Grants (PI or Lead at Kitware)

Overcoming commercial roadblocks in CT perfusion research via high-quality open-source software

NIH Administrative Supplement (PI: Y. Lee UNC; Kitware Lead S. Aylward)

9/2022 – 10/2024                     $99,410

Release open-source software for computing CT perfusion measures used in stroke assessment.


Anatomic Reconstruction for Multi-Task POCUS Automated Interpretation

DARPA              HR00112190077 (PI: S. Aylward)

04/2021 – 2/2023                     $999,999

The objective of this program is to deliver an innovative POCUS AI framework based on anatomic reconstruction that is optimized for learning from limited POCUS-task-specific training data (<15 cases) and that can run in <5 sec. on mobile hardware.


In-Field Detection of Acute Subdural Hematomas Requiring Urgent, Life-Saving Treatment in Severe TBI Patients

CDMRP / USAMRMC / MTEC   MTEC-19-08-MuLTI-0079: MT19008.79 (PI: S. Aylward)

10/2019 – 11/2023                       $1,517,000

We propose to develop the software component of an easy-to-use, automated, non-invasive, light-weight and portable system that can be used in-field to determine if urgent, life-saving treatment is needed to address severely elevated intra-cranial pressure (ICP) associated with traumatic brain injuries. The system estimates ICP using automatic measurements of the optic nerve sheath diameter in ultrasound images. We also propose to compare the performance of our ICP estimation algorithm when it is applied to images from five different commercially available point-of-care ultrasound devices, to help assess imaging system design tradeoffs for machine learning applications.


Stationary Digital Tomosynthesis for Transbronchial Biopsy Guidance

NIH NIBIB        R01EB028283 (PI: Y. Lee, UNC; Kitware Lead: S. Aylward)

9/2019 – 5/2024                           $711,000

We propose to develop a novel intraoperative chest tomosynthesis imaging (ICTI) system to provide real-time, low-dose, and high-resolution 3D image guidance during thoracic medical procedures. The focus of the present work is to develop the instrumentation and to validate the technology for lung nodule localization during guided bronchoscopy. The proposed ICTI system utilizes the novel distributed x-ray source array invented by our team to perform rapid and high-resolution tomosynthesis imaging. The system will be integrated with image reconstruction and display software to provide up-to-date information on the location of the bronchoscope with respect to the lesion of interest.


Automated Assessment of Leptomeningeal Collaterals on CT Angiograms

NIH NINDS      R41NS086295 (PI: S. Aylward, Y. Lee UNC)

6/2015 – 9/2020                           $977,000

The primary goal of this study is to develop an automated scoring system of leptomeningeal collaterals based on CT angiography (CTA) as a quantitative assessment of acute stroke. Leptomeningeal collaterals provide an alternate path of blood flow in the setting of a proximal occlusion. The presence and effectiveness of leptomeningeal collaterals (i.e., collateral status) will vary significantly from patient to patient. Collateral status has been shown to be correlated with both patient outcome and risk of hemorrhagic transformation, suggesting that rapid and accurate assessment of the collaterals would provide a powerful tool for treatment evaluation.


High-Frame Rate 3-D Super-Resolution Ultrasound Microvascular Imaging: An Academic-Industrial Partnership

NIH NCI            R01CA220681 (P. Dayton, UNC; Kitware Lead: S. Aylward)

7/2020 – 7/2022                       $190,000

Ultrasound has tremendous potential in the oncology clinic because of low cost, portability, and safety. However, it suffers from low specificity to malignancy and limited sensitivity to small lesions. In this project, we pair with an industrial partner to develop the next-generation of ultrasound for oncological imaging. Our proposed approach builds on recent discoveries with innovative technological advancement to develop an ultrasound imaging technique which can detect cancer’s fingerprint through detection of malignant microvascular angiogenesis. To do this, we will exploit advances in computational hardware and data processing to achieve super-resolution images of micro-vessels in-vivo, at resolutions an order of magnitude better than possible with commercial systems, at clinically relevant depths, with clinically translatable acoustic parameters, and with FDA approved contrast agents.


