Research‎ > ‎



BWH: Surgicapl Planning Lab

Duke University: Neurosurgery

NeuralNav: Intra-operative ultrasound registration for guiding brain tumor resections
We proposed two new methods for registering pre-operative MRI with intra-operative 3D ultrasound data, during craniotomies for brain tumor resection. These methods will be delivered as part of an extensible NeuralNav toolkit that provides a common API for fetching tracker data and intra-operative images from commercial (VectorVision by BrainLAB) and research (Image-Guided Surgical Toolkit, IGSTK by Georgetown Univ) surgical guidance systems. Development and validation of methods and efforts will be conducted in collaboration with top neurosurgeons, the developers of IGSTK, and BrainLAB.



UNC Biomedical Engineering

Modeling Tumor Micro-environments

New contrast enhanced ultrasound technologies are now enabling detailed images of tissue vascular structure and provide the opportunity to interrogate vascular morphology as an indicator of tumor malignancy based on ultrasound data. Prior work has developed algorithms to glean quantifiable vascular morphology metrics from Magnetic Resonance Imaging data, which have then been shown to be reliable predictors of tumor malignancy and response to therapy in humans. In this project, Kitware will extend this diagnostic capability to the more widely available modality of contrast enhanced ultrasound. Co-PI Dayton has recently demonstrated ultrasonic microvascular mapping using a new type of ultrasound probe which enables the rapid acquisition of 3D high-contrast and high-resolution images of blood vessel structure. If successful, this project will provide a novel and efficient method for clinical ultrasound system manufacturers to implement to assess tumor response to therapy.



UCLA: Laboratory of Neuro Imaging


Traumatic Brain Injury Quantification

Nearly 1.7 million Americans suffer traumatic brain injury (TBI) annually. We propose to develop and validate computational algorithms, based on image segmentation, registration and analysis, which yield quantitative measures to characterize injury, monitor pathology evolution, inform patient prognosis and optimize patient care workflows. This project addresses the current clinical need for informative TBI metrics and the technical need for easy-to-use image analysis tools capable of handling large, heterogenous pathologies that cause severe brain deformations. This project also advances the development of “geometric metamorphosis” which can register images with significant appearance changes caused by structures that grow or contract, such as TBI pathologies. Using geometric metamorphosis, we can derive novel voxel-wise quantifications and visualizations of pathology infiltration and of brain deformations induced by injury or longitudinal brain changes, both within lesions and within GM and WM. These measures will be used to predict outcome and guide clinical decision making. The focus is on final clinical impact and on evaluating the relationship between brain remodeling (e.g. structural changes) with functional recovery or decline.



Liver Ablation Guidance

We aim to develop an operating-room-ready system that provides novel 3D visualizations for needle guidancein soft tissue. The visualizations fuse intra-operative ultrasound (U/S) and pre-operative (pre-op) CTimages by combining two novel technologies: 1) a radically different way of visualizing and interacting with fused U/S-CT images, which we call Spotlight; and 2) an innovative algorithm that automatically and continuously performs U/S-to-CT registration. We will start with open surgical hepatic tumor ablation, but our system’s utility extends well beyond that specific procedure.

The problem: during treatment, surgeons and radiologists must integrate information from several imagemodalities. While CT has excellent diagnostic value, tissues deform intra-operatively due to breathing and surgical manipulation, making pre-op CT a poor representation of intra-op states and positions. On the other hand, intra-op U/S images are real-time, but have a limited field of view and often do not depict pathologies as well as CT. With the current standard of care, the physician alternates between viewing (annotated) pre-op CTand live U/S on separate monitors, with no assistance in correlating features across them.



Multi-scale Spatio-temporal Visualization

In recent years, various terms – the Virtual Physiological Human (VPH), Integrative Biology, Physiome Research – have been used to describe the trend in biomedical research towards the consideration of multi-system processes.

The Multiscale Spatiotemporal Visualisation (MSV) project aims, by international cooperation between the European@neurIST and VPHOP integrated projects, the US National Alliance for Medical Imaging Computing (NA-MIC), and the New Zealand-based IUPS Physiome initiative:
  1. to define an interactive visualisation paradigm for biomedical multiscale data,
  2. to validate it on the large collections produced by the VPH projects, and
  3. to develop a concrete implementation as an open-source extension to the Visualisation Took Kit (VTK), ready to be incorporated by virtually any biomedical modelling software project.



UNC Neuro Image Research
and Analysis Lab (NIRAL)

High-Throughput Small Animal Imaging

While effective methods have been developed for automatically extracting brain morphometry from human MRI scans, few automated quantitative analysis methods exist for small animal MRI.

In this work we propose to develop automatic, reliable, high-throughput MR image analysis methods for small animal, brain morphometry studies. These methods will be made available to investigators via an intuitive web-based interface for collecting, distributing, and automatic processing the imaging data in small animal studies. The web-based data sharing and processing system will also support the inspection of the ongoing processing and the examination of the computed results. This web-based processing system will be generic in nature and could be extended to host and process human MRI data as well as data from other modalities and other applications.

To demonstrate and evaluate the whole system, our collaborators will apply it to several studies of murine and rat brain morphometry currently conducted at UNC. The feedback collected from these studies will directly improve the usability of the proposed system.



COVALIC: Algorithm Validation

Validation remains a difficult task and several tools have emerged to help scientists with validation tasks. The open source Insight Toolkit and Visualization Toolkit provide off the shelf algorithms for medical imaging making comparison with other methods easier. Grand challenges for segmentation and registration, like the ones hosted at the Medical Image Computing and Computer Assisted Intervention, invite researchers to test their algorithms against each other providing a level of validation. However, no complete infrastructure is currently being offer to the research validation for collection and hosting validation tools.

The aim of this proposal is to develop an infrastructure to help scientists to perform validation tasks. While considered an important element towards full clinical validation, the system does not aim to perform a full clinical validation, but rather help research choose the best tools for their clinical application. The proposed system, named COVALIC, provides an online repository of testing and training datasets, an open source framework for validation metrics and an infrastructure for hosting grand challenges and publishing validation results.