Chao Jin

Quantitative analysis of retinal layer optical intensities on three-dimensional optical coherence tomography.

PURPOSEnTo investigate the optical intensities of all retinal layers on three-dimensional (3D) spectral-domain optical coherence tomography (SD-OCT) in normal subjects using an automatic measurement.nnnMETHODSnForty normal subjects underwent Topcon 3D OCT-1000 macula-centered scan. The raw data were automatically segmented into 10 layers using the 3D graph search approach. Then the mean and standard deviation of intensities of each layer were calculated. The image quality index was given by the OCT software. Correlation analysis was performed between the optical intensities in each layer and image quality and subjects age.nnnRESULTSnThe correlation of optical intensities was strong from ganglion cell layer (GCL) to outer nuclear layer (ONL) with r > 0.934; moderate among retinal nerve fiber layer (RNFL), photoreceptor, retinal pigment epithelium (RPE), and choroid (0.410 < r < 0.800); and low in the vitreous (0.251 < r < 0.541). The optical intensities were also correlated with the image quality, r > 0.869 from GCL to ONL, 0.748 < r < 0.802 for RNFL, photoreceptor layer, RPE, and the choroid, r = 0.528 for the vitreous. The optical intensity in RNFL was negatively correlated with age (r = -0.365).nnnCONCLUSIONSnAutomatic assessment of the layers intensities was achieved. In normal subjects, the retinal layers optical intensities were affected by image quality. Normalization with optical intensity of ONL, all areas, or image quality index is recommended. The optical intensity of RNFL decreased with age.

3D Fast Automatic Segmentation of Kidney Based on Modified AAM and Random Forest

In this paper, a fully automatic method is proposed to segment the kidney into multiple components: renal cortex, renal column, renal medulla and renal pelvis, in clinical 3D CT abdominal images. The proposed fast automatic segmentation method of kidney consists of two main parts: localization of renal cortex and segmentation of kidney components. In the localization of renal cortex phase, a method which fully combines 3D Generalized Hough Transform (GHT) and 3D Active Appearance Models (AAM) is applied to localize the renal cortex. In the segmentation of kidney components phase, a modified Random Forests (RF) method is proposed to segment the kidney into four components based on the result from localization phase. During the implementation, a multithreading technology is applied to speed up the segmentation process. The proposed method was evaluated on a clinical abdomen CT data set, including 37 contrast-enhanced volume data using leave-one-out strategy. The overall true-positive volume fraction and false-positive volume fraction were 93.15%, 0.37% for renal cortex segmentation; 83.09%, 0.97% for renal column segmentation; 81.92%, 0.55% for renal medulla segmentation; and 80.28%, 0.30% for renal pelvis segmentation, respectively. The average computational time of segmenting kidney into four components took 20 seconds.

Fast Renal Cortex Localization by Combining Generalized Hough Transform and Active Appearance Models

Automatic localization of objects is one of great important steps in object recognition and analysis, such as segmentation, registration in many medical applications. In this paper, an automated method is proposed to recognize renal cortex on contrast-enhanced abdominal CT images. The proposed method is based on a strategic combination of the Generalized Hough Transform and Active Appearance Model. It consists of two main phases: training and localization. In the training phase, we train the mean shape models of renal cortex by using Active Appearance Model and compute Generalized Hough Transform parameters. In the localization phase, a modified Generalized Hough Transform algorithm is advanced to estimate potential center of gravity for improving the conventional Active Appearance Model matching method, and then a two-pass Active Appearance Model matching method is proposed based on Generalized Hough Transform. The Active Appearance Models and Generalized Hough Transform parameters were trained with 20 CT angiography datasets, and then the proposed method was tested on a clinical data set of 17 CT angiography datasets. The experimental results show that: 1 an overall cortex localization accuracy is 0.9920±0.0038, average distance is 11.00±9.34 pixels. 2 The proposed method is highly efficient such that the overall localization can be finalized within 1.2075±0.3738 seconds for each 2D slice.

Renal cortex localization by combining 3D Generalized Hough Transform and 3D Active Appearance Models

Automatic localization is one of important steps in medical image segmentation. In this paper, a model-based method for three-dimensional image localization is developed. Our method is based on a combination of 3D Generalized Hough Transform and 3D Active Appearance Models. It consists of two main parts: training and localization. The proposed method was tested on a clinical abdomen CT data set, including 27 contrast-enhanced volume data, in which 15 were chose as training data while the other 12 as testing data. The experimental results show that: (1) an overall cortex localization average distance is 12.58±3.26 voxels. (2) The proposed method is highly efficient, the running time is about only 35.70±3.62 seconds for each volume data.

Support vector machine based IS/OS disruption detection from SD-OCT images

In this paper, we sought to find a method to detect the Inner Segment /Outer Segment (IS/OS)disruption region automatically. A novel support vector machine (SVM) based method was proposed for IS/OS disruption detection. The method includes two parts: training and testing. During the training phase, 7 features from the region around the fovea are calculated. Support vector machine (SVM) is utilized as the classification method. In the testing phase, the training model derived is utilized to classify the disruption and non-disruption region of the IS/OS, and calculate the accuracy separately. The proposed method was tested on 9 patients SD-OCT images using leave-one-out strategy. The preliminary results demonstrated the feasibility and efficiency of the proposed method.

