Quan Wen

Lesion Border Detection in Dermoscopy Images Using Ensembles of Thresholding Methods

Dermoscopy is one of the major imaging modalities used in the diagnosis of melanoma and other pigmented skin lesions. Due to the difficulty and subjectivity of human interpretation, automated analysis of dermoscopy images has become an important research area. Border detection is often the first step in this analysis. In many cases, the lesion can be roughly separated from the background skin using a thresholding method applied to the blue channel. However, no single thresholding method appears to be robust enough to successfully handle the wide variety of dermoscopy images encountered in clinical practice.

A Delaunay triangulation approach for segmenting clumps of nuclei

Cell-based fluorescence imaging assays have the potential to generate massive amount of data, which requires detailed quantitative analysis. Often, as a result of fixation, labeled nuclei overlap and create a clump of cells. However, it is important to quantify phenotypic read out on a cell-by-cell basis. In this paper, we propose a novel method for decomposing clumps of nuclei using high-level geometric constraints that are derived from low-level features of maximum curvature computed along the contour of each clump. Points of maximum curvature are used as vertices for Delaunay triangulation (DT), which provides a set of edge hypotheses for decomposing a clump of nuclei. Each hypothesis is subsequently tested against a constraint satisfaction network for a near optimum decomposition. The proposed method is compared with other traditional techniques such as the watershed method with/without markers. The experimental results show that our approach can overcome the deficiencies of the traditional methods and is very effective in separating severely touching nuclei.

Hard versus fuzzy c-means clustering for color quantization

Color quantization is an important operation with many applications in graphics and image processing. Most quantization methods are essentially based on data clustering algorithms. Recent studies have demonstrated the effectiveness of hard c-means (k-means) clustering algorithm in this domain. Other studies reported similar findings pertaining to the fuzzy c-means algorithm. Interestingly, none of these studies directly compared the two types of c-means algorithms. In this study, we implement fast and exact variants of the hard and fuzzy c-means algorithms with several initialization schemes and then compare the resulting quantizers on a diverse set of images. The results demonstrate that fuzzy c-means is significantly slower than hard c-means, and that with respect to output quality, the former algorithm is neither objectively nor subjectively superior to the latter.

A statistical approach for intensity loss compensation of confocal microscopy images

In this paper, a probabilistic technique for compensation of intensity loss in confocal microscopy images is presented. For single‐colour‐labelled specimen, confocal microscopy images are modelled as a mixture of two Gaussian probability distribution functions, one representing the background and another corresponding to the foreground. Images are segmented into foreground and background by applying Expectation Maximization algorithm to the mixture. Final intensity compensation is carried out by scaling and shifting the original intensities with the help of parameters estimated for the foreground. Since foreground is separated to calculate the compensation parameters, the method is effective even when image structure changes from frame to frame. As intensity decay function is not used, complexity associated with estimation of the intensity decay function parameters is eliminated. In addition, images can be compensated out of order, as only information from the reference image is required for the compensation of any image. These properties make our method an ideal tool for intensity compensation of confocal microscopy images that suffer intensity loss due to absorption/scattering of light as well as photobleaching and the image can change structure from optical/temporal section‐to‐section due to changes in the depth of specimen or due to a live specimen. The proposed method was tested with a number of confocal microscopy image stacks and results are presented to demonstrate the effectiveness of the method.

Multiple Interacting Subcellular Structure Tracking by Sequential Monte Carlo Method

With the wide application of green fluorescent protein (GFP) in the study of live cells, there is a surging need for the computer-aided analysis on the huge amount of im- age sequence data acquired by the advanced microscopy devices. One of such tasks is the motility analysis of the multiple subcellular structures. In this paper, an algorithm using sequential Monte Carlo (SMC) method for multiple interacting object tracking is proposed. First, marker resid- ual image is applied to detect individual subcellular struc- ture automatically, and to represent all the objects together using the joint state. Then the interaction between ob- jects in the 2D plane is modeled by augmenting an extra dimension and evaluating the overlapping relationship in the 3D space. Finally, the distribution of the dimension varying joint state is sampled efficiently by Reversible jump Markov chain Monte Carlo (RJMCMC) algorithm with a novel height swap move. The experimental results show that our method is promising.

