Mark G. Eramian

Local S Spectrum Analysis of 1-D and 2-D Data

Abstract The local changes of the spectrum with time are often more interesting than the spectrum of the whole time series. For example, there is an apparent drift in the nominal 28 day fluctuations of sunspot numbers over the period of the sunspot cycle, averaging ∼ 11.1 years. This time-local change in spectrum is due to a combination of Sporers Law and the differential rotation of the sun. Similarly, the space-local variations in the 2-D spectrum on an image conveys visual information on textures, boundaries and shapes. In this paper we use the recently developed S -transform to analyse two segments of the Wolf Sunspot series, a seismogram, and a synthetic 2-D image as examples of applications of the S -transform for time-local and space-local spectral analysis.

Generation of aquifer heterogeneity maps using two-dimensional spectral texture segmentation techniques

Numerical models that solve the governing equations for subsurface fluid flow and transport require detailed quantitative maps of spatially variable hydraulic properties. Recently, there has been great interest in methods that can map the spatial variability of hydraulic properties such as porosity and hydraulic conductivity (permeability). Presently, only limited data on natural permeability spatial structure are available. These data are often based on extensive discrete sampling in outcrops or boreholes. Then methods are used to interpolate between data values to map aquifer heterogeneity. Interpolation methods often mask critical local or intermediate scale heterogeneities. As sediment texture is directly correlated with many hydraulic properties we developed two new texture segmentation algorithms based on a space-local two-dimensional wavenumber spectral method known as the S-Transform. Existing texture segmentation algorithms could not delineate the subtle and continuous texture variations that exist in natural sediments. The S-Transform algorithms successfully delineated geologic structures and grain size patterns in photographs of outcrops in a glacial fluvial deposit; thus, no interpolation methods were required to produce continuous two-dimensional maps of texture facies. The S-Transform method is robust and is insensitive to changes in light intensity, and moisture variations. This makes the algorithm particularly applicable to natural sedimentary outcrops. The effectiveness of our methods are tested by correlating measured relative grain sizes in the images with actual grain size measurements taken from the sedimentary outcrops.

Feature-rich distance-based terrain synthesis

This paper describes a novel terrain synthesis method based on distances in a weighted graph. A height field is determined by least-cost paths in a weighted graph from a set of generator nodes. The shapes of individual terrain features, such as mountains, hills, and craters, are specified by a monotonically decreasing profile describing the cross-sectional shape of a feature. The locations of features in the terrain are specified by placing the generators; secondary ridges are placed by pathing. We show the method to be robust and easy to control, even making it possible to embed images in terrain shadows. The method can produce a wide range of realistic synthetic terrains such as mountain ranges, craters, cinder cones, and hills. The ability to manually place terrain features that incorporate multiple profiles produces heterogeneous terrains that compare favorably to existing methods.

Segmentation of epithelium in H&E stained odontogenic cysts

An algorithm for the automated segmentation of epithelial tissue in digital images of histologic tissue sections of odontogenic cysts (cysts originating from residual odontogenic epithelium) is presented. The algorithm features an image standardization process that greatly reduces variation in luminance and chrominance between images due to variations in sample preparation. Segmentation of the epithelial regions of images uses an algorithm based on binary graph cuts where graph weights depend on probabilities obtained from colour histogram models of epithelium and stroma image regions. Algorithm training used a data set of 38 images of four types of odontogenic cyst and was tested using a separate data set of 35 images of the same four cyst types. The best parameters for the segmentation algorithm were determined using a response‐surface optimizer. The best parameter set resulted in an overall mean (± std. dev.) sensitivity of 91.5 ± 17% and overall mean specificity of 85.1 ± 18.6% on the training set. Particularly good results were obtained for dentigerous and odontogenic keratocysts for which the mean sensitivities/specificities were 91.9 ± 6.15%/97.4 ± 2.15% and 96.1 ± 1.98%/98.7 ± 3.16%, respectively. Our method is potentially applicable to many pathological conditions in similar tissues, such as skin and mucous membranes where there is a clear microscopic distinction between epithelium and connective tissues.

Thyroid Nodule Classification in Ultrasound Images by Fine-Tuning Deep Convolutional Neural Network

With many thyroid nodules being incidentally detected, it is important to identify as many malignant nodules as possible while excluding those that are highly likely to be benign from fine needle aspiration (FNA) biopsies or surgeries. This paper presents a computer-aided diagnosis (CAD) system for classifying thyroid nodules in ultrasound images. We use deep learning approach to extract features from thyroid ultrasound images. Ultrasound images are pre-processed to calibrate their scale and remove the artifacts. A pre-trained GoogLeNet model is then fine-tuned using the pre-processed image samples which leads to superior feature extraction. The extracted features of the thyroid ultrasound images are sent to a Cost-sensitive Random Forest classifier to classify the images into “malignant” and “benign” cases. The experimental results show the proposed fine-tuned GoogLeNet model achieves excellent classification performance, attaining 98.29% classification accuracy, 99.10% sensitivity and 93.90% specificity for the images in an open access database (Pedraza et al. 16), while 96.34% classification accuracy, 86% sensitivity and 99% specificity for the images in our local health region database.

