This document describes the latest efforts in integrating the Lesion Sizing Toolkit (LSTK) into ITK v4 as an external/remote module providing an Open Science dashboard website with a large open image archive of lung cancer CT images for LSTK development and testing.

Fostering Open Science in Lung Cancer Lesion Sizing with ITK module LSTK

This document describes the implementation in ITK of the Incremental Delaunay Triangulation algorithm. Using the Straight Walk in Triangulation function, the exact discrete geometrical orientation predicate, and the itk::QuadEdgeMesh API of ITK , we propose a geometrically exact and robust implementation that, from a given 2-dimensional itk::PointSet, incrementally constructs the corresponding 2-dimensional Delaunay Triangulation as an itk::QuadEdgeMesh.

Incremental Delaunay Triangulation

This paper describes the Statismo framework, which is a framework for PCA based statistical models.Statistical models are used to describe the variability of an object within a population, learned from a set of training samples. Originally developed to model shapes, statistical models are now increasingly used to model the variation in different kind of data, such as for example images, volumetric meshes or deformation fields. Statismo has been developed with the following main goals in mind: 
1) To provide generic tools for learning different kinds of PCA based statistical models, such as shape, appearance or deformations models. 
2) To make the exchange of such models easier among different research groups and to improve the reproducibility of the models. 
3) To allow for easy integration of new methods for model building into the framework. 
To achieve the first goal, we have abstracted all the aspects that are specific to a given model and data representation, into a user defined class. This does not only make it possible to use Statismo to create different kinds of PCA models, but also allows Statismo to be used with any toolkit and data format. To facilitate data exchange, Statismo defines a storage format based on HDF5, which includes all the information necessary to use the model, as well as meta-data about the model creation, which helps to make model building reproducible. The last goal is achieved by providing a clear separation between data management, model building and model representation. In addition to the standard method for building PCA models, Statismo already includes two recently proposed algorithms for building conditional models, as well as convenience tools for facilitating cross-validation studies.
Although Statismo has been designed to be independent of a particular toolkit, special efforts have been made to make it directly useful for VTK and ITK. Besides supporting model building for most data representations used by VTK and ITK, it also provides an ITK transform class, which allows for the integration of Statismo with the ITK registration framework. This leverages the efforts from the ITK project to readily access powerful methods for model fitting.

Statismo - A framework for PCA based statistical models

Many enabling technologies like non-rigid registration in medical image computing rely on the construction of a function by interpolating scattered points; however, the outliers contained in the data and the approximation error make the robust and accurate estimation difficult. This paper presents an ITK implementation of a robust Finite Element (FE) solver, which can effectively deal with the above difficulties. The experiment results of synthetic data and real cases demonstrate the characteristics of the robust solver and its typical application. 

Robust Scattered Data Points Approximation Using Finite Element Biomechanical Model

Ultrasound (US) imaging is a popular and convenient medical imaging modality thanks to its mobility, non-ionizing radiation, ease-of-use, and real-time acquisition. Conventional US imaging is frequently integrated with tracking systems and robotic systems in Image Guided Therapy (IGT) systems. Recently, these systems are also incorporating advanced US imaging such as US elasticity imaging, photoacoustic imaging, and thermal imaging. Real-time synchronous data from multiple sources and bidirectional data communication are essential for integrating components in advanced US IGT research. We previously proposed the MUSiiC ToolKit [1], a modular real-time software toolkit, and OpenIGTLinkMUSiiC [2], a standard communication protocol extended from the OpenIGTLink library [3, 4]. However, this software framework only supported real-time synchronous data from at most two sources and unidirectional communication at the software module level and class level.

In this paper, we propose MUSiiC ToolKit 2.0, an upgraded software framework for interventional advanced US research sup-porting bidirectional communication, real-time US data processing, and real-time data synchronization from multiple sources. MUSiiC ToolKit 2.0 consists of OpenIGTLink 2.0, OpenIGTlinkMUSiiC 2.0, MUSiiCNotes 2.0, and a collection of executable programs designed for US research. OpenIGTLink 2.0 is a standard TCP/IP-based protocol for the integration of medical imaging and IGT systems. OpenIGTLinkMUSiiC 2.0 is the upgraded version of OpenIGTLinkMUSiiC with new active multi-task classes, data interfaces for supporting bidirectional communication and parallel data processing. MUSiiCNotes 2.0 provides US research-oriented task classes, such as US data acquisition, beamforming, envelope detection, scan conversion, and data synchronization. Graphic User Interface (GUI) units are also available for the executable programs in MUSiiCNotes 2.0. Finally, we introduce advanced US applications based on this new software framework. 


