Boundary modelling and shape analysis methods for classification of mammographic masses

The problem of computer-aided classification of benign and malignant breast masses using shape features is addressed. The aim of the study is to look at the exceptions in shapes of masses such as circumscribed malignant tumours and spiculated benign masses which are difficult to classify correctly using common shape analysis methods. The proposed methods of shape analysis treat the object's boundary in terms of local details. The boundaries of masses analysed using the proposed methods were manually drawn on mammographic images by an expert radiologist (JELD). A boundary segmentation method is used to separate major portions of the boundary and to label them as concave or convex segments. To analyse the shape information localised in each segment, features are computed through an iterative procedure for polygonal modelling of the mass boundaries. Features are based on the concavity fraction of a mass boundary and the degree of narrowness of spicules as characterised by a spiculation index. Two features comprising spiculation index (SI) and fractional concavity (f_cc) developed in the present study when used in combination with the global shape feature of compactness resulted in a benign/malignant classification accuracy of 82%, with an area (A_z) of 0.79 under the receiver operating characteristics (ROC) curve with a database of the boundaries of 28 benign masses and 26 malignant tumours. SI alone resulted in a classification accuracy of 80% with A_z of 0.82. The combination of all the three features achieved 91% accuracy of circumscribed versus spiculated classification of masses based on shape.     

Usefulness of Texture Analysis for Computerized Classification of Breast Lesions on Mammograms

This work presents the usefulness of texture features in the classification of breast lesions in 5,518 images of regions of interest, which were obtained from the Digital Database for Screening Mammography that included microcalcifications, masses, and normal cases. Sixteen texture features were used, i.e., 13 were based on the spatial gray-level dependence matrix and 3 on the wavelet transform. The nonparametric K-NN classifier was used in the classification stage. The results obtained from receiver operating characteristic analysis indicated that the texture features can be used for separating normal regions and lesions with masses and microcalcifications, yielding the area under the curve (AUC) values of 0.957 and 0.859, respectively. However, the texture features were not very effective for distinguishing between malignant and benign lesions because the AUC was 0.617 for masses and 0.607 for microcalcifications. The study showed that the texture features can be used for the detection of suspicious regions in mammograms.     

Micro-macro hybrid soft-lithography master (MMHSM) fabrication for lab-on-a-chip applications

We present a novel micro-macro hybrid soft-lithography master (MMHSM) fabrication technique where microdevices having both microscale and macroscale features can be replicated with a single soft-lithography step. A poly(methyl methacrylate) (PMMA) master having macroscale structures was first created by a bench-top milling machine. An imprinting master mold having microscale structures was then imprinted on the PMMA surface through a hot-embossing process to obtain a PMMA master mold. A poly(dimethylsiloxane) (PDMS) master was then replicated from this PMMA master through a standard soft-lithography process. This process allowed both microscale (height: 3–20 μm, width: 20–500 μm) and macroscale (height: 3.5 mm, width: 1.2–7 mm) structures to co-exist on the PDMS master mold, from which final PDMS devices could be easily stamped out in large quantities. Microfluidic structures requiring macroscale dimensions in height, such as reservoirs or fluidic tubing interconnects, could be directly built into PDMS microfluidic devices without the typically used manual punching process. This significantly reduced alignment errors and time required for such manual fabrication steps. In this paper, we successfully demonstrated the utility of this novel hybrid fabrication method by fabricating a PDMS microfluidic device with 40 built-in fluidic interfaces and a PDMS multi-compartment neuron co-culture platform, where millimeter-scale compartments are connected via arrays of 20 μm wide and 200 μm long microfluidic channels. The resulting structures were characterized for the integrity of the transferred pattern sizes and the surface roughness using scanning electron microscopy and optical profilometry.     

