The existing CMake and CTest environment provides an excellent platform for execution and submission of software quality tests. However, there is little support or aid for regression type tests provided to lessen the burden of writing tests and verifying the results. We propose an additional facility to compare measurements generated by a software quality test to a based-line result. Our approach enables the ability to use multiple images as measurements, along with integer, floating point and string values for regression tests. We describe a new testing class framework which provides the functionality of reading, writing, comparing and differentiating named measurements.

A Framework for Improved Regression Testing Based Upon CTest and CDash

In this paper a contour detection method is described and evaluated
on the evaluation data sets of the Cardiac MR Left Ventricle Segmentation
Challenge as part of MICCAI 2009’s 3D Segmentation Challenge for Clinical
Applications. The proposed method, using 2D AAM and 3D ASM, performs a
fully automated detection of the myocardial contours, not requiring any user
interaction. The algorithm’s performance is reported using the metrics provided
by the LV Challenge organization. Endocardial contour detection was classified
as successful in 86% of the images and epicardial contours in 94%. The average
perpendicular distance (APD) of the successful contours was 2.28 mm and
2.29 mm for the endo- and epicardial contours, respectively.

LV Challenge LKEB Contribution: Fully Automated Myocardial Contour Detection

This paper describes a fully automatic system to segment the left ventricle in all slices and all phases of a magnetic resonance cardiac cine study. After localizing the left ventricle blood pool using motion, thresholding and clustering, slices are segmented sequentially. For each slice, deformable registration is used to align all the phases, candidate contours are recovered in the average image using shortest paths, and a minimal surface is built to generate the final contours. The advantage of our method is that the resulting contours follow the edges in each phase and are consistent over time. As part of the MICCAI grand challenge on left ventricle segmentation, we demonstrate using 15 training datasets and 15 validation datasets that the results are very good with average errors around 2 mm and the method is ready for clinical routine.

Fully Automatic Left Ventricle Segmentation in Cardiac Cine MR Images Using Registration and Minimum Surfaces

Treatment planning for high precision radiotherapy of head and neck (H&N) cancer patients requires accurate delineation of critical structures. Manual contouring is tedious and often suffers from large inter- and intra-rater variability. In this paper, we present a fully automated, atlas-based segmentation method and apply it to tackle the H&N CT image segmentation problem in the MICCAI 2009 3D Segmentation
Grand Challenge. The proposed method employs a multiple atlas fusion strategy and a hierarchical atlas registration approach. We also exploit recent advancements in GPU technology to accelerate the deformable atlas registration and to make multi-atlas segmentation computationally feasible in practice. Validation results on the eight clinical datasets distributed by the MICCAI workshop showed that the proposed method gave very accurate segmentation of the mandible and the brainstem, with a volume
overlap close to or above 90% for most subjects. These results suggest that our method is clinically applicable, accurate, and may significantly reduce manual labor and improve contouring efficiency.

Automatic Segmentation of Head and Neck CT Images by GPU-Accelerated Multi-atlas Fusion

This paper presents a new approach to higher dimensional segmentation. We present an extended Active Shape Model (ASM) formulation for the segmentation of multi-contour anatomical structures. We employ coupling and weighting schemes to improve the robustness of ASM segmentation. 3D segmentation is achieved through propagation of a 2D ASM using a learned non-rigid spatial model. This approach does not suﬀer from the training and aligning diﬃculties faced by direct 3D model-based methods. Experimental results are encouraging at this early stage, and future directions of research are provided. 


Segmenting the Left Ventricle in 3D Using a Coupled ASM and a Learned Non-Rigid Spatial Model

As demonstrated by the example videos accompanying this submission of the
multiresolution implementation of Spherical Demons, the registration appears
reasonable. However, we are still unable to replicate the warps from the
stable and well verified implementation of this algorithm
url{http://sites.google.com/site/yeoyeo02/software/sphericaldemonsrelease}.

We find the average warp discrepancies between the original implementation and ITK
implementation to be about 5.3 mm on a sphere of radius 100.0 mm (note that
we do not expect a 100% agreement because of implementation differences, but
5.3 mm is relatively large). The submission also includes cortical surface
meshes of 39 subjects and the corresponding segmentation labels of the cortical
surfaces. Ultimately, the best validation would be to compare the overlap of
these segmentation labels after registration. We welcome fellow ITK developers
to work on this.