Defining Vascular Supply and Territory of Thinned Perforator Flaps: Part Ii. Superior Gluteal Artery Perforator Flap

Background: Superior gluteal artery perforator flaps are surgical options in breast and pressure sore reconstructions. Based on the recipient site, primary thinning of these flaps may be necessary for final optimal contour. As the thinning of a superior gluteal artery perforator flap should be based on the knowledge of perforator vascular territories to prevent vascular compromise, the authors performed an anatomical study to determine the number, location, and diameter of the perforators present in the superior gluteal artery perforator flap. Accompanying veins and acceptable locations for surgical incisions were also determined. Methods: Fourteen superior gluteal artery perforator flaps were harvested from seven cadavers. Perforator flaps were thinned to 8 to 15 mm, except for a 2.5-cm radius around the dissected perforator. Vascular territory areas were quantified before and after thinning by photographic and radiographic methods, and respective vascular territory maps were constructed. Surgical incision “danger zones” of vertical and horizontal axes were determined at specific depths (relative to the skin surface) for each flap. Danger zone measurements were determined with an automatic three-dimensional vascular tree construction using computed tomographic images and several modeling algorithms. Results: Mean perforator artery diameter and number at the fascia level were 0.91 ± 0.07 mm and 2.86 ± 0.77 (mean ± SD), respectively. Perforator pedicles were located midway between the posterior superior iliac spine and the greater trochanter. After thinning, skin surface and whole flap vascular territories were reduced 80.9 percent (photographic) and 76.9 percent (radiographic), respectively, compared with unthinned vascular territory areas. From the skin at 4-, 6-, and 8-mm thicknesses, elliptical danger zones (two vertical segments and two horizontal segments) had overall vertical segment axis length ranges from the pedicles of 59 to 66 mm, 51 to 57 mm, and 49 to 51 mm, respectively. Horizontal axis segment length ranges were 61 to 76 mm, 61 to 66 mm, and 60 to 57 mm for 4-, 6-, and 8-mm skin thicknesses, respectively. Conclusions: The superior gluteal artery perforator flap provides an excellent blood supply to adipose tissue but may be compromised when aggressively thinned. Surgeons may design and harvest partially thinned superior gluteal artery perforator flaps based on the anatomical vascular territory maps provided by this study.

An effective real-time color quantization method based on divisive hierarchical clustering

Color quantization (CQ) is an important operation with many applications in graphics and image processing. Clustering algorithms have been extensively applied to this problem. In this paper, we propose a simple yet effective CQ method based on divisive hierarchical clustering. Our method utilizes the commonly used binary splitting strategy along with several carefully selected heuristics that ensure a good balance between effectiveness and efficiency. We also propose a slightly computationally expensive variant of this method that employs local optimization using the Lloyd–Max algorithm. Experiments on a diverse set of publicly available images demonstrate that the proposed method outperforms some of the most popular quantizers in the literature.

A particle filter framework using optimal importance function for protein molecules tracking

Tagging and tracking protein molecules are a key to a better understanding of proteomics in diverse aspects. In this paper, a common framework of particle filter using optimal importance function is proposed for confocal protein molecules tracking. To deal with the challenges stemming from small size, deformable shape, noisy environment, and multi-modality motion, a stochastic process based particle filter is used. Partial Gaussian state space (PGSS) model is developed as the importance function to incorporate the latest measurement in the state estimation. Experimental results have demonstrated the performance of the proposed algorithm for both Brownian and translational motion.

Colour quantisation using the adaptive distributing units algorithm

Abstract Colour quantisation (CQ) is an important operation with many applications in graphics and image processing. Most CQ methods are essentially based on data clustering algorithms one of which is the popular k-means algorithm. Unfortunately, like many batch clustering algorithms, k-means is highly sensitive to the selection of the initial cluster centres. In this paper, we adapt Uchiyama and Arbib’s competitive learning algorithm to the CQ problem. In contrast to the batch k-means algorithm, this online clustering algorithm does not require cluster centre initialisation. Experiments on a diverse set of publicly available images demonstrate that the presented method outperforms some of the most popular quantisers in the literature.

Color Quantization of Dermoscopy Images Using the K-Means Clustering Algorithm

Color quantization (CQ) is an important operation with various applications in medical image analysis. Most quantization methods are essentially based on data clustering algorithms. However, despite its popularity as a general purpose clustering algorithm, k-means has not received much respect in the CQ literature because of its high computational requirements and sensitivity to initialization. In this chapter, we investigate the performance of a recently proposed k-means based CQ method. Experiments on a diverse set of dermoscopy images of skin lesions demonstrate that an efficient implementation of k-means with an appropriate initialization strategy can in fact serve as a very effective color quantizer.