OIPAV: an integrated software system for ophthalmic image processing, analysis and visualization

OIPAV (Ophthalmic Images Processing, Analysis and Visualization) is a cross-platform software which is specially oriented to ophthalmic images. It provides a wide range of functionalities including data I/O, image processing, interaction, ophthalmic diseases detection, data analysis and visualization to help researchers and clinicians deal with various ophthalmic images such as optical coherence tomography (OCT) images and color photo of fundus, etc. It enables users to easily access to different ophthalmic image data manufactured from different imaging devices, facilitate workflows of processing ophthalmic images and improve quantitative evaluations. In this paper, we will present the system design and functional modules of the platform and demonstrate various applications. With a satisfying function scalability and expandability, we believe that the software can be widely applied in ophthalmology field.

OIPAV: an Integrated Software System for Ophthalmic Image Processing, Analysis, and Visualization

Ophthalmic medical images, such as optical coherence tomography (OCT) images and color photo of fundus, provide valuable information for clinical diagnosis and treatment of ophthalmic diseases. In this paper, we introduce a software system specially oriented to ophthalmic images processing, analysis, and visualization (OIPAV) to assist users. OIPAV is a cross-platform system built on a set of powerful and widely used toolkit libraries. Based on the plugin mechanism, the system has an extensible framework. It provides rich functionalities including data I/O, image processing, interaction, ophthalmic diseases detection, data analysis, and visualization. By using OIPAV, users can easily access to the ophthalmic image data manufactured from different imaging devices, facilitate workflows of processing ophthalmic images, and improve quantitative evaluations. With a satisfying function scalability and expandability, the software is applicable for both ophthalmic researchers and clinicians.

Graph search: active appearance model based automated segmentation of retinal layers for optic nerve head centered OCT images

In this paper, a novel approach combining the active appearance model (AAM) and graph search is proposed to segment retinal layers for optic nerve head(ONH) centered optical coherence tomography(OCT) images. The method includes two parts: preprocessing and layer segmentation. During the preprocessing phase, images is first filtered for denoising, then the B-scans are flattened. During layer segmentation, the AAM is first used to obtain the coarse segmentation results. Then a multi-resolution GS–AAM algorithm is applied to further refine the results, in which AAM is efficiently integrated into the graph search segmentation process. The proposed method was tested on a dataset which contained113-D SD-OCT images, and compared to the manual tracings of two observers on all the volumetric scans. The overall mean border positioning error for layer segmentation was found to be 7.09 ± 6.18μm for normal subjects. It was comparable to the results of traditional graph search method (8.03±10.47μm) and mean inter-observer variability (6.35±6.93μm).The preliminary results demonstrated the feasibility and efficiency of the proposed method.

Fast segmentation of kidney components using random forests and ferns

Purpose: This paper studies the feasibility of developing a fast and accurate automatic kidney component segmentation method. The proposed method segments the kidney into four components: renal cortex, renal column, renal medulla, and renal pelvis. Method: In this article, we have proposed a highly efficient approach which strategically combines random forests and random ferns methods to segment the kidney into four components: renal cortex, renal column, renal medulla, and renal pelvis. The proposed method is designed following a coarse‐to‐fine strategy. The initial segmentation applies random forests and random ferns with a variety of features, and combines their results to obtain a coarse renal cortex region. Then the fine segmentation of four kidney components is achieved using the weighted forests‐ferns approach with the well‐designed potential energy features which are calculated based on the initial segmentation result. The proposed method was validated on a dataset with 37 contrast‐enhanced CT images. Evaluation indices including Dice similarity coefficient (DSC), true positive volume fraction (TPVF), and false positive volume fraction (FPVF) are used to assess the segmentation accuracy. The proposed method was implemented and tested on a 64‐bit system computer (Intel Core i7‐3770 CPU, 3.4 GHz and 8 GB RAM). Results: The experimental results demonstrated the high accuracy and efficiency for segmenting the kidney components: the mean Dice similarity coefficients were 89.85%, 80.60%, 86.63%, and 77.75% for renal cortex, column, medulla, and pelvis, respectively, for right and left kidneys. The computational time of segmenting the whole kidney into four components was about 3 s. Conclusions: The experimental results showed the feasibility and efficacy of the proposed automatic kidney component segmentation method. The proposed method applied an efficient weighted strategy to combine random forests and ferns, making full use of the advantages of both methods. The novel potential energy features help random forests effectively segment the kidney components and the background. The high accuracy and efficiency of our method make it practicable in clinical applications.

CorteXpert: A model-based method for automatic renal cortex segmentation

HighlightsA precise clinical definition of renal cortex.A localization algorithm for the outer and the inner layers of the renal cortex.A purely delineation‐based algorithm, which is not only accurate but also extremely efficient.A non‐uniform graph search method is presented to obtain accurate segmentation. Graphical abstract Figure. No caption available. ABSTRACT This paper introduces a model‐based approach for a fully automatic delineation of kidney and cortex tissue from contrast‐enhanced abdominal CT scans. The proposed framework, named CorteXpert, consists of two new strategies for kidney tissue delineation: cortex model adaptation and non‐uniform graph search. CorteXpert was validated on a clinical data set of 58 CT scans using the cross‐validation evaluation strategy. The experimental results indicated the state‐of‐the‐art segmentation accuracies (as dice coefficient): 97.86% ± 2.41% and 97.48% ± 3.18% for kidney and renal cortex delineations, respectively.