Classification of breast masses in mammograms using genetic programming and feature selection

Mammography is a widely used screening tool and is the gold standard for the early detection of breast cancer. The classification of breast masses into the benign and malignant categories is an important problem in the area of computer-aided diagnosis of breast cancer. A small dataset of 57 breast mass images, each with 22 features computed, was used in this investigation; the same dataset has been previously used in other studies. The extracted features relate to edge-sharpness, shape, and texture. The novelty of this paper is the adaptation and application of the classification technique called genetic programming (GP), which possesses feature selection implicitly. To refine the pool of features available to the GP classifier, we used feature-selection methods, including the introduction of three statistical measures—Student’s t test, Kolmogorov–Smirnov test, and Kullback–Leibler divergence. Both the training and test accuracies obtained were high: above 99.5% for training and typically above 98% for test experiments. A leave-one-out experiment showed 97.3% success in the classification of benign masses and 95.0% success in the classification of malignant tumors. A shape feature known as fractional concavity was found to be the most important among those tested, since it was automatically selected by the GP classifier in almost every experiment.     

Classification of Breast Masses Using Selected Shape,Edge-sharpness,and Texture Features with Linear and Kernel-based Classifiers

Breast masses due to benign disease and malignant tumors related to breast cancer differ in terms of shape, edge-sharpness, and texture characteristics. In this study, we evaluate a set of 22 features including 5 shape factors, 3 edge-sharpness measures, and 14 texture features computed from 111 regions in mammograms, with 46 regions related to malignant tumors and 65 to benign masses. Feature selection is performed by a genetic algorithm based on several criteria, such as alignment of the kernel with the target function, class separability, and normalized distance. Fisher's linear discriminant analysis, the support vector machine (SVM), and our strict two-surface proximal (S2SP) classifier, as well as their corresponding kernel-based nonlinear versions, are used in the classification task with the selected features. The nonlinear classification performance of kernel Fisher's discriminant analysis, SVM, and S2SP, with the Gaussian kernel, reached 0.95 in terms of the area under the receiver operating characteristics curve. The results indicate that improvement in classification accuracy may be gained by using selected combinations of shape, edge-sharpness, and texture features.     

A 3D Radiomics-Based Artificial Neural Network Model for Benign Versus Malignant Vertebral Compression Fracture Classification in MRI

To train an artificial neural network model using 3D radiomic features to differentiate benign from malignant vertebral compression fractures (VCFs) on MRI. This retrospective study analyzed sagittal T1-weighted lumbar spine MRIs from 91 patients (average age of 64.24 ± 11.75 years) diagnosed with benign or malignant VCFs from 2010 to 2019, of them 47 (51.6%) had benign VCFs and 44 (48.4%) had malignant VCFs. The lumbar fractures were three-dimensionally segmented and had their radiomic features extracted and selected with the wrapper method. The training set consisted of 100 fractured vertebral bodies from 61 patients (average age of 63.2 ± 12.5 years), and the test set was comprised of 30 fractured vertebral bodies from 30 patients (average age of 66.4 ± 9.9 years). Classification was performed with the multilayer perceptron neural network with a back-propagation algorithm. To validate the model, the tenfold cross-validation technique and an independent test set (holdout) were used. The performance of the model was evaluated using the average with a 95% confidence interval for the ROC AUC, accuracy, sensitivity, and specificity (considering the threshold = 0.5). In the internal validation test, the best model reached a ROC AUC of 0.98, an accuracy of 95% (95/100), a sensitivity of 93.5% (43/46), and specificity of 96.3% (52/54). In the validation with independent test set, the model achieved a ROC AUC of 0.97, an accuracy of 93.3% (28/30), a sensitivity of 93.3% (14/15), and a specificity of 93.3% (14/15). The model proposed in this study using radiomic features could differentiate benign from malignant vertebral compression fractures with excellent performance and is promising as an aid to radiologists in the characterization of VCFs.

     

