Skip to content. | Skip to navigation

Personal tools

 

BI_Logo          BioImgLab          BI_Logo

thinline copy.gifthinline copy.gifthinline copy.gif

|  Dr. Ayman El-Baz | People | Projects | Publications | Education | Facilities | Lab Awards | Gallery | Alumni | Thesis| Suppl._Materials |  MICCAI Tutorial |   Books |

 


______________________________________________________________________________________________
Novel Image Analysis Framework for Early Diagnosis of Malignant Lung Nodules
______________________________________________________________________________________________



 
 AlBazMov3.gif         

Early diagnosis of lung cancer is critical to improving chances of survival. The five-year survival rate is nearly 50% if it is found at localized state and can reach 85% if it is diagnosed at early stage and surgery is possible. Once the cancer has spread to other organs, the survival rates decline dramatically (20% at regional stage and 2.2% at distant stage). Nevertheless, only 15% of lung cancer cases are found at the localized early stage. A pulmonary nodule is the most common manifestation of lung cancer. Lung nodules are approximately spherical regions of relatively higher density that are visible in X-ray images of the lung. The detection and measurement of the growth rate of pulmonary nodules are important for early diagnosis of malignancy. Large (generally defined as greater than 1 cm in diameter) malignant nodules can be easily detected with traditional imaging equipment and can be diagnosed by needle biopsy or bronchoscopy techniques; whereas, the diagnostic options for small malignant nodules are limited due to problems associated with accessing small tumors, especially if they are located deep in the tissue or faraway from the large airways, as well as the need for additional diagnostic and imaging techniques. One of the most promising techniques for detecting small cancerous nodules relies on characterizing the nodule based on its growth rate. The goal of this project is to develop and clinically validate a new image analysis approach for automatic measuring of the growth rate of the detected lung nodules to aid in early diagnosis of malignant tumors.

 

 


______________________________________________________________________________________________
A New CAD System for Automatic Diagnosis of Autistic Brains
______________________________________________________________________________________________



 
 AlBazMov2.gif         

Autism is a developmental disorder characterized by social deficits, impaired communication, and restricted and repetitive patterns of behavior. Recent neuropathological studies of autism have revealed abnormal anatomy of the cerebral white matter (CWM) in autistic brains. This project proposes a novel approach to classify autistic from normal subjects based on a analyzing the shape of cerebral white matter gyrifications for both normal and autistic subjects. The proposed shape analysis technique consists of three main steps. The first step is to segment cerebral white matter from proton density MRI images using currently and priorly learned visual appearance models for the 3D cerebral white matter in order to control the evolution of deformable boundaries. The appearance prior is modeled with a translation and rotation invariant Markov-Gibbs random field (MGRF) of voxel intensities with a pairwise interaction model. Appearance of the cerebral white matter and their background in current multi-modal proton density MRI images is also represented with a marginal probability distribution of voxel intensities. The cerebral white matter appearance model is isolated from the mixed distribution using its close approximation with a linear combination of discrete Gaussians (LCDGs). The second step is to extract the gyrifications of cerebral white matter from the segmented cerebral white matter. The last step is to perform shape analysis to quantify the thickness of the extracted cerebral white matter gyrifications for both autistic and normal subjects. The preliminary results of the proposed image analysis have yielded promising results that would, in the near future, supplement the use of current technologies for diagnosing. autism.


______________________________________________________________________________________________
A New CAD System for the Evaluation of Kidney Diseases Using DCE-MRI
______________________________________________________________________________________________



 
 AlBazMov1.gif         

Acute rejection is the most common reason for graft failure after kidney transplantation, and early detection is crucial to survival of function in the transplanted kidney. In this research project, we introduce a new framework for automatic classification of normal and acute rejection transplants from Dynamic Contrast Enhanced Magnetic Resonance Images (DCE-MRI). The proposed framework consists of three main steps. The first isolates the kidney from the surrounding anatomical structures by evolving a deformable model based on two density functions; the first function describes the distribution of the gray level inside and outside the kidney region and the second describes the prior shape of the kidney. In the second step, non-rigid registration algorithms are employed to account for the motion of the kidney due to the patient’s breathing. In the third step, the perfusion curves that show transportation of the contrast agent into the tissue are obtained from the segmented cortex of the whole image sequence of the patient. In the final step, we collect four features from these curves and use Bayesian classifiers to distinguish between acute rejection and normal transplants. Applications of the proposed approach yield promising results that would, in the near future, replace the use of current technologies such as nuclear imaging and ultrasonography, which are not specific enough to determine the type of kidney dysfunction.


______________________________________________________________________________________________
Cerebrovascular Segmentation by Accurate Probabilistic Modeling of MRA Images
______________________________________________________________________________________________



 
 ElBazMov4.gif         

We present a fast algorithm for automatic extraction of a 3D cerebrovascular system from magnetic resonance angiography (MRA) data. Blood vessels are separated from background tissues (fat, bones, or grey and white brain matter) by voxel-wise classification based on precise approximation of a multi-modal empirical marginal intensity distribution of the MRA data. The approximation involves a linear combination of discrete Gaussians (LCDG) with alternating signs, and we modify the conventional Expectation-Maximization (EM) algorithm to deal with the LCDG. To validate the accuracy of our algorithm, a special 3D geometrical phantom motivated by statistical analysis of the MRA data is designed. Experiments with both the phantom and 50 real data sets confirm high accuracy of the proposed approach.

Back to the Lab

 

Home

Document Actions
Personal tools