Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

We demonstrate methods for the detection of architectural distortion in prior mammograms of interval-cancer cases based on analysis of the orientation of breast tissue patterns in mammograms. We hypothesize that architectural distortion modifies the normal orientation of breast tissue patterns in mammographic images before the formation of masses or tumors. In the initial steps of our methods, the oriented structures in a given mammogram are analyzed using Gabor filters and phase portraits to detect node-like sites of radiating or intersecting tissue patterns. Each detected site is then characterized using the node value, fractal dimension, and a measure of angular dispersion specifically designed to represent spiculating patterns associated with architectural distortion.Our methods were tested with a database of 106 prior mammograms of 56 interval-cancer cases and 52 mammograms of 13 normal cases using the features developed for the characterization of architectural distortion, pattern classification via quadratic discriminant analysis, and validation with the leave-one-patient out procedure. According to the results of free-response receiver operating characteristic analysis, our methods have demonstrated the capability to detect architectural distortion in prior mammograms, taken 15 months (on the average) before clinical diagnosis of breast cancer, with a sensitivity of 80% at about five false positives per patient.     

Fully automated measurements by light microscopy of tissue sections using a cellular array computer

Software was developed for the acquisition, segmentation and analysis of microscopic OD-images on a VICOM digital image processor, extended with a VISIOMORPH morphoprocessor board. The delineation algorithms for peroxisomes, lysosomes, and nuclei in liver, kidney, and adrenal gland sections start by thresholding the difference between the original image and a low pass filtered version. The resulting binary mask is then processed by morphological operations in order to produce an object overlay. The efficiency of the programs is evaluated by comparing delineated objects at different OD-levels, created by varying the stain or by multiplying the original pixel values with constant factors. Manual delineation on some images is also used as a reference. More complex algorithms are used for the delineation of muscle fibres in ATP-ase-stained sections and immunocytochemically labelled cells in monolayer preparations. Muscle images from parallel sections with different stainings are matched with a coordinate transform, enabling the transfer of the object mask from a single delineated image to the unprocessed images and thus obtain all necessary information for fibre classification. After segmentation, the OD-images and their object overlays are fed into a data extraction program, measuring for each delineated object user-selected features. Data are sent to a VAX for statistical interpretation.     

Direct identification of pure Penicillium species using image analysis

This paper presents a method for direct identification of fungal species solely by means of digital image analysis of colonies as seen after growth on a standard medium. The method described is completely automated and hence objective once digital images of the reference fungi have been established. Using a digital image it is possible to extract precise information from the surface of the fungal colony. This includes color distribution, colony dimensions and texture measurements. For fungal identification, this is normally done by visual observation that often results in a very subjective data recording. Isolates of nine different species of the genus Penicillium have been selected for the purpose. After incubation for 7 days, the fungal colonies are digitized using a very accurate digital camera. Prior to the image analysis each image is corrected for self-illumination, thereby gaining a set of directly corresponding images with respect to illumination. A Windows application has been developed to locate the position and size of up to three colonies in the digitized image. Using the estimated positions and sizes of the colonies, a number of relevant features can be extracted for further analysis. The method used to determine the position of the colonies will be covered as well as the feature selection. The texture measurements of colonies of the nine species were analyzed and a clustering of the data into the correct species was confirmed. This indicates that it is indeed possible to identify a given colony merely by macromorphological features. A classifier (in the normal distribution) based on measurements of 151 colonies incubated on yeast extract sucrose agar (YES) was used to discriminate between the species. This resulted in a correct classification rate of 100% when used on the training set and 96% using cross-validation. The same methods applied to 194 colonies incubated on Czapek yeast extract agar (CYA) resulted in a correct classification rate of 98% on the training set and 71% using cross-validation.     

Improvement of image classification using wavelet coefficients with structured-based neural network

Image classification is a challenging problem in organizing a large image database. However, an effective method for such an objective is still under investigation. A method based on wavelet analysis to extract features for image classification is presented in this paper. After an image is decomposed by wavelet, the statistics of its features can be obtained by the distribution of histograms of wavelet coefficients, which are respectively projected onto two orthogonal axes, i.e., x and y directions. Therefore, the nodes of tree representation of images can be represented by the distribution. The high level features are described in low dimensional space including 16 attributes so that the computational complexity is significantly decreased. 2,800 images derived from seven categories are used in experiments. Half of the images were used for training neural network and the other images used for testing. The features extracted by wavelet analysis and the conventional features are used in the experiments to prove the efficacy of the proposed method. The classification rate on the training data set with wavelet analysis is up to 91%, and the classification rate on the testing data set reaches 89%. Experimental results show that our proposed approach for image classification is more effective.     