乳腺计算机辅助诊断中DCE-MRI图像特征的选择与分析

目的比较动态对比度增强磁共振成像（dynamic contrast—enhanced magnetic resonance imaging,DCE—MRI）图像的形态、纹理和时间强度曲线（time intensity curve,TIC）特征对乳腺病灶良恶性的诊断效果,讨论DCE—MRI图像特征的计算机辅助诊断价值。方法测量224个乳腺病灶样本（良性样本82个,恶性样本142个）的12个形态学特征、56个基于灰度共生矩阵（gray level co—occurrencematrix,GLCM）的纹理特征以及11个TIC特征,采用平均平方距离准则和SVM分类器评估这三类特征的良恶性分辨能力。结果反映病灶血流动力学特性的TIC特征的分类性能最优（SE0．9366,SP0．8293,AUC0．9495）;纹理特征次之（SE0．9225,SP0．7195,AUC0．8835）;形态学特征效果最差（SE0．8451,SP0．6951,AUC0．8079）。研究发现,在上述基础上融合三类特征可优化分类性能。最终结合平滑度、紧致度、熵等9个特征参数进行诊断,对乳腺病灶良恶性的分辨效果最好,AUC达0．9642。结论DCE—MRI的TIC特征对恶性乳腺病灶具有较高的灵敏度,可以提高乳腺计算机辅助诊断的恶性病灶检出率。综合分析形态、纹理和TIC特征可以提高病灶的诊断特异度,降低良性病灶的误诊率。     

Analysis of temporal changes of mammographic features: computer-aided classification of malignant and benign breast masses

A new classification scheme was developed to classify mammographic masses as malignant and benign by using interval change information. The masses on both the current and the prior mammograms were automatically segmented using an active contour method. From each mass, 20 run length statistics (RLS) texture features, 3 speculation features, and 12 morphological features were extracted. Additionally, 20 difference RLS features were obtained by subtracting the prior RLS features from the corresponding current RLS features. The feature space consisted of the current RLS features, the difference RLS features, the current and prior speculation features, and the current and prior mass sizes. Stepwise feature selection and linear discriminant analysis classification were used to select and merge the most useful features. A leave-one-case-out resampling scheme was used to train and test the classifier using 140 temporal image pairs (85 malignant, 55 benign) obtained from 57 biopsy-proven masses (33 malignant, 24 benign) in 56 patients. An average of 10 features were selected from the 56 training subsets: 4 difference RLS features, 4 RLS features, and 1 speculation feature from the current image, and 1 speculation feature from the prior, were most often chosen. The classifier achieved an average training Az of 0.92 and a test Az of 0.88. For comparison, a classifier was trained and tested using features extracted from the 120 current single images. This classifier achieved an average training Az of 0.90 and a test Az of 0.82. The information on the prior image significantly (p = 0.015) improved the accuracy for classification of the masses.     

Microsporidian parasites: a danger facing marine fishes of the Red Sea

Out of 600 marine fish from the Red Sea belonging to three different species that were collected and examined for microsporidian parasites, 87 (14.5%) fish were found to be infected. The infection was recorded as cysts or xenomas embedded in the gut epithelium and the peritoneal cavity of the three fish species. The highest percent of infection with microsporidian parasites was recorded in Saurida tumbil 19.5% (39/200) followed by Pagrus pagrus 15% (45/300) and the lowest percent of infection was recorded in Epinephelus chlorostigma 3% (three out of 100). After rupture of the cysts, the spores were released and examined by light microscopy. Each spore was elongated to ellipsoidal in shape and possessed a posterior vacuole which is characteristic to phylum Microspora. They measure 1.6 ± 0.5 μm (1.5–2.4 μm) × 1.3 ± 0.1 μm (1.3–2.0 μm) in Saurida tumbil and Pagrus pagrus, respectively. The spores of Pleistophora sp recorded from E. chlorostigma were ovoid to pyriform in shape and measure 1.9 ± 0.5 μm (1.8–2.7 μm) × 1.6 ± 0.4 μm (1.5–2.4 μm).     

Modelling the sampling volume for skin blood oxygenation measurements

The absolute quantified measurement of haemoglobin skin blood saturation from collected reflectance spectra of the skin is complicated by the fact that the blood content of tissues can vary both in the spatial distribution and in the amount. These measurements require an understanding of which vascular bed is primarily responsible for the detected signal. Knowing the spatial detector depth sensitivity makes it possible to find the best range of different probe geometries for the measurements of signal from the required zones and group of vessels inside the skin. To facilitate this, a Monte Carlo simulation has been developed to estimate the sampling volume offered by fibre-optic probes with a small source-detector spacing (in the current report 250 μm, 400 μm and 800 μm). The optical properties of the modelled medium are taken to be the optical properties of the Caucasian type of skin tissue in the visible range of the spectrum. It is shown that, for a small source-detector separation (800 μm and smaller), rough boundaries between layers of different refractive index can play a significant role in skin optics. Wavy layer interfaces produce a deeper and more homogeneous distribution of photons within the skin and tend to suppress the direct channelling of photons from source to detector. The model predicts that a probe spacing of 250 μm samples primarily epidermal layers and papillary dermis, whereas spacings of 400–800 μm sample upper blood net dermis and dermis.     

