Han J. W. van Triest

PhenOMIM: An OMIM-based secondary database purported for phenotypic comparison

Phenotypic comparison may provide crucial information for obtaining insights into molecular interactions underlying various diseases. However, few attempts have been made to systematically analyze the phenotypes of hereditary disorders, mainly owing to the poor quality of text descriptions and lack of a unified system of descriptors. Here we present a secondary database, PHENOMIM, for translating the phenotypic data obtained from the Online Mendelian Inheritance in Man (OMIM) database into a structured form. Moreover, a web interface has also been developed for visualizing the data and related information from the OMIM and PhenOMIM databases. The data is freely available online for reviewing and commenting purposes and can be found at http://faculty.neu.edu.cn/bmie/han/PhenOMIM/.

SIMULATION ANALYSIS OF DEFORMATION AND STRESS OF TRACHEAL AND MAIN BROCHIAL WALL FOR SUBJECTS WITH LEFT PULMONARY ARTERY SLING

Left pulmonary artery sling (LPAS) is a kind of severe congenital anomaly, where the stenoses usually occur at trachea and main bronchi for the external compression of the artery sling. Computed tomography (CT) images can provide accurate morphological analysis, but the airflow and its effects on the airway wall are unknown and seldom investigated. In the present study, a uni-directional coupling fluid–structure interaction (UCFSI) method is employed to simulate the deformation and stress of tracheal and main bronchial wall for four LPAS subjects and one health control. Much higher airflow velocity is observed for LPAS subjects due to the stenosis, and the deformation and equivalent stress of airway wall are about 50–900 and 90–1000 times of the health control, respectively. The direction of tracheal shift may be related to the airway shape, and is opposite to the net reaction force. The influences of inlet flow velocity and wall thickness on the deformation and stress are significant and their relationsh...

Automatic 3D segmentation of human airway tree in CT image

With the dawn of modern imaging technologies such as CT, MRI and PET, images play a role of ever increasing importance. Due to the high contrast between air and tissue, X-rays are the imaging modality of choice. Especially CT-scans are increasingly useful for diagnosis of disorders of the lung. Working with the acquired 3D CT data brings new difficulties as it is not trivial to display 3D data on a 2D monitor. One way to display this information is by reconstructing the structures and applying volume rendering on the segmented volumes. In this paper a novel method is presented for the segmentation of the airway tree. The proposed algorithm employs region growing, 3D wave propagation and morphological refinement to segment bronchi. The algorithm has been tested on 24 datasets resulting in airway trees that are successfully segmented up to the sixth generation, while execution times are as low as 2 seconds per airway tree.

Automatic detection of neovascularization in retinal images using extreme learning machine

Diabetic Retinopathy is one complication of diabetes, which can cause blindness. Diabetic retinopathy can be divided into Non-Proliferative Diabetic Retinopathy (NPDR) and Proliferative Diabetic Retinopahy (PDR), and neovascularization is a key symbol to make diagnosis between them. An automatic detection of neovascularization in retinal images using extreme learning machine is proposed. Furthermore, we use a series of filter banks to get the features of neovascularization from retinal images. The detection framework is evaluated with images annotated by expert ophthalmologists based on the images selected from several public retinal image databases. The experimental results illustrate that the framework can mark and show the suspected neovascularization regions to ophthalmologists, and thus support for their decision making.

Arterial radius estimation from microscopic data using a new algorithm for circle parameter estimation

We develop and test a new method for automatic determination of vessel wall diameters from image stacks obtained using two-photon laser scanning microscopy (TPLSM) on viable arteries in perfusion flow chambers. To this extent, a new method is proposed for estimating the parameters of a circle describing the inner diameter of the blood vessels. The new method is based on the Hough transform and the observation that three points that are not colinear uniquely define a circle. By only storing the estimated center location, the computational and memory costs of the Hough transform can be greatly reduced. We test the algorithm on 20 images and compare the result with a ground-truth established by human volunteers and a standard least-squares technique. With errors of 3 to 5%, the algorithm enables accurate estimation of the vessel diameters from image stacks containing only small parts of the vessel cross section. Combined with TPLSM imaging of anatomical vessel wall properties, potentially, the algorithm enables the correlation of structural and functional properties of large intact arteries simultaneously, without requirements for additional experiments.

