Evaluation of the Cardiac Global Function from Tagged Magnetic Resonance Images /
Mona Ragab Sayed
Evaluation of the Cardiac Global Function from Tagged Magnetic Resonance Images / Mona Ragab Sayed - 2013 - 103 p. ill. 21 cm.
Supervisor: Ahmed Fahmy
Thesis (M.A.)—Nile University, Egypt, 2013 .
"Includes bibliographical references"
Contents:
1. INTRODUCTION ................................................................................................. 1
1.1 Problem and motivation ................................................................................. 1
1.2 Imaging Techniques ....................................................................................... 2
1.3 Diseases overview .......................................................................................... 3
1.4 Thesis arrangement ........................................................................................ 6
2. BACKGROUND ................................................................................................... 8
2.1 Introduction .................................................................................................... 8
2.2 The Human Heart ........................................................................................... 8
2.2.1 Heart Anatomy ........................................................................................ 8
2.2.2 17-Segment Model ................................................................................ 11
2.3 Cardiac MRI ................................................................................................. 14
2.3.1 Global cardiac function ......................................................................... 15
2.3.2 Regional cardiac function ..................................................................... 18
2.4 Summary ...................................................................................................... 21
3. RELATED WORK .............................................................................................. 22
3.1 Introduction .................................................................................................. 22
3.2 Previous trials ............................................................................................... 23
3.2.1 Comparison between tagged MRI and standard cine MRI for evaluation of left ventricular ejection fraction ...................................................................... 24
3.2.2 Tagged cardiac MRI: Detection of myocardial boundaries by texture analysis …………………………………………………………………………27
3.2.3 Tagged Cardiac MR Image Segmentation by Contrast Enhancement and Texture Analysis .................................................................................................. 31
3.2.4 A novel tag removal technique for tagged cardiac MRI ....................... 36
3.3 Summary ................
4. PROPOSED TECHNIQUE ................................................................................. 38
4.1 Introduction .................................................................................................. 38
4.2 Tag Removal using Hybrid Texture-Frequency based method ................... 38
4.2.1 Texture-based contrast enhancement .................................................... 40
4.2.2 Frequency-based tag removal ............................................................... 41
4.2.3 Tagless image formation ....................................................................... 43
4.3 Testing and Validation ................................................................................. 44
4.3.1 Testing Dataset...................................................................................... 44
4.3.2 Analysis of standard cine image sequences .......................................... 44
4.3.3 Analysis of tagged image sequences ..................................................... 46
4.3.4 Statistical analysis ................................................................................. 48
4.4 Summary ...................................................................................................... 49
5. RESULTS AND DISCUSSION.......................................................................... 50
6. OTHER EXPERIMENTS AND DISCUSSION ................................................. 55
6.1 Introduction .................................................................................................. 55
6.2 Experiment 1: Comparison between tagged MRI and standard cine MRI for evaluation of left ventricular ejection fraction ........................................................ 55
6.3 Experiment 2: Tagged cardiac MRI: Detection of myocardial boundaries by texture analysis ........................................................................................................ 56
6.4 Experiment 3: Tagged Cardiac MR Image Segmentation by Contrast Enhancement and Texture Analysis ........................................................................ 65
6.5 Experiment 4: Evaluation of the Proposed Technique in Comparison with Texture-based and Frequency based techniques ..................................................... 69
6.6 Summary ...................................................................................................... 71
7. CONCLUSION AND FUTURE WORK ............................................................ 72
REFERENCES ........................................................................................................... 75
APPENDICES ..............................
Abstract:
Heart diseases lead the causes of human death. Some of these diseases cause suddenly death without any indicators to the symptoms or causes after death. As Centers for Disease Control and Prevention in USA stated, cardiovascular diseases account for more than one-third (33.6%) of all U.S deaths. Cardiovascular diseases may also lead to disabilities. In 2007, 150,000 who are younger than age 65 died. Nearly 4 million people are also reporting disability from these causes. Some of the diseases that lead to death are ventricular fibrillation, failure of ventricle, coronary artery and Hypertrophic Cardiomyopathy (HCM).