Accelerating Community-Driven Medical Innovation with VTK

NIH NIBIB        R01EB014955 (PI: S. Aylward, K. Martin)

5/2019 – 4/2023                       $2,200,000

The Visualization Toolkit (VTK) is an open source, freely available software library for the interactive display and processing of medical images. It is being used in most major medical imaging research applications, e.g., 3D Slicer and Osirix, and in several commercial medical applications, e.g., BrainLAB’s VectorVision surgical guidance system. VTK development began in 1993 and since then an extensive community of users and developers has grown around it. However, the rapid advancement of cloud computing, GPU hardware, deep learning algorithms, and VR/AR systems require corresponding advances in VTK so that the research and products that depend on VTK continue to deliver leading edge healthcare technologies. With the proposed updates, not only will existing applications continue to provide advanced healthcare, but new, innovative medical applications will also be inspired.


Automated Assessment of Leptomeningeal Collaterals on CT Angiograms

NIH NINDS      R42NS086295 (PI: S. Aylward, Y. Lee, UNC, and M. Niethammer, UNC)

9/2018 – 9/2020                       ~$1,000,000

Leptomeningeal collateral vessels provide connections between vascular territories in the brain when occlusions occur, e.g., when a person suffers a stroke. The number and connectivity of collateral vessels varies significantly from patient to patient, and they have been shown to play an important role in stroke outcomes, providing blood flow to tissue at risk. Rapid and effective patient-specific determination of the presence and connectivity of collaterals is needed in acute stroke situations to determine which treatments will be effective. We propose to develop an automated system to evaluate collateral vessel trees from computed tomography angiography (CTA) images and to offer that system as an algorithms-as-a-service product.


Slicer+PLUS: Collaborative, open-source software for ultrasound analysis

NIH NIBIB        1R01EB021396   (PI: S. Aylward, G. Fitchinger, Queens University, Canada)

7/2016 – 6/2020                       $2,400,000

Integrate ITK, Slicer, and PLUS as open-source toolkits to define a platform for low-cost ultrasound application research and development.


Web-Based Infrastructure for Comparison and Validation of Image Computing Methods

NIH NIBIB        R41EB011796 (PI: S. Aylward, G. Gerig, U of Utah)

4/2015 – 4/2018                       $977,000

We propose to develop the infrastructure for and deploy a commercial installation of an Algorithm Evaluation Service (AES) that will (1) assist commercial company in determining which medical data analysis algorithms they should integrate into their products, and (2) provide algorithm researchers with better access to clinically relevant data and with a better understanding of clinical and commercial needs.


Image-based Quantification and Analysis of Longitudinal Lung Nodule Detection (PI: Cahill, RPI)

NIH NIBIB        1R41EB015775-01 (PI: N. Cahill RIT; Kitware Lead: S. Aylward)

4/2013 – 4/2015                       $64,000

Tracking morphological changes in ground-glass nodules for early detection of cancer and assessment of response to treatment.


In-field FAST Procedure Support and Automation

NIH/NIBIB       1R43EB016621-01 (PI: S. Aylward)

4/2013 – 4/2015                       $397,000

Assist first-responders and those with limited medical experience in the detection of blood in the abdomen using a portable ultrasound system, after blunt abdominal trauma.


Accelerating Community-driven Medical Innovation with VTK

NIH NIBIB        1R01EB014955-01A1 (PI: S. Aylward, B. Geveci, W. Schroeder)

4/2013 – 4/2018                       $2,000,000

Update Kitware’s Visualization Toolkit (VTK) to serve as a platform for pervasive computation and visualization: from phone, to tablet, to desktop, to web server, to cloud, and to super-computer.


Multimodality Image-based Assessment System for Traumatic Brain Injury

NIH/NINDS     1R41NS081792-01 (PI: S. Aylward, M. Niethammer UNC, J. van Horn USC)

4/2013 – 4/2014                       $159,000

Longitudinal analysis of traumatic brain injury MRI scans for the determination of patient response to therapy and predict outcome.


Micro-tumor detection by quantifying tumor-induced vascular abnormalities

NIH NCI            1R01CA170665-01 (PI: P. Dayton, UNC; Kitware Lead: S. Aylward)

9/2012 - 8/2017                        ~$150,000

Data show that substantial changes in microvasculature structure occur after the arrival of only 10s to 100s of tumor cells, that these changes extend to vessels that are relatively large (hundreds of microns in diameter), and that microvascular changes extend well beyond tumor margins, even soon after the onset of disease. These unique microvascular “cancer signatures” provide us with a means to overcome traditional resolution limitations which otherwise impair micro-tumor detection.


