Anatomy packing with hierarchical segments: an algorithm for segmentation of pulmonary nodules in CT images.

BACKGROUND: This paper proposes a semantic segmentation algorithm that provides the spatial distribution patterns of pulmonary ground-glass nodules with solid portions in computed tomography (CT) images. METHODS: The proposed segmentation algorithm, anatomy packing with hierarchical segments (APHS), performs pulmonary nodule segmentation and quantification in CT images. In particular, the APHS algorithm consists of two essential processes: hierarchical segmentation tree construction and anatomy packing. It constructs the hierarchical segmentation tree based on region attributes and local contour cues along the region boundaries. Each node of the tree corresponds to the soft boundary associated with a family of nested segmentations through different scales applied by a hierarchical segmentation operator that is used to decompose the image in a structurally coherent manner. The anatomy packing process detects and localizes individual object instances by optimizing a hierarchical conditional random field model. Ninety-two histopathologically confirmed pulmonary nodules were used to evaluate the performance of the proposed APHS algorithm. Further, a comparative study was conducted with two conventional multi-label image segmentation algorithms based on four assessment metrics: the modified Williams index, percentage statistic, overlapping ratio, and difference ratio. RESULTS: Under the same framework, the proposed APHS algorithm was applied to two clinical applications: multi-label segmentation of nodules with a solid portion and surrounding tissues and pulmonary nodule segmentation. The results obtained indicate that the APHS-generated boundaries are comparable to manual delineations with a modified Williams index of 1.013. Further, the resulting segmentation of the APHS algorithm is also better than that achieved by two conventional multi-label image segmentation algorithms. CONCLUSIONS: The proposed two-level hierarchical segmentation algorithm effectively labelled the pulmonary nodule and its surrounding anatomic structures in lung CT images. This suggests that the generated multi-label structures can potentially serve as the basis for developing related clinical applications.

A Heuristic Framework for Image Filtering and Segmentation: Application to Blood Vessel Immunohistochemistry.

The blood vessel density in a cancerous tissue sample may represent increased levels of tumor growth. However, identifying blood vessels in the histological (tissue) image is difficult and time-consuming and depends heavily on the observer's experience. To overcome this drawback, computer-aided image analysis frameworks have been investigated in order to boost object identification in histological images. We present a novel algorithm to automatically abstract the salient regions in blood vessel images. Experimental results show that the proposed framework is capable of deriving vessel boundaries that are comparable to those demarcated manually, even for vessel regions with weak contrast between the object boundaries and background clutter.

Automatic immunostaining vessel image filtering for visual search efficiency.

Blood vessel abstraction is an important procedure for quantitative analysis of blood vessel densities determined by immunostaining the tumor cells. Due to the weak contrast of object boundaries and background clutter, it is difficult to identify the vessel and non-vessel region clearly. In this paper we present a novel algorithm to automatically abstract salient regions in blood vessel images using Gaussian perceptually color space for producing the detail and large-scale layer. The first component of Gaussian color model is used to represent the large-scale layer after bilateral filtering. The detail layer of salient region in the blood vessel image is obtained by normalizing the color of the color layer in the image. Using these two features, we can reconstruct the image using intensity-color coupling. The abstraction result is then processed by luminance quantization algorithm to provide both boundary and region information of blood vessel images. This proposed algorithm has been applied on a wide range of complex blood vessel images with promising results.

ACCOMP: Augmented cell competition algorithm for breast lesion demarcation in sonography.

PURPOSE: Fully automatic and high-quality demarcation of sonographical breast lesions remains a far-reaching goal. This article aims to develop an image segmentation algorithm that provides quality delineation of breast lesions in sonography with a simple and friendly semiautomatic scheme. METHODS: A data-driven image segmentation algorithm, named as augmented cell competition (ACCOMP) algorithm, is developed to delineate breast lesion boundaries in ultrasound images. Inspired by visual perceptual experience and Gestalt principles, the ACCOMP algorithm is constituted of two major processes, i.e., cell competition and cell-based contour grouping. The cell competition process drives cells, i.e., the catchment basins generated by a two-pass watershed transformation, to merge and split into prominent components. A prominent component is defined as a relatively large and homogeneous region circumscribed by a perceivable boundary. Based on the prominent component tessellation, cell-based contour grouping process seeks the best closed subsets of edges in the prominent component structure as the desirable boundary candidates. Finally, five boundary candidates with respect to five devised boundary cost functions are suggested by the ACCOMP algorithm for user selection. To evaluate the efficacy of the ACCOMP algorithm on breast lesions with complicated echogenicity and shapes, 324 breast sonograms, including 199 benign and 125 malignant lesions, are adopted as testing data. The boundaries generated by the ACCOMP algorithm are compared to manual delineations, which were confirmed by four experienced medical doctors. Four assessment metrics, including the modified Williams index, percentage statistic, overlapping ratio, and difference ratio, are employed to see if the ACCOMP-generated boundaries are comparable to manual delineations. A comparative study is also conducted by implementing two pixel-based segmentation algorithms. The same four assessment metrics are employed to evaluate the boundaries generated by two conventional pixel-based algorithms based on the same set of manual delineations. RESULTS: The ACCOMP-generated boundaries are shown to be comparable to the manual delineations. Particularly, the modified Williams indices of the boundaries generated by the ACCOMP algorithm and the first and second pixel-based algorithms are 1.069 +/- 0.024, 0.935 +/- 0.024, and 0.579 +/- 0.013, respectively. If the modified Williams index is greater than or equal to 1, the average distance between the computer-generated boundaries and manual delineations is deemed to be comparable to that between the manual delineations. CONCLUSIONS: The boundaries derived by the ACCOMP algorithm are shown to reasonably demarcate sonographic breast lesions, especially for the cases with complicated echogenicity and shapes. It suggests that the ACCOMP-generated boundaries can potentially serve as the basis for further morphological or quantitative analysis.

Development of NTU standard Chinese Brain Template: morphologic and functional comparison with MNI template using magnetic resonance imaging.

BACKGROUND AND PURPOSE: The brain structure mismatch between western and eastern people may lead to an inappropriate interpretation of neurocognitive studies. To minimize this interracial misinterpretation, we developed the National Taiwan University Chinese Brain Template (NTU-CBT). METHODS: 102 (M/F = 55/47) healthy Chinese subjects were recruited and received 3T MR brain scans. The template development processes were based on the construction process of Montreal Neurological Institute (MNI) template. Further pilot functional magnetic resonance imaging (fMRI) studies with blocked design visual stimulation and foot tapping task were performed on 3 volunteers and applied to both MNI template and NTU-CBT for analyses. RESULTS: 7 subjects were excluded due to motion artifacts. The average brain size of 95 (M/F = 50/45) subjects was 16.0 cm in length, 13.9 cm in width and 11.3 cm in height, which was 88.9%, 97.9% and 84.3% of the size of MNI template, respectively. Maximum dimensional differences came from the height of superior brain and the length of posterior brain. The average activation voxel volume of the fMRI studies applying to NTU-CBT was 80.7% of that to MNI template in visual stimulation, and 72.8% in foot tapping task. Noticeable mismatches were noted between interpolating original data to NTU-CBT and MNI template. CONCLUSIONS: Morphologic differences between MNI template and NTU-CBT do lead to spatial mismatch in functional studies, especially at cortical regions of superior and posterior brain. With the development of NTU-CBT, we look forward to more accurate interpretation in neurocognitive studies for Chinese subjects.