基于Local Jet变换空间纹理特征的肺结节分类方法

为了在纹理特征下改善肺结节良、恶性的模式识别,提出一种基于local jet变换空间的纹理特征提取方法。首先利用二维高斯函数的前三阶偏微分函数将结节原图像变换到local jet纹理图像空间,然后利用纹理描述子在该空间提取特征参数。以灰度值的前四阶矩和基于灰度共生矩阵的特征参数作为纹理描述子,分别提取结节原图像和变换后纹理图像的特征参数,以BP神经网络作为分类器,对同一纹理描述子下的2个不同图像空间的经核主成分分析优化后的特征参数集进行结节良、恶性分类。以157个肺结节(51个良性,106个恶性)作为实验数据进行对比实验,结果显示:两种纹理描述子基于local jet变换空间提取的特征参数分别获得82.69%和86.54%的分类正确率,较原图像空间提高6%～8%,同时AUC值提高约10%。实验结果表明,基于local jet变换空间提取的纹理特征可以有效地改善肺结节良、恶性的模式识别。     

Silicon-based microelectrodes for neurophysiology, micromachined from silicon-on-insulator wafers

A process is described for the fabrication of silicon-based microelectrodes for neurophysiology using bonded and etched-back silicon-on-insulator (BESOI) wafers. The probe shapes are defined without high levels of boron doping in the silicon; this is considered as a step towards producing probes with active electronics integrated directly beneath the electrodes. Gold electrodes, of 4μm by 4μm to 50μm by 50μm are fabricated on shanks (cantilever beams) 6μm thick and which taper to an area approximately 100μm wide and 200μm long, which are inserted into the tissue under investigation. The passive probes fabricated have been successfully employed to make acute recordings from locust peripheral nerve.     

Quantitative Ultrasound Analysis for Classification of BI-RADS Category 3 Breast Masses

The accuracy of an ultrasound (US) computer-aided diagnosis (CAD) system was evaluated for the classification of BI-RADS category 3, probably benign masses. The US database used in this study contained 69 breast masses (21 malignant and 48 benign masses) that at blinded retrospective interpretation were assigned to BI-RADS category 3 by at least one of five radiologists. For computer-aided analysis, multiple morphology (shape, orientation, margin, lesions boundary, and posterior acoustic features) and texture (echo patterns) features based on BI-RADS lexicon were implemented, and the binary logistic regression model was used for classification. The receiver operating characteristic curve analysis was used for statistical analysis. The area under the curve (Az) of morphology, texture, and combined features were 0.90, 0.75, and 0.95, respectively. The combined features achieved the best performance and were significantly better than using texture features only (0.95 vs. 0.75, p value?=?0.0163). The cut-off point at the sensitivity of 86 % (18/21), 95 % (20/21), and 100 % (21/21) achieved the specificity of 90 % (43/48), 73 % (35/48), and 33 % (16/48), respectively. In conclusion, the proposed CAD system has the potential to be used in upgrading malignant masses misclassified as BI-RADS category 3 on US by the radiologists.     

Micro-Flow Visualization of Red Blood Cell-Enhanced Platelet Concentration at Sudden Expansion

Microscopic steps and crevices are inevitable features within prosthetic blood-contacting devices. This study aimed to elucidate the thrombogenicity of the associated microscopic flow features by studying the transport of fluorescent platelet-sized particles in a suspension of red blood cells (RBCs) flowing through a 100 μm:200 μm sudden expansion. Micro-flow visualization revealed a strong influence of hematocrit upon the path of RBCs and spatial concentration of particles. At all flow rates studied (Re = 8.3–41.7) and hematocrit 20% and lower, RBC streamlines were found to detach from the microchannel wall creating an RBC-depleted zone inside the step that was much larger than the cells themselves. However, the observed distribution of particles was relatively homogeneous. By contrast, the RBC streamlines of samples with hematocrit equal to or greater than 30% more closely followed the contour of the microchannel, yet exhibited enhanced concentration of particles within the corner. The corresponding size of the cell depletion layer was comparable with the size of the cells. This study implies that local platelet concentration in blood within the physiological range of hematocrit can be elevated within the flow separation region of a sudden expansion and implicates the role of RBCs in causing this effect.     

