Chaolu Feng

ISLES 2015 - A public evaluation benchmark for ischemic stroke lesion segmentation from multispectral MRI

&NA; Ischemic stroke is the most common cerebrovascular disease, and its diagnosis, treatment, and study relies on non‐invasive imaging. Algorithms for stroke lesion segmentation from magnetic resonance imaging (MRI) volumes are intensely researched, but the reported results are largely incomparable due to different datasets and evaluation schemes. We approached this urgent problem of comparability with the Ischemic Stroke Lesion Segmentation (ISLES) challenge organized in conjunction with the MICCAI 2015 conference. In this paper we propose a common evaluation framework, describe the publicly available datasets, and present the results of the two sub‐challenges: Sub‐Acute Stroke Lesion Segmentation (SISS) and Stroke Perfusion Estimation (SPES). A total of 16 research groups participated with a wide range of state‐of‐the‐art automatic segmentation algorithms. A thorough analysis of the obtained data enables a critical evaluation of the current state‐of‐the‐art, recommendations for further developments, and the identification of remaining challenges. The segmentation of acute perfusion lesions addressed in SPES was found to be feasible. However, algorithms applied to sub‐acute lesion segmentation in SISS still lack accuracy. Overall, no algorithmic characteristic of any method was found to perform superior to the others. Instead, the characteristics of stroke lesion appearances, their evolution, and the observed challenges should be studied in detail. The annotated ISLES image datasets continue to be publicly available through an online evaluation system to serve as an ongoing benchmarking resource (www.isles‐challenge.org). HighlightsEvaluation framework for automatic stroke lesion segmentation from MRIPublic multi‐center, multi‐vendor, multi‐protocol databases releasedOngoing fair and automated benchmark with expert created ground truth setsComparison of 14+7 groups who responded to an open challenge in MICCAISegmentation feasible in acute and unsolved in sub‐acute cases Graphical abstract Figure. No caption available.

Segmentation of the Left Ventricle Using Distance Regularized Two-Layer Level Set Approach

We propose a novel two-layer level set approach for segmentation of the left ventricle (LV) from cardiac magnetic resonance (CMR) short-axis images. In our method, endocardium and epicardium are represented by two specified level contours of a level set function. Segmentation of the LV is formulated as a problem of optimizing the level set function such that these two level contours best fit the epicardium and endocardium. More importantly, a distance regularization (DR) constraint on the level contours is introduced to preserve smoothly varying distance between them. This DR constraint leads to a desirable interaction between the level contours that contributes to maintain the anatomical geometry of the endocardium and epicardium. The negative influence of intensity inhomogeneities on image segmentation are overcome by using a data term derived from a local intensity clustering property. Our method is quantitatively validated by experiments on the datasets for the MICCAI grand challenge on left ventricular segmentation, which demonstrates the advantages of our method in terms of segmentation accuracy and consistency with anatomical geometry.

Non-locally regularized segmentation of multiple sclerosis lesion from multi-channel MRI data

Segmentation of multiple sclerosis (MS) lesion is important for many neuroimaging studies. In this paper, we propose a novel algorithm for automatic segmentation of MS lesions from multi-channel MR images (T1W, T2W and FLAIR images). The proposed method is an extension of Li et al.s algorithm in [1], which only segments the normal tissues from T1W images. The proposed method is aimed to segment MS lesions, while normal tissues are also segmented and bias field is estimated to handle intensity inhomogeneities in the images. Another contribution of this paper is the introduction of a nonlocal means technique to achieve spatially regularized segmentation, which overcomes the influence of noise. Experimental results have demonstrated the effectiveness and advantages of the proposed algorithm.