Quantitative ultrasound analysis of vascular morphology for cancer assessment

NIH/NCI           1R43CA165621-01 (PI: S. Aylward and P. Dayton, UNC)

7/2012 – 6/2014                       $150,000

Develop vessel analysis methods for a novel dual-frequency ultrasound system that uses a micro-bubble contrast agent. Will be used to stage tumors and assess response to treatment.


A Needle Guidance System for Hepatic Tumor Ablation That Fuses Real-Time Ultrasound

NIH NCI            1R44CA143234 (PI: S. Razzaque, InnerOptic; Kitware Lead: S. Aylward)

4/2012 – 3/2015                       $416,000         

Register intra-operative ultrasound with pre-operative CT and display using augmented reality, for liver lesion RFA.


The National Alliance for Medical Image Computing (NA-MIC)

NIH NIBIB        U54 EB005149 (PI: R. Kikinis, BWH.  Software Engineering Core PI: S. Aylward)

10/2010 – 9/2015                     $2,220,000

NA-MIC was created to implement a robust and flexible open-source infrastructure for developing and applying advanced imaging technologies across a range of important biomedical research disciplines.


Image Registration for Ultrasound-Based Neurosurgical Navigation

NIH/NCI           1R01CA138419-01A1 (PI: S. Aylward and W. Wells, BWH)

7/2009 - 6/2014                            $2,200,000

Register pre-operative MRI with intra-operative ultrasound to guide neurosurgical de-bulking of tumors.


Multiscale Spatio-temporal Visualization

EU                  FP7 Consortium Funding (D. Testi, SCS, Italy; Kitware Lead: S. Aylward)

1/2010 - 12/2012                      $312,000

Develop consensus on the requirements and specifications for a next-generation toolkit for managing, processing, and visualizing data that spans scale and space at multiple, vastly different scales. Participants include Super Computing Solutions, University of Bedfordshire, Universitat Pompeu Fabra, The University of Auckland, and Kitware.


 High Throughput web-base Image Analysis of Mouse Brain MR Imaging Studies

NIH NINDS      R42 NS059095-03 (M. Styner, UNC; Kitware Lead: S. Aylward)

7/2009 - 6/2012                            $1,100,000

Process multiple MRI protocols for brain morphology quantification, over the web, for the study of mice.


ITK Maintenance, and Development of ITKv4

NLM              Contract (PI: L. Ibanez and S. Aylward)

6/2009 - 10/2011                      $1,100,000

Maintain the infrastructure that supports ITK and the ITK community. Promote and support the adoption and use of ITK. Support the refactoring and extension of ITK for ITKv4.


EXPOSE: TRUST in Integrated Circuits

DARPA (M. Bajura, USC, Kitware Lead: S. Aylward)

12/2009 - 10/2011                   $541,000

Analysis of x-ray images of integrated circuits to detective deviations from specification sent to manufacturer.


Micro-to-Macro Multimodality Atlas Formation

NIH NIBIB        1R41EB004737-01 (PI: S. Aylward)

10/04-09/05                            $100,000

Reconstruct confocal microscopy images of consecutive vibratome sections into 3D, linking anatomical and functional details at multiple scales within and across subjects.  Provides a software application that enables researchers with limited software and image processing expertise to create and interactively visualize multi-scale, multimodality, image-based atlases.


Medical Image Segmentation (Phase I and II awards)

Airforce Research Labs   SBIR (PI: S. Aylward)

2/2006 - 2/2009                        $1,177,000      

Develop applications for medical image segmentation, with an emphasis on workflow within and between the applications.


Neural Networks and Clustering

NLM  N01 467-MZ-402068 (PI: S. Aylward)

10/04-09/06                            $100,000

Implement neural network method in the NLM’s Insight Toolkit for medical image segmentation and registration.


BatchMake and Validation Dashboards

NLM  N01 467-MZ-402071 NLM (PI: S. Aylward)

10/04-09/06                            $100,000

Provide a simple scripting language for conducting method validation studies.   Reports generated as HTML pages.   Scripts can be executed on individual machines or distributed to the GRID.