Combination of Radiological and Gray Level Co-occurrence Matrix Textural Features Used to Distinguish Solitary Pulmonary Nodules by Computed Tomography

The objective of this study was to investigate the method of the combination of radiological and textural features for the differentiation of malignant from benign solitary pulmonary nodules by computed tomography. Features including 13 gray level co-occurrence matrix textural features and 12 radiological features were extracted from 2,117 CT slices, which came from 202 (116 malignant and 86 benign) patients. Lasso-type regularization to a nonlinear regression model was applied to select predictive features and a BP artificial neural network was used to build the diagnostic model. Eight radiological and two textural features were obtained after the Lasso-type regularization procedure. Twelve radiological features alone could reach an area under the ROC curve (AUC) of 0.84 in differentiating between malignant and benign lesions. The 10 selected characters improved the AUC to 0.91. The evaluation results showed that the method of selecting radiological and textural features appears to yield more effective in the distinction of malignant from benign solitary pulmonary nodules by computed tomography.     

Evaluation of an algorithm for arterial lumen diameter measurements by means of ultrasound

We have developed an algorithm for arterial luminal diameter measurement by means of ultrasound and evaluated the algorithm on agar vessel phantoms and in vivo. The algorithm utilises relative threshold detection on the inner slopes of the arterial walls before the resolution is improved by solving the equation of a straight line between the samples around the threshold value. Further, correction distances added to compensate for the underestimation when using the inner slopes were found to be 304 μm for the near wall and 415 μm for the far wall. The measured mean diameters of ten consecutive images of 3-, 6- and 9-mm phantoms were 3,006 μm (SD 4), 5,918 μm (SD 1) and 8,936 μm (SD 2), respectively. The mean differences between the images were 0.19, 0.04 and 0.37 μm, respectively. In vivo, the intra- and inter-observer variabilities were −64 μm (2SD 358) and −57 μm (2SD 366), respectively.     

An Efficient Fractal Method for Detection and Diagnosis of Breast Masses in Mammograms

In this paper, we present an efficient fractal method for detection and diagnosis of mass lesion in mammogram which is one of the abnormalities in mammographic images. We used 110 images that were carefully selected by a radiologist, and their abnormalities were also confirmed by biopsy. These images included circumscribed benign, ill-defined, and spiculated malignant masses. Firstly, we discriminated lesions automatically using new fractal dimensions. The results which were examined by different types of breast density showed that the proposed method was able to yield quite satisfactory detection results. Secondly, noting that contours of masses playing the most important role in diagnosis of different mass types, we defined new fractal features based on information extraction from the contours. This information is able to identify the roughness in mass contours and determines the extent of spiculation or smoothness of the masses. In this manner, in classification of the spiculated malignant masses from the circumscribed benign tumors, we achieved highly satisfactory results, i.e., 0.98 measured in terms of area under ROC curve (AUC). In this paper, it is also shown that the roughness in contours is a suitable characteristic feature for diagnosis of ill-defined malignant tumors with AUC equal to 0.94 in their classification. The extracted information was also found to be useful in the classification of early malignancies whereas in the classification of spiculated and ill-defined malignant masses in their early stage from those of benign tumors, we achieved high accuracy of 0.99 and 0.90 for AUC, respectively.     

