and the technology will be commercially available to early adopters that secure approval from their Institutional Review Board (IRB). This is possible because this application will leverage an ongoing clinical trial being performed by Pathfinder Therapeutics Incorporated (PTI) that is in the process of testing their image-guided liver surgery (IGLS) system to start in the latter half of 2007. PTI has agreed to share their clinical data to support the novel tissue deformation correction strategy proposed herein. The hypothesis that models can be used to correct for deformation within IGLS will be supported by three specific aims which involve: (1) the development of deformation compensation strategy that involves a combined registration and shape correction technique that will reside on an adaptable deformation correction compute node, (2) the retrospective testing of this approach on data from three separate clinical trials, and (3) an investigation to improve the computer model for the updating process. Phase II for this application would involve upgrading the systems of the early adopters to include our deformation correction compute node and then prospectively test its fidelity clinically. Primary and metastatic cancer within the liver is becoming increasingly common. There is significant evidence that intra-abdominal liver surgery improves survival times for patients afflicted with metastatic disease. Currently the patient population is limited largely due to the complexity of this procedure. Better visualization and guidance would provide surgeons more confidence and would increase the number of surgical candidates and outcome for these patients. If this application is successful, it would lead to the first commercially available image-guided liver surgery system capable of soft-tissue deformation correction. The proposed "deformation correction compute node" would have more widespread impact by being readily adaptable to other surgical systems with similar data. In addition, the strategy would also be compatible with minimally invasive surgeries provided that information regarding organ shape can be acquired. Currently, the only commercial means to correct for soft-tissue deformation is to use intraoperative magnetic resonance (iMR) and computed tomography (iCT). These systems are of considerable expense, require staff, and can be costly to maintain. Due to their cumbersome nature, the patient through-put is also considerably less than a conventional operating room. iCT has been available since the mid-1980's and iMR has been available since the mid- 1990's, yet there are still only a handful of systems being used throughout the world. While these are disadvantages, it should be noted that these systems will not be dispensed with and will continue to be developed. However, it is highly probable that these facilities will become referral centers for the most critical cases rather than available as a mainstream technology. The strategy of augmenting an existing image- guidance system with a "deformation correction compute node" is very low cost, may be as effective as the iMR/iCT solution, and is translatable to any medical center with an image-guided surgery system. This application will play an important role in remedying a disconnection between these sparse referral centers and the vast assortment of local medical centers available to the general population.

This program aims to revitalize undergraduate computing education through the development of a computational science minor targeted to undergraduate majors in science and engineering. These majors represent a broad community of learners for whom computation is an increasingly critical tool. Modern scientific and engineering applications of significant complexity require high-performance computing solutions, and scientists and engineers require computational thinking competencies to achieve such solutions.

The project introduces concurrent, parallel, and distributed computing concepts, techniques, and patterns early in the curriculum. The advent of multi-core processors at the commodity level, necessitated by the efforts to prolong Moore's law, have made understanding these topics a critical learning outcome. The overall effect of this project will thus be to teach computational thinking competencies, modern software design methods, high-performance computing, and scientific computing to a broad community of learners sorely in need of them. By renovating the curriculum with non-computer science majors in mind, computer science majors will also benefit significantly because concurrent, parallel, and distributed computational methods will be infused into the curriculum earlier than they are normally encountered. The computer science curriculum will also be revitalized by introducing real-world examples from science and engineering that have computational interest.

The demographics of science and engineering are different enough from traditional computer science that underrepresented groups in computing will receive significant exposure to core ideas of computational thinking and computing. The diffusion of computational techniques throughout a variety of disciplines will also change the way computational thinking and computation are perceived and taught within the core computer science discipline. A rigorous evaluation plan throughout all phases of the project will measure quantitatively the changes in the preparation of undergraduates for scientific computing by the proposed computational sciences minor, leading to a greater likelihood of being adopted or adapted by other institutions. By improving the computational skills of scientists and engineers, at Vanderbilt and elsewhere, the project will achieve the broader impact of improving science education in the United States, making students far better prepared for the work force and advanced graduate training.