Cardiovascular diseases problem is one of the most widespread health problem. It is important to search for and enhance techniques that help in detection and evaluation of the heart. Magnetic Resonance Imaging (MRI) is a popular technique for non-invasive assessment of human organs. Cardiac MR (CMR) is of major interest for the diagnosis and treatment of cardiovascular diseases of the cardiac function.
Cine Steady State Free Processing (SSFP) imaging is MRI technique that generates high contrast sequences of images which represent the whole cardiac cycle. These images are used to evaluate the global function of the heart. Ventricular volumes, masses and Ejection fraction are measured as indicators to the functionality of the whole heart. Nevertheless, this technique cannot be used to describe the contractile properties until tagging techniques have appeared.
Tagged Magnetic Resonance (MR) images are considered the gold standard for evaluating the cardiac regional function. The cine MR images are manipulated to have new images with visible dark lines over the tissue at myocardium region. These lines called tags. These tags are tracked over the cardiac cycle. Regional function is estimated by quantitative measurements of tags deformation that occurs while heart moving.
Using two different imaging techniques to evaluate the myocardium function is inefficient and inconvenient to the patient and the doctor. Therefore, there is a need for a technique to estimate the global function from the tagged MR images without incurring extra scanning time. Nevertheless, the low contrast in tagged images hinders the segmentation process, and hence, the global function may be inaccurate.
In this thesis, new method for enhancing the myocardium-to-blood contrast in tagged MR images is proposed. First, the tag pattern in each input tagged MR image is removed using a technique based on the image texture and the frequency components of the tag pattern to produce two tagless images. Then, these two images are combined to obtain a tagless MR image with high blood-to-myocardium contrast. The proposed method is applied to five patients where the global functional parameters are calculated and compared to those calculated from standard untagged cine MR images of the same subjects. The results show that the proposed method has a high potential to be used in evaluating the global functional parameters from tagged MR images.
Text in English, abstracts in English.
Informatics-IFM
Dissertation, Academic
610
Evaluation of the Cardiac Global Function from Tagged Magnetic Resonance Images / Mona Ragab Sayed - 2013 - 103 p. ill. 21 cm.
Supervisor: Ahmed Fahmy
Thesis (M.A.)—Nile University, Egypt, 2013 .
"Includes bibliographical references"
Contents:
1. INTRODUCTION ................................................................................................. 1
1.1 Problem and motivation ................................................................................. 1
1.2 Imaging Techniques ....................................................................................... 2
1.3 Diseases overview .......................................................................................... 3
1.4 Thesis arrangement ........................................................................................ 6
2. BACKGROUND ................................................................................................... 8
2.1 Introduction .................................................................................................... 8
2.2 The Human Heart ........................................................................................... 8
2.2.1 Heart Anatomy ........................................................................................ 8
2.2.2 17-Segment Model ................................................................................ 11
2.3 Cardiac MRI ................................................................................................. 14
2.3.1 Global cardiac function ......................................................................... 15
2.3.2 Regional cardiac function ..................................................................... 18
2.4 Summary ...................................................................................................... 21
3. RELATED WORK .............................................................................................. 22
3.1 Introduction .................................................................................................. 22
3.2 Previous trials ............................................................................................... 23
3.2.1 Comparison between tagged MRI and standard cine MRI for evaluation of left ventricular ejection fraction ...................................................................... 24
3.2.2 Tagged cardiac MRI: Detection of myocardial boundaries by texture analysis …………………………………………………………………………27
3.2.3 Tagged Cardiac MR Image Segmentation by Contrast Enhancement and Texture Analysis .................................................................................................. 31