Segmentation of longitudinal brain MR images using bias correction embedded fuzzy c-means with non-locally spatio-temporal regularization

An automated method is proposed to segment brain tissues in longitudinal MR images.The method has an inherent mechanism to deal with intensity inhomogeneities.Spatio-temporal regularization is used to ensure segmentation consistency.Results are consistent in segmenting an arbitrary number of image series. We propose an automated method for segmentation of brain tissues in longitudinal MR images. In the proposed method, images acquired at each time point are first separately segmented into white matter, gray matter, and cerebrospinal fluid by bias correction embedded fuzzy c-means. Intensities differences are then defined as similarities of each voxel to the cluster centroids. After being normalized in inter-class, the similarities are incorporated into a non-local means de-noising formula to regularize the segmentation in both spatial and temporal dimensions. Non-locally regularization results are used to compute final membership functions for the segmentation. To improve time performance, we accelerate the modified de-noising algorithm using CUDA and obtain a 200 × performance improvement. Quantitative comparison with the state-of-the-art methods on BrainWeb dataset demonstrate advantages of the proposed method in terms of segmentation accuracy and the ability to consistently segment brain tissues in an arbitrary number of longitudinal brain MR image series.

Image segmentation and bias correction using local inhomogeneous iNtensity clustering (LINC)

Image segmentation is still an open problem due to the existing of intensity inhomogeneity and noise. To accurately segment images with these biases, a local inhomogeneous intensity clustering (LINC) model is proposed. In LINC, a linear combination of a given set of smooth orthogonal basis functions is used to estimate the bias field. A local clustering criterion function is first defined to cluster the nearly homogeneous intensities in a relatively small neighborhood of each pixel. An energy functional is then defined by integrating the function with respect to the neighborhood center. This energy together with a regularization term and an arc length term are incorporated into a variational level set formulation in which de-nosing is implicitly included due to the implied convolution. Image segmentation and bias correction can be simultaneously achieved by updating variables of the final energy functional iteratively till it is stable or a predetermined iteration number is reached. The proposed model LINC has been extensively tested on both synthetic and real images. Experimental results and comparison with state-of-the-art methods demonstrate the advantages of the proposed model in terms of segmentation accuracy, bias field correction, dealing with noise, and robustness to initialization. HighlightsA level set method is proposed for image segmentation and bias correction.The bias field is simulated by linearly combining a given set of basis functions.The proposed method is able to segment images with intensity inhomogeneity.The proposed method is also able to correct the bias field from the image.Experiments demonstrate that our method is also robust to noise and initialization.

Image segmentation using CUDA accelerated non-local means denoising and bias correction embedded fuzzy c-means (BCEFCM)

Due to intensity overlaps between interested objects caused by noise and intensity inhomogeneity, image segmentation is still an open problem. In this paper, we propose a framework to segment images in the well-known image model in which intensities of the observed image are viewed as a product of the true image and the bias field. In the proposed framework, a CUDA accelerated non-local means denoising method is first used to remove noise from the image. Then, a bias correction embedded fuzzy c-means (BCEFCM) method is proposed to segment the image and correct the bias field simultaneously. To ensure the slowly and smoothly varying property of the bias field, we convolve it with a normalized kernel as soon as it is updated in each iteration. The proposed framework has been extensively tested on both selected synthetic and real images and public BrainWeb and IBSR datasets. Experimental results and comparison analysis demonstrate that the proposed framework is not only able to deal with noise and correct the bias field but it is also faster and more accurate than state-of-the-art methods. Graphical abstractDisplay Omitted HighlightsA framework is proposed for segmenting images with noise and bias field.The noise is first removed by a CUDA accelerated non-local means denoising method.The noise removed image is then segmented by bias correction embedded FCM.The framework is able to segment images with noise and correct the bias field.The framework is faster and more accurate than state-of-the-art methods.

Segmentation of Ischemic Stroke Lesions in Multi-spectral MR Images Using Weighting Suppressed FCM and Three Phase Level Set

Accurate segmentation of ischemic lesions is still a challenging task. In this paper, we propose a framework to extract ischemic lesions from multi-spectral MR images. In the proposed framework, MR images of each modality are first segmented into brain tissues and ischemic lesions by weighting suppressed fuzzy c-means. Preliminary lesion segmentation results are then fused among all the imaging modalities by majority voting. The fused segmentation results are finally refined by a three phase level set method. The level set formulation is defined on multi-spectral images with the capability of dealing with intensity inhomogeneities. The proposed framework has been applied to the MICCAI 2015 ISLES challenge. According to the ranking rules of the challenge, the proposed framework took the second place and the fourth place in sub-acute lesion segmentation and acute stroke estimation, respectively.