LBP-Based Segmentation of Defocus Blur

Defocus blur is extremely common in images captured using optical imaging systems. It may be undesirable, but may also be an intentional artistic effect, thus it can either enhance or inhibit our visual perception of the image scene. For tasks, such as image restoration and object recognition, one might want to segment a partially blurred image into blurred and non-blurred regions. In this paper, we propose a sharpness metric based on local binary patterns and a robust segmentation algorithm to separate in- and out-of-focus image regions. The proposed sharpness metric exploits the observation that most local image patches in blurry regions have significantly fewer of certain local binary patterns compared with those in sharp regions. Using this metric together with image matting and multi-scale inference, we obtained high-quality sharpness maps. Tests on hundreds of partially blurred images were used to evaluate our blur segmentation algorithm and six comparator methods. The results show that our algorithm achieves comparative segmentation results with the state of the art and have big speed advantage over the others.

Enhancement of Textural Differences Based on Morphological Component Analysis

This paper proposes a new texture enhancement method which uses an image decomposition that allows different visual characteristics of textures to be represented by separate components in contrast with previous methods which either enhance texture indirectly or represent all texture information using a single image component. Our method is intended to be used as a preprocessing step prior to the use of texture-based image segmentation algorithms. Our method uses a modification of morphological component analysis (MCA) which allows texture to be separated into multiple morphological components each representing a different visual characteristic of texture. We select four such texture characteristics and propose new dictionaries to extract these components using MCA. We then propose procedures for modifying each texture component and recombining them to produce a texture-enhanced image. We applied our method as a preprocessing step prior to a number of texture-based segmentation methods and compared the accuracy of the results, finding that our method produced results superior to comparator methods for all segmentation algorithms tested. We also demonstrate by example the main mechanism by which our method produces superior results, namely that it causes the clusters of local texture features of each distinct image texture to mutually diverge within the multidimensional feature space to a vastly superior degree versus the comparator enhancement methods.

Rapamycin reduces fibroblast proliferation without causing quiescence and induces STAT5A/B-mediated cytokine production.

Rapamycin is a well-known inhibitor of the Target of Rapamycin (TOR) signaling cascade; however, the impact of this drug on global genome function and organization in normal primary cells is poorly understood. To explore this impact, we treated primary human foreskin fibroblasts with rapamycin and observed a decrease in cell proliferation without causing cell death. Upon rapamycin treatment chromosomes 18 and 10 were repositioned to a location similar to that of fibroblasts induced into quiescence by serum reduction. Although similar changes in positioning occurred, comparative transcriptome analyses demonstrated significant divergence in gene expression patterns between rapamycin-treated and quiescence-induced fibroblasts. Rapamycin treatment induced the upregulation of cytokine genes, including those from the Interleukin (IL)-6 signaling network, such as IL-8 and the Leukemia Inhibitory Factor (LIF), while quiescent fibroblasts demonstrated up-regulation of genes involved in the complement and coagulation cascade. In addition, genes significantly up-regulated by rapamycin treatment demonstrated increased promoter occupancy of the transcription factor Signal Transducer and Activator of Transcription 5A/B (STAT5A/B). In summary, we demonstrated that the treatment of fibroblasts with rapamycin decreased proliferation, caused chromosome territory repositioning and induced STAT5A/B-mediated changes in gene expression enriched for cytokines.

Image-based search and retrieval for biface artefacts using features capturing archaeologically significant characteristics

Archaeologists are currently producing huge numbers of digitized photographs to record and preserve artefact finds. These images are used to identify and categorize artefacts and reason about connections between artefacts and perform outreach to the public. However, finding specific types of images within collections remains a major challenge. Often, the metadata associated with images is sparse or is inconsistent. This makes keyword-based exploratory search difficult, leaving researchers to rely on serendipity and slowing down the research process. We present an image-based retrieval system that addresses this problem for biface artefacts. In order to identify artefact characteristics that need to be captured by image features, we conducted a contextual inquiry study with experts in bifaces. We then devised several descriptors for matching images of bifaces with similar artefacts. We evaluated the performance of these descriptors using measures that specifically look at the differences between the sets of images returned by the search system using different descriptors. Through this nuanced approach, we have provided a comprehensive analysis of the strengths and weaknesses of the different descriptors and identified implications for design in the search systems for archaeology.