Statistical Detection of Colors in Dermoscopic Images With a Texton-Based Estimation of Probabilities.

Color has great diagnostic significance in dermatoscopy. Several diagnosis methods are based on the colors detected within a lesion. Malignant lesions frequently show more than three colors, whereas in benign lesions, three or fewer colors are usually observed. Black, red, white, and blue-gray are found more frequently in melanomas than in benign nevi. In this paper, a method to automatically identify the colors of a lesion is presented. A color label identification problem is proposed and solved by maximizing the posterior probability of a pixel to belong to a label, given its color value and the neighborhood color values. The main contribution of this paper is the estimation of the different terms involved in the computation of this probability. Two evaluations are performed on a database of 200 dermoscopic images. The first one evaluates if all the colors detected in a lesion are indeed present in it. The second analyzes if each pixel within a lesion is assigned the correct color label. The results show that the proposed method performs correctly and outperforms other methods, with an average F-measure of 0.89, an accuracy of 0.90, and a Spearman correlation of 0.831.

Rules of tissue packing involving different cell types: human muscle organization.

Natural packed tissues are assembled as tessellations of polygonal cells. These include skeletal muscles and epithelial sheets. Skeletal muscles appear as a mosaic composed of two different types of cells: the "slow" and "fast" fibres. Their relative distribution is important for the muscle function but little is known about how the fibre arrangement is established and maintained. In this work we capture the organizational pattern in two different healthy muscles: biceps brachii and quadriceps. Here we show that the biceps brachii muscle presents a particular arrangement, based on the different sizes of slow and fast fibres. By contrast, in the quadriceps muscle an unbiased distribution exists. Our results indicate that the relative size of each cellular type imposes an intrinsic organization into natural tessellations. These findings establish a new framework for the analysis of any packed tissue where two or more cell types exist.

Machine Learning Methods for Binary and Multiclass Classification of Melanoma Thickness From Dermoscopic Images.

Thickness of the melanoma is the most important factor associated with survival in patients with melanoma. It is most commonly reported as a measurement of depth given in millimeters (mm) and computed by means of pathological examination after a biopsy of the suspected lesion. In order to avoid the use of an invasive method in the estimation of the thickness of melanoma before surgery, we propose a computational image analysis system from dermoscopic images. The proposed feature extraction is based on the clinical findings that correlate certain characteristics present in dermoscopic images and tumor depth. Two supervised classification schemes are proposed: a binary classification in which melanomas are classified into thin or thick, and a three-class scheme (thin, intermediate, and thick). The performance of several nominal classification methods, including a recent interpretable method combining logistic regression with artificial neural networks (Logistic regression using Initial variables and Product Units, LIPU), is compared. For the three-class problem, a set of ordinal classification methods (considering ordering relation between the three classes) is included. For the binary case, LIPU outperforms all the other methods with an accuracy of 77.6%, while, for the second scheme, although LIPU reports the highest overall accuracy, the ordinal classification methods achieve a better balance between the performances of all classes.

Model-based classification methods of global patterns in dermoscopic images.

In this paper different model-based methods of classification of global patterns in dermoscopic images are proposed. Global patterns identification is included in the pattern analysis framework, the melanoma diagnosis method most used among dermatologists. The modeling is performed in two senses: first a dermoscopic image is modeled by a finite symmetric conditional Markov model applied to L∗a∗b∗ color space and the estimated parameters of this model are treated as features. In turn, the distribution of these features are supposed that follow different models along a lesion: a Gaussian model, a Gaussian mixture model, and a bag-of-features histogram model. For each case, the classification is carried out by an image retrieval approach with different distance metrics. The main objective is to classify a whole pigmented lesion into three possible patterns: globular, homogeneous, and reticular. An extensive evaluation of the performance of each method has been carried out on an image database extracted from a public Atlas of Dermoscopy. The best classification success rate is achieved by the Gaussian mixture model-based method with a 78.44% success rate in average. In a further evaluation the multicomponent pattern is analyzed obtaining a 72.91% success rate.

Topological progression in proliferating epithelia is driven by a unique variation in polygon distribution.