Please note that this ITK implementation of the algorithm is
currently being reviewed and tested in the NAMIC Sandbox at:
url{http://svn.na-mic.org/NAMICSandBox/trunk/QuadEdgeMeshSphericalDiffeomorphicDemonsRegistration/}.


If you want access to the stable and well verified implementation of this
algorithm, please use the MATLAB code available at:
url{http://sites.google.com/site/yeoyeo02/software/sphericaldemonsrelease}.


This document describes a contribution to the Insight Toolkit intended to support the process of performing deformable registration on two Meshes. The method implemented here is restricted to Meshes with a Spherical geometry and topology, and with scalar values associated to their nodes. The code described here is an implementation of the paper “Spherical Demons: Fast Diffeomorphic Landmark-Free Surface Registration” by Yeo, Sabuncu, Vercauteren, Ayache, Fischl and Golland [3, 4].

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.

Spherical Demons Registration of Spherical Surfaces

This document presents a set of classes which implement a region growing algorithm whose criterion for growth is different than existing ITK region growing algorithms. The region is grown into the query pixel based only on the difference between the query pixel and neighboring pixels that are already members of the region. This allows the region to be grown along paths that change gradually over time and end up looking nothing like they originally did. This is usefully in many situation, including growing surface patches around bends using growth criteria on the normal vectors. The usage and structure of these classes follow the style of itkConnectedThresholdImageFilter. We propose these classes as addition to the Insight Toolkit.

Region Growing Using a Criterion on the Region Boundary

In this paper, color image segmentation based on fuzzy logic has been studied. Anefficient Fuzzy logic inference engine based on magnetic fields has been implemented in this project so as to intelligently extract color information in the given image and classify it into the predominant color pyramid with the help of artificial color magnets. This also forms the basis for the priority based edge detection. Sobel operator is used in this project which is intelligently fused with the result of the above mentioned method to enhance the output so as to obtain a priority based enhanced edge detection output. 
Experimental results have demonstrated the effectiveness and superiority of the proposed method after extensive set of color images was tested.


Fuzzy Magnetic Color Image Segmentation and Advanced Edge Detection

This paper presents the description on Healthcare organizations are required to focus on Total quality improve:- Rendering acceptable, quality health services to patients at affordable price within reasonable price, within in a reasonable time; Applying zero errors to all patients services; maintaining a continuous error prevention program; Training employees in medical care on such aspects as error prevention,reducing delay time and providing prompt reasonable to patients needs; management system have always improvement in such systems to realize the true nature of the quality of healthcare and to be motivated towards improving this quality.In spite of billions of dollars of money spent worldwide, most of the healthcare is seen to be ineffective, inefficient and inadequate. Therefore there is a crying need to bring about a paradigm shift in the quality of health care delivery and to monitor and sustain it. It is obvious that those institutions, which are quality conscious and are committed to continuous quality improvement, will gain the highest consumer acceptance and will flourish at the expense of others. The 'Quality Revolution', as it is sometimes referred to, is nothing but putting the patient at the heart of health care and wrapping the care around it, rather than the other way around. Quality Measuring and assessing service is important, particularly when multiple sources of variation are present. Analyzing all medical processes to remove rework and waste could build health care quality and lead to significant reductions in patient cost. The use of quality-assurance programs and statistical tools can be directly applied to health-care organizations with improved quality of patient's objectives and the results of care from the patient's viewpoint.

Total Quality Management in Healthcare

Segmentation of the left ventricle is important in assessment of cardiac functional parameters. Currently, manual segmentation is the gold standard for acquiring these parameters and can be time-consuming. Therefore, accuracy and automation are two important criteria in improving cardiac image segmentation methods. In this paper, we present a comprehensive approach that utilizes various features of cine MR images and combines multiple image processing methods including thresholding, edge detection, mathematical morphology, deformable model as well as image filtering. The segmentation is performed automatically with minimized interaction to optimize segmentation results. This approach provides cardiac radiologists a practical method for accurate segmentation of the left ventricle.