This document describes an ITK class implementing an Adaptive Moment Estimator (Adam) optimizer algorithm within the Insight Toolkit ITK www.itk.org. Adam is an adaptive gradient descent optimizer, which independently adaptively estimates the gradient descent step for each parameter, at each iteration, based on stored past gradients. The optimizer stores exponentially decaying averages of past gradients to estimate first moment (the mean) and the second moment (the variance) of the gradients to formulate update rule for present iteration. The Adam optimizer compares favorably to other adaptive learning-method algorithms, converges faster, and is robust to saddle point. This paper is accompanied with the source code, input data, parameters and output data that the authors used for validating the algorithm described in this paper. 

Adaptive Moment Estimator (Adam) Optimizer in ITK v3

Recently, the scientific community has been proposing several automatic algorithms to biomedical image segmentation procedure, being an interesting and helpful approach to assist both technicians and radiologists in this time-consuming and subjective task. One of these interesting and widely used image segmentation method could be the voxel intensity-based algorithms, e.g. image histogram threshold methods, which have been intensively improved in the past decades. Recently, an interesting approach that gained focus is the logistic classification (LC) for object detection in biomedical images. Even though the general concept behind the LC method is fairly known, the proper method's optimization still commonly adjusted by hand which naturally adds a level of uncertainty and subjectivity in the general segmentation performance. Therefore, an empirical LC optimization is presented, offering a ITK class that performs the LC parameters optimization based on empirical input data analysis. It is worth mentioning that the LogisticContrastEnhancementImageFilter class showed here is also applied on others computational problems, being briefly explained in this document.

An Empirical Optimization to Logistic Classification Model

In this paper, we describe a set of filters, implemented in the Insight Toolkit www.itk.org, for converting an image from Cartesian co-ordinate space to Polar co-ordinate space and vice-versa. Cartesian to Polar conversion of an image is a useful operation in preprocessing stage of certain image-processing algorithm where feature of interest has simplified representation in the polar space. This paper is accompanied with the source code, input data, parameters and output data that the authors 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.

CartesianToPolar and PolarToCartesian Image Filters

K-Means clustering is an excellent technique for clustering points when the number of clusters is known. We present a implementation (vtkKMeanClustering) of the algorithm written in a VTK context. We also implement the K-Means++ initialization method which finds the global optimum much more frequently than a naive/random initialization.

The code is currently hosted at http://github.com/daviddoria/KMeansClustering .

A K-Means++ Clustering Implementation for VTK

It is convenient to acquire data in a format which VTK can read directly. To enable this, we provide a wrapper of the Hokuyo UTM-30LX interface in a class vtkHokuyo. The code is currently hosted at git@github.com:daviddoria/vtkHokuyo.git. This class only works on Linux systems.

A VTK Interface for the Hokuyo UTM 30LX Laser Range Finder

Being able to quickly compute the inverse of a deformation ﬁeld is often useful in the context of medical image analysis. While ITK supports this functionality, the current algorithms are slow and do not always yield accurate results. In this paper we describe an ITK implementation of a ﬁxed point algorithm for the approximate inversion of deformation ﬁelds that was recently proposed by M. Chen and co-workers. The algorithm has been shown to be both faster and more accurate than those currently implemented in ITK.


Inverting deformation fields using a fixed point iteration scheme

The Random Sample Consensus (RANSAC) algorithm for robust parameter value estimation has been applied to a wide variety of parametric entities (e.g. plane, the fundamental matrix). In many implementations the algorithm is tightly integrated with code pertaining to a specific parametric object. In this paper we introduce a generic RANSAC implementation that is independent of the estimated object. Thus, the user is able to ignore outlying data elements potentially found in their input. To illustrate the use of the algorithm we implement the required components for estimating the parameter values of a hyperplane and hypersphere.


Random Sample Consensus (RANSAC) Algorithm, A Generic Implementation

In this document we present an ITK image filter that outputs a new image with identical voxel contents but with modified physical space representation. The physical space representation of an image is modified to reflect the physical space transform on the input image. The advantage of this filter is that it is free from interpolation error.
This paper is accompanied with the source code, input data, parameters and output data that the au- thors 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.

Introduction to ITK Resample In-Place Image Filter


We think about interoperability only in today’s terms. Looking ahead, the demands that future technologies will make on health information exchange could be large, as could the health because of interoperability. This paper would present the description on Healthcare Information System organizations and communities as a standard to underpin clinical information exchange. A number of signiﬁcant problems results are directly from the way the computable entries in clinical documents exchange represented currently. In this paper, the information exchange with other system and also supporting communication of electronic healthcare records across organization boundaries. The CCD is an XML-based specification for exchange of clinical summary information. The HL7 Reference Information Models expressed in the Clinical Document Architecture Release 2 (CDA R2), an information exchange specification generic to any type of clinical information. Hence, the patient’s health information is easily communicated points of care and improves quality services. 


HL7 CCD - Clinical Interoperability to Improve Quality and the Point-of-Care of EHR

We have created a wrapper package named SimVTK that allows VTK, and third-party classes derived from VTK classes, to be seamlessly integrated with MATLAB's Simulink interface. Our package generates a loadable Simulink module for each VTK class, which is then represented as a "block" on the Simulink canvas, and can be connected with other blocks to form a pipeline. Each block can be double-clicked to bring up a dialog box that allows introspection and control of the VTK class. After a VTK pipeline has been built, it can be run interactively from within Simulink. The outputs of the pipeline can be displayed in an interactive render window, written to a disk file, or exported to a MATLAB variable. Within Simulink, the VTK pipeline can also be connected to the ITK pipeline through the use of our SimITK package. We generate the VTK-Simulink wrapper automatically though the use of CMake build scripts, with XML as an intermediate representation of the VTK class interfaces.
