Validation of Registration for Renal Dynamic Contrast Enhanced MRI Imaging
Journal Title: International Journal of Intelligent Systems and Applications in Engineering - Year 2016, Vol 4, Issue 3
Abstract
In Dynamic Contrast Enhanced Resonance Imaging (DCE-MRI), abdomen is scanned repeatedly and rapidly after injection of a contrast agent. During data acquisition, collected images suffer from the motion induced by the patient if he/she moves or breathes heavily during the scan. Therefore, these images should be aligned accurately to correct the motion. Recently, mutual information (MI) registration has become the first tool to register renal DCE-MRI images before any further processing. However, MI registration is sensitive to initial conditions and optimization methods, and it is bound to fail under certain conditions such as extreme movement or noise in the image. Therefore, if automated image analysis for renal DCE-MRI is to enter the clinical settings, it is necessary to have validation strategies that show the limitations of registration models on known datasets. In this study, two methods are introduced for the validation of registration of renal DCE-MRI images. The first method demonstrates how to use the inverse transform to generate realistic looking DCE-MRI kidney images and use them in validation. The second method shows how to generate checkerboard images and how to evaluate the goodness of registration for real DCE-MRI images. These validation methods can be incorporated into the registration studies to quantitatively and qualitatively demonstrate the success and the limitations of registration models.
Authors and Affiliations
Seniha Esen Yuksel *| Hacettepe University, Department of Electrical and Electronics Engineering, Ankara, Turkey.
Keywords
Related Articles
- EP ID EP800
- DOI 10.18201/ijisae.45496
- Views 451
- Downloads 23
Predicting Student Success in Courses via Collaborative Filtering
Based on their skills and interests, students’ success in courses may differ greatly. Predicting student success in courses before they take them may be important. For instance, students may choose elective courses that...
Using Word Embeddings for Ontology Enrichment
Word embeddings, distributed word representations in a reduced linear space, show a lot of promise for accomplishing Natural Language Processing (NLP) tasks in an unsupervised manner. In this study, we investigate if the...
Developing a Fuzzy Logic Decision Support System for Strategic Planning in Industrial Organizations
Internal – External (IE), Strategic Position and Action Evaluation (SPACE), Boston Consulting Group (BCG), and Grand Strategy matrices are important tools in generating and evaluating alternative output strategies which...
Comparative Study of Krill Herd, Firefly and Cuckoo Search Algorithms for Unimodal and Multimodal Optimization
Today, in computer science, a computational challenge exists in finding a globally optimized solution from an enormously large search space. Various metaheuristic methods can be used for finding the solution in a large s...
Improving Intrusion Detection using Genetic Linear Discriminant Analysis
The objective of this research is to propose an efficient soft computing approach with high detection rates and low false alarms while maintaining low cost and shorter detection time for intrusion detection. Our results...