In this article, we will present a novel interactive image registration technique using the 3D volumetric view as guidance throughout the registration process. The registration criterion used in this technique is the homogeneity of the color distribution on any given anatomic landmark, usually the skin/muscle voxels. This registration criterion has been validated using two different approaches, including comparison with automatic registration using mutual information based on the Insight Toolkits (ITK) and comparison with registered MR images with different pulse sequences acquired directly from the same MR scanner less than 5 minutes. An investigation is also made to compare this technique with the conventional interactive image registration using three orthogonal planar (3P) views, which is predominantly used in the current treatment planning in clinic. This 3D technique has shown several advantages, including improved registration accuracy and accelerated registration speed. It has also found that the 3P pixel registration technique often "inherits" a global positioning error, due to its partial 3D visual presentation. In addition, unlike the 3P technique, which generally cannot be used to register more than 2 images simultaneously and low resolution PET images, this 3D technique can register up to 4 imaging modalities simultaneously, including PET images, with both global view guidance and detail checking capability.

3D Voxel-Based Volumetric Image Registration with Volume-View Guidance

This paper focuses on the important role of open-source software in the development of a fully-automatic and accurate registration system for images of Quadrature Tomographic Microscopy. The system is correspondence-based, using the framework provided by open-source RGRL(Rensselear Generalized Registration Library). The library implements several well-known techniques and the ability to mix-and-match techniques for components of the framework allows testing of ideas and
development of the robust system in a timely fashion.

The Role of Open-Source Software in the Study of Embryogenesis

A novel fiber tracking algorithm (GTRACT) was developed to enhance tracking through ambiguous regions where cross fibers, fiber merging or fanning may be occurring. The software was developed using several cross platform open-source toolkits (ITK, VTK, and FLTK). The algorithm was evaluated using a freely available digital phantom dataset provided by King's College London. The results show that the GTRACT algorithm performed significantly better than standard streamline approaches and is less affected by noise.

Guided Diffusion Tensor Tractography with GTRACT: A Validation Study

The usage of image processing has become an important and continuous growing field in medical imaging. Although a large amount of algorithms is available, only few are routinely applied within applications for clinical diagnosis support and treatment planning. New tools are required that allow for efficient evaluation and comparison of algorithms as well as the 
generation of dedicated user interfaces with sophisticated visualization for clinical applications. We propose a generic integration of ITK algorithms into the visual programming and rapid prototyping platform MeVisLab. One of the main limitation of ITK is its lack of visualization and interaction functionality through a dedicated development platform. MeVisLab provides a modular visual programming interface with a comprehensive suite of image processing and visualization tools. The linkage of ITK algorithms to MeVisLab is based on a two step process that consists of a generic XML description for each ITK algorithm and an automatic code generator to create new MeVisLab modules, which will be made publicly available within the free version of MeVisLab. Each module can handle different image types and data dimensions dynamically at runtime. Example applications using registration, segmentation, and visualization modules are presented in order to show the capabilities of the described integration concept.

A Framework for Algorithm Evaluation and Clinical Application Prototyping using ITK

The Medical Imaging Interaction Toolkit (MITK) is an opensource toolkit for the development of interactive medical image analysis software. MITK is based on the open-source Insight Toolkit (ITK) and Visualization Toolkit (VTK) and extends them with features required for interactive systems. ITK is used for the algorithmic scope and general infrastructure, VTK for visualization. Key features of MITK are the coordination of multiple 2D and 3D visualizations of arbitrary data, a general interaction concept including undo/redo, and its extendibility and flexibility to create tailored applications due to its toolkit character and
different layers of hidden complexity. The paper gives a brief introduction into the overall concepts and goals of the MITK approach. Suggestions and participation are welcome. MITK is available at www.mitk.org.

The MITK Approach

The Allen Brain Atlas (ABA) project aims to create a
cellular-resolution, genome-wide map of gene expression in the adult
mouse brain. The resulting in situ hybridization (ISH) image data
will be available free-of-charge to the public. Additionally, we are
developing an informatics pipeline to support searching of the data by
anatomic region and expression level and/or pattern. This paper
describes a robust, high-throughput registration scheme to
automatically annotate hierarchical brain structures in the ISH
imagery.

Automated high-throughput registration for localizing 3D mouse brain gene expression using ITK

Active contour segmentation and its robust implementation using level sets
have been studied thoroughly in the medical image analysis literature. Despite
the availability of these powerful methods, clinical research still largely
relies on manual slice-by-slice outlining for anatomical structure
segmentation. To bridge the gap between methodological advances and clinical
routine, we developed ITK-SNAP: an open source application intended to make
level set segmentation easily accessible to a wide range of users with various
levels of mathematical expertise. We briefly describe this new tool and report
the results of a validation study in which ITK-SNAP was compared to manual
segmentation of the caudate in the context of an ongoing child neuroimaging
autism study.

User-Guided Level Set Segmentation of Anatomical Structures with ITK-SNAP

A modification is made to Otsu thresholding method. It provides better results for ICV (Intracranial volume) segmentation on one kind of T2-weighted MRI data.

A Modification to Otsu Thresholding Method for ICV Segmentation

Methods for segmentation of skull infiltrated tumors in Computed
Tomography (CT) images using Insight Segmentation and Registration
Toolkit ITK (www.itk.org) are presented. Pipelines of filters and
algorithms from ITK are validated on the basis of different criteria: sensitivity,
specificity, dice similarity coefficient, Chi-squared, and Hausdorff
distance measure. The method to rate segmentation results in relation
to validation metrics is presented together with analysis of importance of
different goodness measures. Results for one simulated dataset and three
patient are presented.

Segmentation of Skull-infiltrated Tumors Using ITK: Methods and Validation

For over two decades, the National Library of Medicine (NLM)
has provided support for the collection of biomedical image data for use
throughout the biomedical image, visualization and analysis community.
Data collected during the Visible Human Project has been utilized to
advance development of medical image research, education, and other
ventures. One goal of our research has been to further diversify such
image data and make it openly available to the biomedical imaging community.
The approach is to provide open access to a diverse collection
of biomedical image data that can be used for the development and validation
of new image processing and analysis techniques. With support
from NLM, over 100 datasets were incorporated into the NLM-Mayo
data collection. There is variation in species, anatomy, pathology, scale,
and modality. In addition to providing linical qualitymedical image
data, the collection also includes newly acquired datasets of several animals,
including a whole mouse with both T and R data volumes.
This unique collection of data was categorized and organized into an intuitive
web-based browser which allows a user to rapidly access descriptive
information as well as the actual data volumes. The data collection
will be made available by the NLM for distribution, vis-a-vis its Visible
Human Project (VHP). Because the landscape of biomedical imaging
continues to change with new, advanced image acquisition systems and
techniques, continuously updating the VHP data collection seems prudent.
Additional new and varied image data will be incorporated into our
collection and disseminated to researchers in the medical image analysis
community.