Improvement of mammographic mass characterization using spiculation meausures and morphological features

We are developing new computer vision techniques for characterization of breast masses on mammograms. We had previously developed a characterization method based on texture features. The goal of the present work was to improve our characterization method by making use of morphological features. Toward this goal, we have developed a fully automated, three-stage segmentation method that includes clustering, active contour, and spiculation detection stages. After segmentation, morphological features describing the shape of the mass were extracted. Texture features were also extracted from a band of pixels surrounding the mass. Stepwise feature selection and linear discriminant analysis were employed in the morphological, texture, and combined feature spaces for classifier design. The classification accuracy was evaluated using the area Az under the receiver operating characteristic curve. A data set containing 249 films from 102 patients was used. When the leave-one-case-out method was applied to partition the data set into trainers and testers, the average test Az for the task of classifying the mass on a single mammographic view was 0.83 +/- 0.02, 0.84 +/- 0.02, and 0.87 +/- 0.02 in the morphological, texture, and combined feature spaces, respectively. The improvement obtained by supplementing texture features with morphological features in classification was statistically significant (p = 0.04). For classifying a mass as malignant or benign, we combined the leave-one-case-out discriminant scores from different views of a mass to obtain a summary score. In this task, the test Az value using the combined feature space was 0.91 +/- 0.02. Our results indicate that combining texture features with morphological features extracted from automatically segmented mass boundaries will be an effective approach for computer-aided characterization of mammographic masses.     

Design considerations in the development and application of microdisc electrode arrays (MDEAs) for implantable biosensors

The use of microlithographically fabricated Microdisc Electrode Arrays (MDEAs) in the development of implantable voltammetric biosensors necessitates design criteria that balances the overall footprint of the device with the advantages to be derived from large separation distances between non-interacting microdisc elements. Using the dynamic electroanalytical techniques of Multiple Scan Rate Cyclic Voltammetry (MSRCV) experiments with finite element simulations and Electrochemical Impedance Spectroscopy with equivalent circuit modeling, three unique MDEA designs; MDEA 050 (r = 25 μm, 5,184 discs), MDEA 100 (r = 50 μm, 1,296 discs) and MDEA 250 (r = 125 μm, 207 discs) of constant critical dimensions (center-to-center d/r = 4) and area (A = 0.1 cm²) were studied in 1.0 mM ferrocene monocarboxylic acid (FcCO₂H) solution (in 0.1 M Tris/0.1 M KCl buffer, pH = 7.2). The critical disc-to-disc spacing (d/r) required to archive 67% of maximal current response was defined as optimal. Based on the predictive model, new MDEA designs; MDEA 001 (r = 0.5 μm, 127,324 discs), MDEA 002.5 (r = 1.25 μm, 20,372 discs), MDEA 005 (r = 2.5 μm, 5,093 discs), MDEA 010 (r = 5 μm, 1,273 discs), MDEA 015 (r = 7.5 μm, 566 discs), MDEA 020 (r = 10 μm, 318 discs) were simulated at 10 and 100 mV/s. The final disc count of each MDEA was dictated by the need to maintain a comparable electroactive area between the MDEAs, which was chosen to be 0.001 cm², which in turn was dictated by the need to generate sufficient electrochemical current to be comfortably measured by common electrochemical detectors.     

The Impact of Voxel Size-Based Inaccuracies on the Mechanical Behavior of Thin Bone Structures

Computed tomography (CT)-based measures of skeletal geometry and material properties have been widely used to develop finite element (FE) models of bony structures. However, in the case of thin bone structures, the ability to develop FE models with accurate geometry derived from clinical CT data presents a challenge due to the thinness of the bone and the limited resolution of the imaging devices. The purpose of this study was to quantify the impact of voxel size on the thickness and intensity values of thin bone structure measurements and to assess the effect of voxel size on strains through FE modeling. Cortical bone thickness and material properties in five thin bone specimens were quantified at voxel sizes ranging from 16.4 to 488 μm. The measurements derived from large voxel size scans showed large increases in cortical thickness (61.9–252.2%) and large decreases in scan intensity (12.9–49.5%). Maximum principal strains from FE models generated using scans at 488 μm were decreased as compared to strains generated at 16.4 μm voxel size (8.6–64.2%). A higher level of significance was found in comparing intensity (p = 0.0001) vs. thickness (p = 0.005) to strain measurements. These findings have implications in developing methods to generate accurate FE models to predict the biomechanical behavior of thin bone structures.