Simultaneous extraction of endocardial and epicardial contours of the left ventricle by distance regularized level sets

PURPOSE Segmentation of the cardiac left ventricle (LV) is still an open problem and is challenging due to the poor contrast between tissues around the epicardium and image artifacts. To extract the endocardium and epicardium of the cardiac left ventricle accurately, the authors propose a two-layer level set approach for segmentation of the LV from cardiac magnetic resonance short-axis images. METHODS In the proposed method, the endocardium and epicardium are represented by two specified level contours of a level set function. Segmentation of the LV is formulated as a problem of optimizing the level set function such that these two level contours best fit the epicardium and endocardium, subject to a distance regularization (DR) term to preserve a smoothly varying distance between them. The DR term introduces a desirable interaction between the two level contours of a single level set function, which contributes to preserve the anatomical geometry of the epicardium and endocardium of the LV. In addition, the proposed method has an intrinsic ability to deal with intensity inhomogeneity in MR images, which is a common image artifact in MRI. RESULTS Their method is quantitatively validated by experiments on the datasets for the MICCAI 2009 grand challenge on left ventricular segmentation and the MICCAI 2013 challenge workshop on segmentation: algorithms, theory and applications (SATA). To overcome discontinuity of 2D segmentation results at some adjacent slices for a few cases, the authors extend distance regularized two-layer level set to 3D to refine the segmentation results. The corresponding metrics for their method are better than the methods in the MICCAI 2009 challenge. Their method was ranked at the first place in terms of Hausdorff distance and the second place in terms of Dice similarity coefficient in the MICCAI 2013 challenge. CONCLUSIONS Experimental results demonstrate the advantages of their method in terms of segmentation accuracy and consistency with the heart anatomy.

CUDA accelerated uniform re-sampling for non-Cartesian MR reconstruction.

A grid-driven gridding (GDG) method is proposed to uniformly re-sample non-Cartesian raw data acquired in PROPELLER, in which a trajectory window for each Cartesian grid is first computed. The intensity of the reconstructed image at this grid is the weighted average of raw data in this window. Taking consider of the single instruction multiple data (SIMD) property of the proposed GDG, a CUDA accelerated method is then proposed to improve the performance of the proposed GDG. Two groups of raw data sampled by PROPELLER in two resolutions are reconstructed by the proposed method. To balance computation resources of the GPU and obtain the best performance improvement, four thread-block strategies are adopted. Experimental results demonstrate that although the proposed GDG is more time consuming than traditional DDG, the CUDA accelerated GDG is almost 10 times faster than traditional DDG.

CUDA accelerated method for motion correction in MR PROPELLER imaging

In PROPELLER, raw data are collected in N strips, each locating at the center of k-space and consisting of Mx sampling points in frequency encoding direction and L lines in phase encoding direction. Phase correction, rotation correction, and translation correction are used to remove artifacts caused by physiological motion and physical movement, but their time complexities reach O(Mx×Mx×L×N), O(N×RA×Mx×L×(Mx×L+RN×RN)), and O(N×(RN×RN+Mx×L)) where RN×RN is the coordinate space each strip gridded onto and RA denotes the rotation range. A CUDA accelerated method is proposed in this paper to improve their performances. Although our method is implemented on a general PC with Geforce 8800GT and Intel Core(TM)2 E6550 2.33GHz, it can directly run on more modern GPUs and achieve a greater speedup ratio without being changed. Experiments demonstrate that (1) our CUDA accelerated phase correction achieves exactly the same result with the non-accelerated implementation, (2) the results of our CUDA accelerated rotation correction and translation correction have only slight differences with those of their non-accelerated implementation, (3) images reconstructed from the motion correction results of CUDA accelerated methods proposed in this paper satisfy the clinical requirements, and (4) the speed up ratio is close to 6.5.