DESCRIPTION (provided by applicant): To a degree, the use of soft-tissue modeling for updating image-guided navigational systems has not been embraced by the mainstream scientific community. It has only just recently found application within the neuronavigation community (although no commercial systems are available yet) and is still under investigation. Much of this frustration is not due to the growing number of methodologies but rather to a misunderstanding of the goals of model-updating, and an inability to test and validate. With respect to the former, it is naove to believe that modeling can account for all fine-scale deformations. However, the question to be answered is, within the confines of surgical margin, can model-updating significantly impact surgical resection? This is a central research hypothesis to be investigated within this application. What sets this application apart is that if the milestones are achieved, the outcome could result in a soft-tissue deformation correction system for image- guided liver surgery systems that could be immediately commercially available for patient care. More specifically, at the conclusion of this work, an image-guided liver surgical system capable of deformation correction will be generated, a preliminary experience with the fidelity of those corrections will be established, and the technology will be commercially available. This is possible because this application will leverage an ongoing relationship with Pathfinder Therapeutics Incorporated (PTI) that is in the process of testing their image-guided liver surgery (IGLS) system and an independent clinical evaluator at Memorial Sloan Kettering Cancer Center. PTI has agreed to share their technological platform with the PD as well as provide open systems and support to support the integration of the novel tissue deformation correction strategy proposed herein. Members of the PTI team have already contributed to the methodologies within this application and have vested scientific interest to continue within the scope of this application with the PD. The hypothesis that models can be used to correct for deformation within IGLS will be supported by three specific aims which involve: (1) incorporate the non-rigid correction compute node controller into the Pathfinder Therapeutics Inc. guidance platform, test, and then deploy to our clinical site, (2) evaluate intraoperative deformation correction using the compute node controller in a clinical trial, and (3) begin to investigate the controller/system within the context of minimally invasive procedures. Previous work involved the development of our registration methods and deformation correction compute node controller. We have succeeded in this endeavor and have shown promise. We are poised to complete the next phase with the deployment into a commercially available IGLS system, and the testing at an independent clinical site in two 25 patient studies as well as address minimally invasive procedures. In addition, it should be noted that while the controller will be integrated into a specific platform, the technology itself is amenable to integration with any image-guided surgery platform.

DESCRIPTION (provided by applicant): The objective of this proposal is to design, construct, and evaluate a steerable needle to debulk intracerebral hemorrhages (ICH) less invasively, from within. This is motivated by the fact that 1 in 50 people will have an ICH in their lifetime, with 40% mortality [50] regardless of whether surgical decompression is attempted. Despite its ability to decompress at-risk tissue, one reason surgery does not produce better overall clinical outcomes may be because of the large volume of healthy brain tissue disrupted simply to access the surgical site. To provide a means of achieving decompression more safely, we propose to create a system capable of debulking the hematoma that results from an ICH without wide exposure, through a needle-sized entry path. Our new steerable needle system will debulk the hematoma from within using a superelastic, precurved aspiration cannula that is deployed within the hematoma through a straight outer needle. The tip of this aspiration cannula is maneuvered within the hematoma by coordinated linear motion and axial rotation of both needle and cannula. The aims of this project involve integrating steerable needle hardware with computed tomography (CT) guidance and use of a deformation model to minimize the number of CT images required, and a demonstration of the system's effectiveness in a porcine model. To achieve these aims, this project brings together a multidisciplinary team combining expertise in neurosurgery (Co-I Weaver), electromechanical design of surgical devices (PI Webster), biomechanical modeling (PI Miga), computer science (Co-I Burgner), surgical research including animal studies (Co-I Williams), and image-guided surgery (all investigators). The endpoint of this R21 project will be a successful demonstration of the complete system in an animal model, which will pave the way for future R01-funded human clinical studies.

? DESCRIPTION (provided by applicant): Over the past several decades, it has been shown that breast lumpectomy is as effective as mastectomy (total or radical) when clear margins are achieved, i.e. no cancer is evident in the margins as encapsulated cancerous breast tissue is resected. Specifically, clear margins have a 5-year recurrence rate of approximately 2-7% (the same as the more radical procedures) while that risk increases to as much as 22% if positive resection margins are present. The difficulty with determining surgical margins intraoperatively, i.e. tumor localization, is that geometric and spatial cues are quickly lost in the OR presentation which differs considerably from the pendant presentation for most diagnostic imaging studies. Generating surgical technologies that could improve the fidelity of resection would have dramatic impact to this considerable population of patients (nearly 230,000 women per year in the United States alone with 80% being detected at stages 0, 1, or 2) especially when considering recurrence rates. The combination of biomechanical computational models of soft tissue deformation with intraoperative guidance and imaging technologies to create interactive displays to help navigate and localize the tumor would be an important step forward in improving the outcomes of lumpectomy. Our hypothesis in this application is that tumor localization is a major confounding factor for improving MR-driven resections as well other complementary disease detection methods (e.g. radioguided probes, optical imaging technologies, etc.). The specific aims of the application are: (1) generate a platform technology with accompanying computational model-enhanced approach to align preoperative MR imaging data to the surgical field for conducting BCS, and (2) acquire feasibility data using two bystander patient studies (n=6 each) measuring the extent and nature of breast deformations and assess initial model-enhanced registration framework.

