A novel platform for realizing software prototypes in the area of medical image processing, analysis and visualization is introduced: The \"Medical Image Processing Platform\" (MIPP). A novel feature is the combination of ITK/VTK with Java/Eclipse RCP, four potential technologies complementing each other. While ITK and VTK provide comprehensive functionality for image processing and visualization, Java and Eclipse RCP facilitate the development of a modern and rich application.

RCP offers everything for creating plug-in based software. The benefits of integrating ITK and VTK in such a modular architecture are reusability and extensibility. Functionality usually found in software for medical image processing and visualization (2D and 3D viewers, DICOM and application data management, etc.) is provided by the MIPP platform as plug-ins, can be easily reused by new applications and needn\'t be re-implemented. Developers can extend the platform\'s features by contributing new plug-ins and, again, make these available for others.

The paper describes the architecture of the MIPP platform and outlines its most important functionality. Finally, three clinical software systems implemented with MIPP are presented.

MIPP: A Medical Image Processing Platform based on ITK, VTK and Eclipse RCP

Connected component is a well known and very useful notion in binary case. As often in mathematical morphology, this notion can be extended to the grayscale case, and allow to perform lot of the useful transforms based on binary connected components in the grayscale images. This article describe the component tree, a data structure able to ef?ciently represent the grayscale connected components in an grayscale image, as well as the algorithm used to build the component tree. All the source codes are provided, as well as a full set of tests and several usage examples of the new classes.

Component tree: an efficient representation of grayscale connected components

FusionViewer is an open source and platform independent viewer that has been specifically designed for PET/CT image display. The combination of PET and CT images offers complementary functional and anatomical information. The application (FusionViewer) facilitates efficient visualization and analysis of PET/CT studies via different viewing modes (linked cursor display, alpha-blend mode, checkerboard mode and split window mode). FusionViewer is implemented in Java and uses the Java OpenGL (JOGL) library and the Insight Segmentation and Registration Toolkit (ITK) library, which make it both a fast and cross-platform application. Its intuitive graphical user interface makes it easy to be used by physicians, radiologists, and research scientists. Several analysis and display tools are already available (navigator, zoom, pan, screen snapshot, ROI, and line measure tool; alpha-blending, checkerboard display, and split window display). Along with PET/CT, several other modalities where co-registered images are often visualized simultaneously have benefited from the use of this software.

FusionViewer: An Open Source Toolkit for Viewing Multimodality Images

This paper describes the ITK implementation of a minimal path extraction framework based on Fast Marching arrival functions. The method requires the user to provide three inputs: 1. a meaningful speed function to generate an arrival function, 2. path information in the form of start, end, and way-points (which the path must pass near), and 3. an optimizer which steps along the resultant arrival function perpendicular to the Fast Marching front. A number of perspectives for choosing speed functions and optimizers are given, as well as examples using synthetic and real images.

Fast Marching Minimal Path Extraction in ITK

This document describes a simple helper class intended for making easy to initialize the grid parameters of a BSplineDeformableTranform. This paper is accompanied with the source code, input data, parameters and output data that we used for validating the algorithm described in this paper. This adheres to the fundamental principle that scientific publications must facilitate reproducibility of the reported results.

Helper class for initializing the grid parameters of a BSpline deformable transform by using an image as reference

ly in ITK, the only way to find the contour of the objects in a binary image is to use the BinaryErodeImageFilter, with a kernel of radius 1. This filter is a generic filter, made to support any shape and size of structuring element, and thus is not optimized for the particular case needed to detect the contours. Moreover, that filter is not multithreaded, so it can't get the performance improvements allowed by the multiprocessor systems. As a result, the contour detection can be quite time consuming currently -- for example, in SignedMaurerDistanceMapImageFilter, the contour detection takes about 33% of the execution time.
This contribution comes with a new filter which highly improve the performance of the contour detection in the binary image, and a second filter which allow the detection of the countours in label images with similar performance.

Efficient contour detection in binary and label images

We present and implement an error estimation protocol in the Insight Toolkit (ITK) for assessing the
accuracy of image alignment. We base this error estimation on a robust version of the HausdorffDistance
(HD) metric applied to the recovered edges of the images. The robust modifications we introduce to
the HD metric significantly reduce the amount of outliers in the local distance error estimation. We
evaluate the accuracy of our protocol on synthetically deformed images. We provide the source code
and datasets to reproduce this evaluation. The proposed method is shown to improve error assessment
when it is compared with conventional HD methods. This approach has many applications including
local estimation and visual assessment of registration error and registration parameter selection.

The Use of Robust Local Hausdorff Distances in Accuracy Assessment for Image Alignment of Brain MRI

Morphological erosion and dilation filters employ a structuring function, with flat structuring functions being the most common example. This article will use the term \"parabolic\", but much of the literature uses \"quadratic\" structuring functions are less well known but theoretically very important and practically very useful. This paper briefly introduces morphology using parabolic structuring functions, describes the ITK classes used to implement them and includes a number of sample applications.

Morphology with parabolic structuring elements

Semi-automatic image segmentation algorithms depend on interaction with the user to accurately define a region of interest within an image. Once such method is a dynamic programming approach called {em Intelligent Scissors} developed by Mortenson and Barret cite{Mortensen1992,Mortensen1995,Barrett1996,Barrett1997}. Standard interaction involves the user-placement of a seed point on or near the boundary of the object to be extracted. Using a gradient-based cost function, a {em live-wire} image path from the seed point to a subsequently placed boundary point is determined. As this free point is manipulated with the mouse cursor, the live-wire boundary, which extends from the seed point to the varying free point, locks onto nearby edges within the image. Although some applications do not require such user-interaction (e.g. cite{Hu2001}) warranting inclusion within the toolkit, incorporating live-wire capabilities into ITK-SNAP cite{Yushkevich2006} has been discussed and, therefore, we would like to, as a preliminary step, vet the code within the ITK community.

Live-Wire-ing the Insight Toolkit with Intelligent Scissors

Texture analysis provides quantitative information describing properties in images such as coarseness and smoothness. Two common quantification schemes are based on co-occurence matrices and run-length matrices. Although the co-occurence measures are readily available in the Insight Toolkit, no such set of classes exists for run-length measures. This submission is meant to remedy this deficiency by providing a set of classes which are modeled after the ITK co-occurence measures classes.