A subject-specific correction of four-dimensional flow magneticresonance images for cerebrospinal fluid flow measurements
Başlık çevirisi mevcut değil.
- Tez No: 780971
- Danışmanlar: Belirtilmemiş.
- Tez Türü: Doktora
- Konular: Nöroloji, Radyoloji ve Nükleer Tıp, Neurology, Radiology and Nuclear Medicine
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 2022
- Dil: İngilizce
- Üniversite: Osaka Unıversıty
- Enstitü: Yurtdışı Enstitü
- Ana Bilim Dalı: Belirtilmemiş.
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: Belirtilmemiş.
Özet
Flow magnetic resonance imaging (Flow MRI) is a non-invasive method that is capable of visualizing in-vivo human body fluid flow velocities. The four-dimensional (4D) flow MRI supplies a time series of velocity components in three directions (superior-inferior, anterior-posterior, left-right) during the cardiac cycle, and enables flow visualization, recording, and measurement of not only the blood flow in the cardiovascular system but also cerebrospinal fluid (CSF) flow in the neurological system. However, there is still no consensus about the usage of 4D flow MRI for analyzing slow flows like CSF, because of non-negligible phase-offset errors caused by its acquisition principles. Since the 4D flow MRI is a unique tool to understand CSF pathophysiology for managing diseases, this study considers an automated approach for the correction of phase offset errors in the CSF flow velocity map. Furthermore, objective methods were suggested for analyzing CSF flow by 4D flow MRI after correction of phase offset errors. In Chapter 2, 4D flow MRI basics and limitations are explained. The MRI is based on nuclear magnetic resonance (NMR) of the nucleus, and in flow MRI magnetic field gradients are used for calculating the velocity of mobile hydrogen nuclei. Phase-contrast MRI is the most popular flow MRI which is acquired by using bipolar gradients for eliminating magnetic field inhomogeneities on nuclei. The range of the velocity is determined by the strength of the bipolar gradients. Measuring slow flows requires low-velocity encoding which needs large bipolar gradients. However, these large gradients cause unexpected electromagnetic induction resulting in nonnegligible phase offset errors. Besides, the lowspatiotemporal resolution also leads to errors like partial volume artifacts. In Chapter 3, an automated correction method for the phase offset errors of 4D flow MRI was presented. Since the phase offset errors were assumed to be constant in time and smoothly distributed spatially in previous existing studies, the spatial distribution of the phase offset errors was modeled as the second-degree polynomial function using robust regression analysis. Flow velocity maps were corrected by the estimated phase-offset errors by robust regression analysis. The method was applied to the CSF flow maps obtained by 9 idiopathic normal-pressure hydrocephalus (iNPH) patients and 9 healthy subjects' 4D flow MRI data. The residual standard errors between the original data and the estimated phase-offset errors were analyzed for the evaluation of estimation consistency and found under 1.7 mm/s for each case. This chapter provides an approach to the correction of phase offset errors for enabling more robust and advanced CSF studies. In Chapter 4, flow rates in aqueduct cerebri were evaluated from the CSF velocity maps corrected in Chapter 3. The aqueduct cerebri was segmented and aliasing artifacts and partial volume artifacts were corrected before calculations. For these processes, semiautomated and automated approaches were suggested with aim of creating an objective environment for the analysis of CSF flow parameters. Flow rates were calculated for the analyzing CSF flow in this study, and significant differences were found at peak diastolic flow rates between iNPH patients and healthy cases. (p=0.02, Mann-Whitney's U test). Additionally, the mean aqueduct cerebri area at the axial plane was found significantly different (p=0.01, Mann-Whitney's U test). In the final chapter, the dissertation is summarized, and future perspectives were given. This study aims to create an objective and automated image process for improving CSF studies with 4D PC-MRI. In this way, we hope future studies could build parameters for understanding CSF pathophysiology in an objective environment.
Özet (Çeviri)
Özet çevirisi mevcut değil.
Benzer Tezler
- Gaz türbin kanadı üzerinde iki boyutlu sıcaklık dağılımının çıkartılması
Two dimensional temperature distribution on a gas turbine blade
İBRAHİM KOÇ
- Doğal dolaşımlı, kendinden depolu güneş toplayıcısının ısıl analizi
The Thermal analysis of a novel built-in-stroge solar water heater
İ.NECMİ KAPTAN
- SQL/DS ve CSP/AD ile kütüphane uygulaması geliştirme
Design and implementation of a library application with SQL/DS and CSP/AD
İSRAFİL AVCI
Yüksek Lisans
Türkçe
1990
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrolİstanbul Teknik ÜniversitesiDOÇ.DR. BÜLENT ÖRENCİK
- Gaz türbin kanadı üzerinde iki boyutlu sıcaklık dağılımının çıkartılması
Two dimensional temperature distribution on gas turbine blade
MURAT ÇAKAN
- Kare piramit yapılı modüler çok işlemcili bir gerçek zaman sistemi
A Modular real-time multi-processor system with square pyramid structure
B.TEVFİK AKGÜN
Doktora
Türkçe
1991
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrolİstanbul Teknik ÜniversitesiPROF.DR. A. EMRE HARMANCI