DICOM and Digital Libraries for Data Archiving

NLM  N01 467-MZ-402070 NLM (PI: S. Aylward)

10/04-09/06                            $100,000

Add DICOM query and retrieve capabilities to the NLM’s Insight Toolkit by interfacing it with DCMTK.


Insight Software Consortium

NLM  N01 467-MZ-402073 (PI: S. Aylward)

10/04-09/06                            $100,000

Provide HTML pages for the Insight Software Consortium, host a conference on open-source software, and provide tutorials on ITK.


3D Cerebral Vessel Location for Surgical Planning

NIH NCI            R01EB000219 NIH-NCI (PI: E. Bullitt, UNC; Kitware Lead: S. Aylward)

4/2005 – 3/2008                       $2,300,000

We propose to build upon earlier work by providing a potentially clinically useful, symbolic description of blood vessels supplying vascular CNS tumors, includes both solid tumors and arteriovenous malformations.


Vessel Visualization and Measurement for Partial Organ Transplant Planning and Evaluation

Whitaker Foundation Young Investigator Award (PI: S. Aylward)

10/2002 - 9/2005                      $500,000

The specific goal of this project is to develop and evaluate a software package that facilitates three partial-liver transplant image analysis tasks: (1) Understanding the donor’s and the recipient’s anatomy and vascular network for surgical path planning. (2) Specifying a surgical path and predicting both donor and recipient outcome based on the surgical path complexity, graft volume, and the path’s effect on the liver’s vascular network. (3) Assessing liver regeneration in the donor and the recipient by quantifying volume and vessel network change by registering the donor’s pre-operative liver vessel network with the donor’s and the recipient’s post-operative liver vascular networks six weeks following the operation.


3D Image-Guided Endovascular Surgery

NIH HLBI          R01HL69808 (PI: E. Bullitt, UNC; Kitware Lead: S. Aylward)

4/2002 - 3/2008                        $2,500,000

The goal of this project is to integrate state-of-the-art visualization methods with novel vessel representation methods and novel methods for reconstructing objects into 3D from bi-plane fluoroscopy for the guidance of TIPS procedures.


Mixed reality for planning and practicing liver lesion radiofrequency ablations

Microsoft Research     (Recipient: S. Aylward)

4/2002                                      $56,000

Five advanced graphics workstations and a haptic (force-feedback robotic arm) have been donated by Microsoft to the CADDLab to support the development of a virtual surgery simulation environment for radiofrequency ablation training.


3D vascular visualization for surgical planning and guidance

Dell Computer Corporation      Equipment Gift (Recipient S. Aylward)

4/2002                                      $10,000

Two high-end workstations have been donated by Dell Corporation to support the 3D vessel visualization work of the CADDLab. These systems will be used to complete the development and delivery of the living donor liver transplant planning software to clinicians.


ITK: Segmentation and Registration Toolkit for the Visible Human Data

NIH/NLM         Contract (PI: S. Aylward)

9/1999 - 9/2003                        $1,230,000

With six other institutions, we are working to define the standard software library for medical image processing algorithm implementation and distribution. Software library will be distributed with the visible human data.


Vessel Visualization for Living-Related Donor Partial Liver Transplant Planning and Evaluation

Microsoft         Equipment Gift (Recipient: S. Aylward)

10/1999                                    $15,000

Three high-end computer graphics workstations were donated to facilitate the development of software and a haptic system for training and surgical simulation for partial liver transplantation.


Phase II: Predicting Time to Brain Tumor Recurrence using Neural Networks and 1H-MRS

Holderness Foundation: Medical Student Research Preceptor Award

Med. Student: Gerstle   (Preceptor: S. Aylward)

10/98-12/98                            $6,000

This project is focusing on the role of neural networks in improving diagnosis via proton magnetic resonance spectroscopy for determining the patient specific effectiveness of brain tumor treatments.


Phase I: Predicting Time to Brain Tumor Recurrence using Neural Networks and 1H-MRS

Lineberger Comprehensive Cancer Research Center, UNC

Innovative Cancer Research Award (PI: S. Aylward)

08/98-10/98                            $3,000

This project is focusing on the role of neural networks in improving diagnosis via proton magnetic resonance spectroscopy for determining the patient specific effectiveness of brain tumor treatments.