Performance evaluation of a machine vision system for insect pests identification of field crops using artificial neural networks

Feature extraction is a crucial part of advanced image recognition systems. In this research, an autonomous detection device was designed and developed for insect pest detection to improve the ability of intelligent systems in order to annihilate harmful insect pests in agricultural crop fields. Device included a dark chamber, a CCD digital camera, a LDR lightening module and a personal computer. The proposed programme for precise insect pest detection was based on an image processing algorithm and artificial neural networks (ANNs). After image acquisition, the insect pests’ images were extracted from original images with Canny filtration. Afterwards, four morphological and three textural features from the obtained images were measured and normalised. Performance of ANN model was tested successfully for Beet armyworm (Spodoptera exigua) recognition in images using back-propagation supervised learning method and inspection data. Results showed that proposed system was able to identify S. exigua in the images from other species. Such this machine vision system can be used in autonomous field robots to achieve a modern farmer’s assistant.     

A thresholding method for automatic cell image segmentation.

An algorithm for automatic segmentation of PAP-stained cell images and its digital implementation is described. First, the image is filtered in order to eliminate the granularily and small objects in the image which may upset the segmentation procedure. In a second step, information on gradient and compactness is extracted from the filtered image and stored in three histograms as functions of the extinction. From these histograms, two extinction thresholds are computed. These thresholds are suitable to separate the nucleus from the cytoplasm, and the cytoplasm from the background in the filtered image. Masks are determined in this way, and finally used to analyse the nucleus and the cytoplasm in the original image.     

Automatic Detection and Classification of Mammograms Using Improved Extreme Learning Machine with Deep Learning

Background and objectiveBreast cancer, the most intrusive form of cancer affecting women globally. Next to lung cancer, breast cancer is the one that provides a greater number of cancer deaths among women. In recent times, several intelligent methodologies were come into existence for building an effective detection and classification of such noxious type of cancer. For further improving the rate of early diagnosis and for increasing the life span of victims, optimistic light of research is essential in breast cancer classification. Accordingly, a new customized method of integrating the concept of deep learning with the extreme learning machine (ELM), which is optimized using a simple crow-search algorithm (ICS-ELM). Thus, to enhance the state-of-the-art workings, an improved deep feature-based crow-search optimized extreme learning machine is proposed for addressing the health-care problem. The paper pours a light-of-research on detecting the input mammograms as either normal or abnormal. Subsequently, it focuses on further classifying the type of abnormal severities i.e., benign type or malignant.Materials and methodsThe digital mammograms for this work are taken from the Curated Breast Imaging Subset of DDSM (CBIS-DDSM), Mammographic Image Analysis Society (MIAS), and INbreast datasets. Herein, the work employs 570 digital mammograms (250 normal, 200 benign and 120 malignant cases) from CBIS-DDSM dataset, 322 digital mammograms (207 normal, 64 benign and 51 malignant cases) from MIAS database and 179 full-field digital mammograms (66 normal, 56 benign and 57 malignant cases) from INbreast dataset for its evaluation. The work utilizes ResNet-18 based deep extracted features with proposed Improved Crow-Search Optimized Extreme Learning Machine (ICS-ELM) algorithm.ResultsThe proposed work is finally compared with the existing Support Vector Machines (RBF kernel), ELM, particle swarm optimization (PSO) optimized ELM, and crow-search optimized ELM, where the maximum overall classification accuracy is obtained for the proposed method with 97.193% for DDSM, 98.137% for MIAS and 98.266% for INbreast datasets, respectively.ConclusionThe obtained results reveal that the proposed Computer-Aided-Diagnosis (CAD) tool is robust for the automatic detection and classification of breast cancer.     