? DESCRIPTION (provided by applicant): Over the past several decades, dramatic breakthroughs in biomedical science have been witnessed within laboratory research. The ability to translate those discoveries as well as to make new discoveries within human investigations has been a challenge and has been often characterized as the bottleneck of clinical research. Added to this context has been the dramatic changes to healthcare organizational environments, constraints on delivery, efficiency, and reimbursements to include major structural changes to the investment in healthcare research, both federally, and industrially. Lastly, as a result of dramatic systemic changes in healthcare funding structure, the impact on higher educational graduates' careers, specifically doctoral graduates, has been quite profound. Within that changing dynamic landscape, we hypothesize that the fundamental bottlenecks associated with clinical translational research can be dramatically loosened with the training of engineers intimately familiar with human treatment and trained in the inception of novel technology-based platforms. We further hypothesize that continued scientific discoveries within the human environment as well as novel treatment approaches are highly dependent on these technology-based instrumentation platforms. The purpose of this training program is to create a new cadre of researchers capable of creating, developing, implementing, clinically evaluating, and translating methods, devices, algorithms, and systems designed with a clear focus at one particular application of medicine, namely, to facilitate surgical/interventional processes and their outcomes. Thematically, our trainees and training program will have a central focus - innovative platform technologies for treatment and discovery. While this training program addresses pressing problems in biomedical research, namely the translation and facilitation of human investigative systems, the program also speaks to improving higher education career trajectories by providing a novel professional development atmosphere. Briefly described, the training program is a year 2, 3 program with an initial year 1 requiring participation in a novel professional development and a broad therapeutic bioengineering course. With application and acceptance to trainee status, trainees experience an immersion in the clinical environment involving literature reviews, needs assessments, and novel proposals to clinical barriers towards treatment and discovery. Continued course work in areas associated with image-guided procedures, interventional imaging, interventional medical image processing and analysis, robotics and medical device design, modeling & simulation, and interventional therapeutics will provide trainees with almost limitless options for the creation of novel technologies. This all takes place within an environment that is highly promoting of technology transfer as well as scientific discovery. We believe Vanderbilt to be one of the most unique environments that includes incredible close proximity of surgical and interventional suites, robotic and medical device fabrication capabilities, guidance and analysis laboratories, extensive therapeutic investigations, imaging infrastructure, large animal surgical facilities, and a clinical cadre with a history of pioneering invention with our engineers. This is a Training Program for Innovative Engineering Research in Surgery and Intervention.

Project Summary The importance of preserving renal function during surgical interventions through nephron-sparing (?partial nephrectomy?) surgical techniques has garnered considerable and growing recent recognition. Total kidney removal (?radical nephrectomy?) permanently compromises renal function and leads to increased morbidity and mortality, with substantial negative impact on long-term patient quality of life. In contrast, partial nephrectomy improves long-term outcomes by sparing a maximal amount of healthy kidney tissue. However, minimally invasive partial nephrectomy is rarely attempted because it is extremely challenging to accomplish. The surgeon must make critical intraoperative decisions (which can mean life or death for the patient), based on imprecisely mentally inferred and registered three-dimensional anatomical relationships. The surgeon must view a series of 2D preoperative images and build a mental 3D model of patient anatomy. He/she must then intraoperatively guess how this remembered anatomical map should be registered to the patient. The inherent inaccuracies in this process prevent many surgeons from attempting minimally invasive partial nephrectomy, since positive margins (or inadvertent damage to internal kidney structures) are disastrous outcomes, while taking a large negative margin defeats the purpose of the partial nephrectomy procedure. Our overall goal is to create an image-guided surgical system that makes localization, dissection, and isolation of critical vascular and organ structures, as well as correct margin selection, easier and more accurate for the surgeon (and thus the procedure safer and more effective for the patient), thereby increasing the number of surgeons and hospitals able to adopt nephron sparing techniques. Toward this goal, our specific objective in this proposal is to test the hypothesis that image guidance can increase the accuracy and/or time- efficiency of the surgery. To test this hypothesis, we propose three Specific Aims: Aim 1 implements image guidance on the da Vinci surgical robot platform. Aim 2 addresses extensive phantom validation studies using anatomically accurate synthetic organs with realistic material properties. Aim 3 consists of an in vivo human subject pilot study we call a ?bystander study? because it can be achieved with essentially zero risk to human subjects. The endpoint of this R01 will be a fully validated system, and the necessary experimental data to power a large-scale clinical comparative efficacy study. This study will be able to take place soon after the conclusion of this R01, due to the wide availability of the da Vinci robot, and the fact that all a surgeon or hospital must have to participate is the robot itself ? no additional pieces of equipment or infrastructural changes will be necessary.