Two novel methods for juxta-pleural nodule segmentation based on CT images

The shape, size and growth rate of lung nodules are the most important indicators for the malignancy of a lung cancer and the basis for assessment of lung cancer treatment effect. Therefore, accurate segmentation of the lung nodules is of great significance for the diagnosis and treatment of the lung cancers. In this paper, two novel methods are proposed to extract juxta-pleural nodules in CT image data for subsequent volume assessment. The algorithm takes the form of user interaction process, such as the selection of the seed point and the adjustment of the volume of interest, which can make best use of the knowledge of the radiologists. The first method combining contour finding and arc chord ratio thresholding and the second method combining the ray casting and line fitting are both designed for segmentation of the juxta-pleural nodules. The algorithm is tested on datasets from 39 patients with a total of 53 juxta-pleural nodules. Evaluated by the senior radiologists, the two methods both gained satisfactory results with segmentation accuracy exceeding 90% on average. It shows the algorithm is helpful for the segmentation, volume measurements and evaluation of juxta-pleural nodules.

Automatic segmentation of juxta-pleural tumors from CT images based on morphological feature analysis.

Extraction of lung tumors is a fundamental step for further quantitative analysis of the tumor, but is challenging for juxta-pleural tumors due to the adhesion to the pleurae. An automatic algorithm for segmentation of juxta-pleural tumors based on the analysis of the geometric and morphological features was proposed. Initially, the lung is extracted by means of thresholding using 2D Otsus method. Next a center point is suggested to find a starting point and endpoint of outward facing pleura. A model based on the variation of incline angle was adopted to identify potentially affected regions, and to full segment juxta-pleural tumors. The results were compared with the manual segmentation by two radiologists. Averaged for ten experimental datasets, the accuracy calculated by Dice index between the results of the algorithm and by the two radiologists is 91.2%. It indicates the proposed method has comparable accuracy with the experts (the inter-observer variability is 92.4%), but requests much less manual interactions. The proposed algorithm can be used for segmenting juxta-pleural tumors from CT images, and help improve the diagnosis, pre-operative planning and therapy response evaluation.

OMIM data - standardized input and pair-wise comparison

The Online Mendelian Inheritance in Man (OMIM) database is a useful reference for human genetic diseases. By standardization of its data structure, we hereby provide a user-friendly interface purported for standardized input of new patient data. Automatic diagnosis and pathway discoveries maybe achieved through pair-wise comparison and association analysis of tree data. A website has been provided at http://faculty.neu.edu.cn/bmie/han/PhenOMIM/. To illustrate the usefulness of this system, we have explored potential genes which may interact with a MLL2 gene which was recently found to underlie Kakubi syndrome. Furthermore, clinical features of a new syndrome identified from the literature have been input and compared with the PhenOMIM list of syndromes, by which the usefulness of this work was assessed, and a differential diagnosis was made. We expect this software will be of substantial value for doctors as well as researchers in the field of genetic diseases.

The analysis and development of computed 2D optical flow velocity based on phantom data

Optical Flows is the study of the relationship with image intensity changes in time and the motion of objects. In this paper, the gradient-based optical flow method is studied to determine the velocity of observed objects and the parameters which can influence the accuracy of the velocity are identified and discussed. According to the characteristics of this Algorithm, an improved iterative method is proposed to let the method be more flexible to ground truth data. To understand the principles behind this phenomenon, we create a phantom sequence to analyze and experiment the algorithm from both mathematical and geometrical aspect. The precision of our method can reach an excellent level with a fast computing speed. The experiments on the phantom images showed that the method work quickly and accurately.