MUSiiC ToolKit 2.0: Bidirectional Real-time Software Framework for Advanced Interventional Ultrasound Research

Modern medical imaging analyses often involve the concatenation of multiple steps, and neuroimaging analysis is no exception. The Java Image Science Toolkit (JIST) has provided a framework for both end users and engineers to synthesize processing modules into tailored, automatic multi-step processing pipelines (“layouts”) and rapid prototyping of module development. Since its release, JIST has facilitated substantial neuroimaging research and fulfilled much of its intended goal. However, key weaknesses must be addressed for JIST to more fully realize its potential and become accessible to an even broader community base. Herein, we identify three core challenges facing traditional JIST (JIST-I) and present the solutions in the next generation JIST (JIST-II). First, in response to community demand, we have introduced seamless data visualization; users can now click ‘show this data’ through the program interfaces and avoid the need to locating files on the disk. Second, as JIST is an open-source community effort by-design; any developer may add modules to the distribution and extend existing functionality for release. However, the large number of developers and different use cases introduced instability into the overall JIST-I framework, causing users to freeze on different, incompatible versions of JIST-I, and the JIST community began to fracture. JIST-II addresses the problem of compilation instability by performing continuous integration checks nightly to ensure community implemented changes do not negatively impact overall JIST-II functionality. Third, JIST-II allows developers and users to ensure that functionality is preserved by running functionality checks nightly using the continuous integration framework. With JIST-II, users can submit layout test cases and quality control criteria through a new GUI. These test cases capture all runtime parameters and help to ensure that the module produces results within tolerance, despite changes in the underlying architecture. These three “next generation” improvements increase the fidelity of the JIST framework and enhance utility by allowing researchers to more seamlessly and robustly build, manage, and understand medical image analysis processing pipelines.

Next Generation of the JAVA Image Science Toolkit (JIST) Visualization and Validation

This document describes the workflow for unwarping echo planar MR images (EPI), in particular diffusion-weighted images, using an acquisition method with reversed phase encoding and the tools of the Computational Morphometry Toolkit (CMTK). This technique requires an additional acquisition (additional b=0 image for diffusion imaging) with reversed phase encoding direction, but no field map. Simple 6-direction diffusion data are provided with this article for demonstration.

Unwarping Echo Planar Images Using CMTK

As part of the ITK v4 project efforts, we have developed ITK filters for physics-based non-rigid registration (PBNRR), which satisfies the following requirements: account for tissue properties in the registration, improve accuracy compared to rigid registration, and reduce execution time using GPU and multi-core accelerators. The implementation has three main components: (1) Feature Point Selection, (2) Block Matching (mapped to both multi-core and GPU processors), and (3) a Robust Finite Element Solver. The use of multi-core and GPU accelerators in ITK v4 provides substantial performance improvements. For example, in average for the non-rigid registration of brain MRIs, the performance of the Block Matching filter is about 12 times faster when 12 hyperthreaded multi-cores are used and about 540 times faster when the Quadro 6000 with 448 threads is used in Dell Workstation.

An ITK Implementation of Physics-based Non-rigid Registration Method

Intensity inhomogeneities often occur in medical images, especially when using magnetic resonance imaging. In these images, the standard Chan-and-Vese levelset segmentation method may not work properly, as it assumes constant intensity distributions for foreground and background. Recently, a novel method was published that models the intensities as piece-wise smooth, and thus is more suitable to segment images with intensity homogeneities. However, this method was not yet implemented in ITK. This submission introduces our implementation of the region-scalable-fitting levelset segmentation method within the ITKv4 levelset framework.


An implementation of the minimization of region-scalable fitting energy levelsets

This document describes a contribution to the Insight Toolkit intended to support the analysis of the principal components of data sets, optionally point data associated with the vertices of a mesh.
This paper is accompanied with the source code, input data, parameters and output data that we used for validating the implementation described in this paper. This adheres to the fundamental principle that scientific publications must facilitate reproducibility of the reported results.