3.2.4 A novel tag removal technique for tagged cardiac MRI ....................... 36
3.3 Summary ................
4. PROPOSED TECHNIQUE ................................................................................. 38
4.1 Introduction .................................................................................................. 38
4.2 Tag Removal using Hybrid Texture-Frequency based method ................... 38
4.2.1 Texture-based contrast enhancement .................................................... 40
4.2.2 Frequency-based tag removal ............................................................... 41
4.2.3 Tagless image formation ....................................................................... 43
4.3 Testing and Validation ................................................................................. 44
4.3.1 Testing Dataset...................................................................................... 44
4.3.2 Analysis of standard cine image sequences .......................................... 44
4.3.3 Analysis of tagged image sequences ..................................................... 46
4.3.4 Statistical analysis ................................................................................. 48
4.4 Summary ...................................................................................................... 49
5. RESULTS AND DISCUSSION.......................................................................... 50
6. OTHER EXPERIMENTS AND DISCUSSION ................................................. 55
6.1 Introduction .................................................................................................. 55
6.2 Experiment 1: Comparison between tagged MRI and standard cine MRI for evaluation of left ventricular ejection fraction ........................................................ 55
6.3 Experiment 2: Tagged cardiac MRI: Detection of myocardial boundaries by texture analysis ........................................................................................................ 56
6.4 Experiment 3: Tagged Cardiac MR Image Segmentation by Contrast Enhancement and Texture Analysis ........................................................................ 65
6.5 Experiment 4: Evaluation of the Proposed Technique in Comparison with Texture-based and Frequency based techniques ..................................................... 69
6.6 Summary ...................................................................................................... 71
7. CONCLUSION AND FUTURE WORK ............................................................ 72
REFERENCES ........................................................................................................... 75
APPENDICES ..............................
Abstract:
Heart diseases lead the causes of human death. Some of these diseases cause suddenly death without any indicators to the symptoms or causes after death. As Centers for Disease Control and Prevention in USA stated, cardiovascular diseases account for more than one-third (33.6%) of all U.S deaths. Cardiovascular diseases may also lead to disabilities. In 2007, 150,000 who are younger than age 65 died. Nearly 4 million people are also reporting disability from these causes. Some of the diseases that lead to death are ventricular fibrillation, failure of ventricle, coronary artery and Hypertrophic Cardiomyopathy (HCM).
Cardiovascular diseases problem is one of the most widespread health problem. It is important to search for and enhance techniques that help in detection and evaluation of the heart. Magnetic Resonance Imaging (MRI) is a popular technique for non-invasive assessment of human organs. Cardiac MR (CMR) is of major interest for the diagnosis and treatment of cardiovascular diseases of the cardiac function.
Cine Steady State Free Processing (SSFP) imaging is MRI technique that generates high contrast sequences of images which represent the whole cardiac cycle. These images are used to evaluate the global function of the heart. Ventricular volumes, masses and Ejection fraction are measured as indicators to the functionality of the whole heart. Nevertheless, this technique cannot be used to describe the contractile properties until tagging techniques have appeared.
Tagged Magnetic Resonance (MR) images are considered the gold standard for evaluating the cardiac regional function. The cine MR images are manipulated to have new images with visible dark lines over the tissue at myocardium region. These lines called tags. These tags are tracked over the cardiac cycle. Regional function is estimated by quantitative measurements of tags deformation that occurs while heart moving.
Using two different imaging techniques to evaluate the myocardium function is inefficient and inconvenient to the patient and the doctor. Therefore, there is a need for a technique to estimate the global function from the tagged MR images without incurring extra scanning time. Nevertheless, the low contrast in tagged images hinders the segmentation process, and hence, the global function may be inaccurate.
In this thesis, new method for enhancing the myocardium-to-blood contrast in tagged MR images is proposed. First, the tag pattern in each input tagged MR image is removed using a technique based on the image texture and the frequency components of the tag pattern to produce two tagless images. Then, these two images are combined to obtain a tagless MR image with high blood-to-myocardium contrast. The proposed method is applied to five patients where the global functional parameters are calculated and compared to those calculated from standard untagged cine MR images of the same subjects. The results show that the proposed method has a high potential to be used in evaluating the global functional parameters from tagged MR images.
Text in English, abstracts in English.
Informatics-IFM
Dissertation, Academic
610