Morphogenesis is consequence of lots of small coordinated variations that occur during development. In proliferating stages, tissue growth is coupled to changes in shape and organization. A number of studies have analyzed the topological properties of proliferating epithelia using the Drosophila wing disc as a model. These works are based in the existence of a fixed distribution of these epithelial cells according to their number of sides. Cell division, cell rearrangements or a combination of both mechanisms have been proposed to be responsible for this polygonal assembling. Here, we have used different system biology methods to compare images from two close proliferative stages that present high morphological similarity. This approach enables us to search for traces of epithelial organization. First, we show that geometrical and network characteristics of individual cells are mainly dependent on their number of sides. Second, we find a significant divergence between the distribution of polygons in epithelia from mid-third instar larva versus early prepupa. We show that this alteration propagates into changes in epithelial organization. Remarkably, only the variation in polygon distribution driven by morphogenesis leads to progression in epithelial organization. In addition, we identify the relevant features that characterize these rearrangements. Our results reveal signs of epithelial homogenization during the growing phase, before the planar cell polarity pathway leads to the hexagonal packing of the epithelium during pupal stages.

Linearized multidimensional earth-mover's-distance gradient flows.

This paper presents the first framework capable of performing active contour segmentation using Earth Mover's Distance (EMD) to measure dissimilarity between multidimensional feature distributions. EMD is the best known and understood cross-bin histogram distance measure, and as such it allows for meaningful comparisons between distributions, unlike bin-to-bin measures that only account for discrepancies on a bin-to-bin basis. Because EMD is obtained with linear programming techniques, its differential structure with respect to variations in bin weights as the active contour evolves is expressed through sensitivity analysis. Euler-Lagrange equations are then derived from the computed sensitivity at every iteration to produce gradient descent flows. We validate our approach with color image segmentation, in comparison with state-of-the-art Bhattacharyya (bin-to-bin) and 1D EMD (cross-bin) active contours. Some unique advantages of cross-bin comparison are highlighted in our segmentation results: better perceptual value and increased robustness with respect to the initialization.

Neuromuscular disease classification system.

Diagnosis of neuromuscular diseases is based on subjective visual assessment of biopsies from patients by the pathologist specialist. A system for objective analysis and classification of muscular dystrophies and neurogenic atrophies through muscle biopsy images of fluorescence microscopy is presented. The procedure starts with an accurate segmentation of the muscle fibers using mathematical morphology and a watershed transform. A feature extraction step is carried out in two parts: 24 features that pathologists take into account to diagnose the diseases and 58 structural features that the human eye cannot see, based on the assumption that the biopsy is considered as a graph, where the nodes are represented by each fiber, and two nodes are connected if two fibers are adjacent. A feature selection using sequential forward selection and sequential backward selection methods, a classification using a Fuzzy ARTMAP neural network, and a study of grading the severity are performed on these two sets of features. A database consisting of 91 images was used: 71 images for the training step and 20 as the test. A classification error of 0% was obtained. It is concluded that the addition of features undetectable by the human visual inspection improves the categorization of atrophic patterns.

Quantifiable diagnosis of muscular dystrophies and neurogenic atrophies through network analysis.

BACKGROUND: The diagnosis of neuromuscular diseases is strongly based on the histological characterization of muscle biopsies. However, this morphological analysis is mostly a subjective process and difficult to quantify. We have tested if network science can provide a novel framework to extract useful information from muscle biopsies, developing a novel method that analyzes muscle samples in an objective, automated, fast and precise manner. METHODS: Our database consisted of 102 muscle biopsy images from 70 individuals (including controls, patients with neurogenic atrophies and patients with muscular dystrophies). We used this to develop a new method, Neuromuscular DIseases Computerized Image Analysis (NDICIA), that uses network science analysis to capture the defining signature of muscle biopsy images. NDICIA characterizes muscle tissues by representing each image as a network, with fibers serving as nodes and fiber contacts as links. RESULTS: After a 'training' phase with control and pathological biopsies, NDICIA was able to quantify the degree of pathology of each sample. We validated our method by comparing NDICIA quantification of the severity of muscular dystrophies with a pathologist's evaluation of the degree of pathology, resulting in a strong correlation (R = 0.900, P <0.00001). Importantly, our approach can be used to quantify new images without the need for prior 'training'. Therefore, we show that network science analysis captures the useful information contained in muscle biopsies, helping the diagnosis of muscular dystrophies and neurogenic atrophies. CONCLUSIONS: Our novel network analysis approach will serve as a valuable tool for assessing the etiology of muscular dystrophies or neurogenic atrophies, and has the potential to quantify treatment outcomes in preclinical and clinical trials.