Quantitative assessment of the Cerebral Vascular System 

Quantitative MRI has numerous applications for diagnosis, follow-up and therapy response assessment. Many parameters provide indirect information regarding physical, physiological and microstructural properties of the tissue.

 

However, accurate quantification is often dependent on the subject's physiology, as well as acquisition and analysis methods. In our research we develop acquisition and analysis methods to extract accurate and reproducible MR parameters.

T1 Mapping using variable flip angle SPGR data with flip angle correction

Classification of tumor area using combined DCE and DSC MRI in patients with glioblastoma

Human cerebral blood volume measurements using dynamic contrast enhancement in comparison to dynamic susceptibility contrast MRI

Differentiation between Progressive Disease and Treatment Necrosis in Patients with High Grade Brain Tumors using DCE MRI

Territorial segmentation and flow measurement

The ability to examine the branches and perfusion territory of a specific artery is crucial in the diagnosis and evaluation of many cerebrovascular diseases. Territorial imaging can be used to identify the presence and source of collateral flow to a brain area at risk of ischemia, assess stenosis severity, differentiate stroke types and assess the success of surgical interventions.

 

Currently, territorial imaging using MRI is limited due to special acquisition sequence requirements.

In this study we developed a post-processing method for territorial segmentation based on routine MRI Time-resolved Angiography with Interleaved Stochastic Trajectories (TWIST) data. In addition, the method extracts the volumetric blood flow rate in each territory, providing a quantitative measure. This was replicated both within and between subjects (Figure 1).

 

The sensitivity of the method to pathological conditions including Moyamoya and patients with carotid stenosis is under investigation. Figure 2 shows an example of a patient with Moyamoya, before and after surgery (Figure 2).

As can be seen, following surgery there is an increase in the collateral blood flow to the hypoperfused areas by the external carotid.

Automatic volumetric MRI measurements of the fetal brain and its structures

Accurate volumetric measurement of the fetal brain and its structures is of great importance for the assessment of fetal growth and early identification of various developmental disorders, such as Intra Uterine Growth Restriction (IUGR).

 

MRI is increasingly being used in fetal imaging with many advantages over the commonly used Ultrasound. However, volume is mainly estimated with linear measurements and advanced analysis methods for volumetric measurements from MRI are currently limited, often requiring the acquisition of unique data.

 

Together with Prof Leo Joskowicz’s group from the Hebrew University, we have developed and implemented a semi-automatic segmentation method based on a seeded region growing algorithm for volumetric measurements of the fetal brain from MRI scans.

 

The developed method is simple, fast, accurate, reproducible, user independent, and is easily applicable with retrospective data. We used this method to create, for the first time, a normal volumetric growth chart of the fetal brain based on 200 fetuses. 

The sensitivity of the method was demonstrated on fetuses with IUGR. Currently we are working on developing additional methods to segment other structures and the entire fetal body.

A normal growth chart based on volumetric data of 200 normal fetal brains (black dots). Data from IUGR fetuses are shown in red triangles

DUSTER - DCE Up-Sampled Temporal Resolution

Territorial segmentation and flow measurement

Automatic volumetric MRI measurements of the fetal brain

and its structures

 
 
DUSTER - DCE Up-Sampled Temporal Resolution

Dynamic contrast enhanced (DCE) imaging was initially proposed for the assessment of tissue permeability, and has been widely used for the characterization of tumor biology and therapy response assessment; it was additionally suggested as an endpoint for early-stage assessment of antiangiogenic and antivascular therapy response. DUSTER (DCE Up-Sampled Temporal Resolution)1 was developed in our lab and enables fully automatic analysis of DCE data using the Extended Tofts Model, incorporating correction for bolus arrival time.

 

DUSTER produces parametric maps of Ktrans, Kep, vp, ve and BAT using model selection. Analysis includes calculation of baseline T1 maps using DESPOT1, with FAs correction, motion correction, brain extraction, removal and compensation of noisy time-points, raw-signal-to-T1-to-CTC conversion, B1 inhomogeneity correction, artery localization, AIF extraction at temporal super- resolution and model fitting with model selection2.

DUSTER has been successfully applied to more than 300 cases, and has been used in several studies by our group3-5.The tool runs on Linux/Windows and has been demonstrated on data obtained from GE, Philips and Siemens.

 

The program is designed as an open-source software, and is available upon request. It provides a user-friendly interface for exploration at all levels of analysis. Currently, we are working on incorporating ACoPeD6 (AIF-corrected-perfusion-DCE-MRI) within this tool, which will include flow in the model.

Key Publications

DUSTER: Dynamic contrast enhance up-sampled temporal resolution analysis method

Optimization of Two-Compartment-Exchange-Model Analysis for Dynamic Contrast-Enhanced MRI Incorporating Bolus Arrival Time

 
 

Prof. DAFNA  BEN-BASHAT (PhD)

Deputy Director of the Sagol Brain Institute, Senior Researcher at the Sackler Faculty of Medicine & Sagol School of Neuroscience at Tel Aviv University.

Visiting Professor at the Northern Jiangsu People's Hospital Clinical Medical School of Yangzhou University.

Prof. Ben-bashat is the Leading Investigator of the #Advanced Brain Imaging research team at the Sagol Brain Institute.