Project Summary The importance of preserving renal function during surgical interventions through nephron-sparing (?partial nephrectomy?) surgical techniques has garnered considerable and growing recent recognition. Total kidney removal (?radical nephrectomy?) permanently compromises renal function and leads to increased morbidity and mortality, with substantial negative impact on long-term patient quality of life. In contrast, partial nephrectomy improves long-term outcomes by sparing a maximal amount of healthy kidney tissue. However, minimally invasive partial nephrectomy is rarely attempted because it is extremely challenging to accomplish. The surgeon must make critical intraoperative decisions (which can mean life or death for the patient), based on imprecisely mentally inferred and registered three-dimensional anatomical relationships. The surgeon must view a series of 2D preoperative images and build a mental 3D model of patient anatomy. He/she must then intraoperatively guess how this remembered anatomical map should be registered to the patient. The inherent inaccuracies in this process prevent many surgeons from attempting minimally invasive partial nephrectomy, since positive margins (or inadvertent damage to internal kidney structures) are disastrous outcomes, while taking a large negative margin defeats the purpose of the partial nephrectomy procedure. Our overall goal is to create an image-guided surgical system that makes localization, dissection, and isolation of critical vascular and organ structures, as well as correct margin selection, easier and more accurate for the surgeon (and thus the procedure safer and more effective for the patient), thereby increasing the number of surgeons and hospitals able to adopt nephron sparing techniques. Toward this goal, our specific objective in this proposal is to test the hypothesis that image guidance can increase the accuracy and/or time- efficiency of the surgery. To test this hypothesis, we propose three Specific Aims: Aim 1 implements image guidance on the da Vinci surgical robot platform. Aim 2 addresses extensive phantom validation studies using anatomically accurate synthetic organs with realistic material properties. Aim 3 consists of an in vivo human subject pilot study we call a ?bystander study? because it can be achieved with essentially zero risk to human subjects. The endpoint of this R01 will be a fully validated system, and the necessary experimental data to power a large-scale clinical comparative efficacy study. This study will be able to take place soon after the conclusion of this R01, due to the wide availability of the da Vinci robot, and the fact that all a surgeon or hospital must have to participate is the robot itself ? no additional pieces of equipment or infrastructural changes will be necessary.

Summary The scientific premise underpinning this application is that technologies relating the precise physical location of therapy and co-located spatially-encoded disease information, i.e. usually provided by imaging, will enable safer less invasive liver surgical procedures thus increasing surgical candidacy and allowing for more accurate interventions. Going further, this improvement in localization will facilitate the analysis of outcomes for evaluating efficacy, the comparison of approaches, the rating of improvements, and the impact of imaging biomarkers to drive therapy decisions. In this application, while there is potentially broad clinical impact, the specific focus is facilitating and quantifying the spatial accuracy and fidelity of localized image-based information to enhance navigational assistance during laparoscopic liver surgery. Realization would be a fundamental advance in the domain of procedural medicine. The application hypothesis is that biomechanical model-based registration coupled to non-contact digitization can account for soft-tissue deformations and localize intraparenchymal liver targets to an unprecedented error of less than 3 mm of error on average. In addition, we will begin to explore the impact of this technology on locoregional therapies. What also sets this application apart is the strong translational trajectory of the technology through our proposed integration with the on FDA cleared liver guidance system. Another distinction for the application is that we have employed an assessment framework enlisting one of the best hepatopancreatobiliary services in the world at Memorial Sloan Kettering Cancer Center under the supervision of Dr. William R. Jarnagin for the independent evaluation of the technology. This decade+ long collaborative relationship among the co-PIs has been extremely productive (see Multi-PI plan) and is among the most scientifically substantive in the field of image guided liver navigation. The specific aims of the application involve: (1) the integration of deformation correction into a commercial guidance system, (2) the investigation of a novel approach to the laparoscopic indication which includes advances in a non-contact digitization and novel enhancements for modeling insufflation, ligament support, and model heterogeneity, (3) three clinical bystander studies that quantitatively characterize improvements in digitization, and the laparoscopic approach, and (4) an exploratory study that initiates the investigation of the use of the platform for locoregional therapies. In summary, while the focus of the application is a fundamental advance in liver cancer surgical care, it also offers a paradigm shift by providing a novel investigational framework in human systems within the context of procedural medicine.