基于地形限制特征的泾河流域遥感地表覆被分类

由于在分类方法和空间分辨率等方面存在局限性,基于粗分辨率遥感数据的传统非监督分类结果在不同地物过渡带内往往误差较大。该文提出了基于地形限制特征的分类方法,在非监督分类的基础上,将非监督分类结果按照像元进行细分,并运用地形限制条件对细分后的像元进行二次判别分类。结果表明,分类精度明显提高,其中,农田和居民点分类精度的提高最为明显。这一方法使得完全同质的单元可以进行属性的变更,改善了像元空间分辨率差造成的误差;而地形限制特征的引入减少了传统非监督分类的不确定性,使模糊区域的分类有了较为明确的区分特征,提高了分类的精度。     

Deep Learning-Based Multi-Label Tissue Segmentation and Density Assessment from Mammograms

Objectives: Breast cancer is the most commonly diagnosed type of cancer among women and a common cause of cancer-related deaths. Early diagnosis and treatment of breast cancer is critical in disease prognosis. Breast density is known to have a correlation with breast cancer. In recent years, there has been an increasing interest in the investigation of computer-aided methods for early diagnosis of breast cancer. In this study, a new fully-automated deep learning-based cascaded model was proposed for breast density assessment. In the first stage, the segmentation of adipose, fibroglandular, and pectoral muscle tissues from the digitized film mammograms of the Digital Database for Screening Mammography (DDSM) was investigated using various types of U-nets. Features extracted from the breast tissue segmentation predictions were then used to assess breast density in the second stage. Material and methods: 66 and 296 mediolateral oblique mammograms were selected from DDSM dataset for segmentation and breast density assessment systems, respectively. Different U-nets with varying number of layers and filters were implemented and the model having the highest performance was determined. U-net performance was investigated using categorical cross-entropy, Dice, Tversky, Focal Tversky, and logarithmic cosine-hyperbolic Dice loss functions. The performances of U-nets having different types of connections were investigated. The performances of U-nets having pre-trained weights from VGG16, VGG19, and ResNet50 networks in the encoding path were also investigated. Segmentation results were improved by using an image processing pipeline based on morphological operators. Segmentation performance was presented in terms of accuracy, balanced accuracy, intersection over union, and Dice's similarity coefficient (DSC) metrics. The segmentation system predictions were then used to estimate mammographic density using a machine learning pipeline by extracting features related to the fibroglandular tissue percentage. Results: Using ResNet50-U-net on the test data, average DSC scores of 82.71%, 73.39%, and 95.30% were obtained for adipose, fibroglandular, and pectoral muscle tissue segmentation, respectively. The mammogram segmentation results are 3%-12% better than the current state-of-the-art DSC in the literature when considering all of the foreground tissues concurrently. A breast density classification accuracy of 76.01% was achieved on a separate mammogram dataset, which is comparable to the recent studies in the literature. Conclusion: The proposed system can be used for automatic segmentation of mammogram into adipose, fibroglandular, and pectoral muscle tissues. The segmentation model enables the estimation of the fibroglandular-adipose tissue interface, which is recently found to be an important region for breast cancer investigations. The proposed fully-automatic breast density assessment system has a comparable performance to the ones in the literature.     

胰腺内镜超声图像纹理特征提取与分类研究

提出了胰腺内镜超声图像的纹理特征提取与分类方法,可应用于胰腺癌内镜超声图像的计算机辅助诊断。对胰腺内镜超声图像采用数字图像处理算法提取9大类共69个纹理特征。使用类间距作为可分性判据,实现特征的初步筛选,之后使用顺序前进搜索算法进一步筛选特征,并由支撑向量机实现分类。对216例病例随机选取训练集和测试集,通过多次随机实验表明。本文提出的算法实现了较高的分类准确率,为胰腺癌的临床诊断提供有价值的参考意见。     

Feature extraction from three-dimensional images in quantitative microscopy

Different methods are investigated in selecting and generating the appropriate microscope images for analysis of three-dimensional objects in quantitative microscopy. Traditionally, the ‘best’ focused image from a set is used for quantitative analysis. Such an objectively determined image is optimal for the extraction of some features, but may not be the best image for the extraction of all features. Various methods using multiple images are here developed to obtain a tighter distribution for all features.Three different approaches for analysis of images of stained cervical cells were analyzed. In the first approach, features are extracted from each image in the set. The feature values are then averaged to give the final result. In the second approach, a set of varying focused images are reconstructed to obtain a set of in-focus images. Features are then extracted from this set and averaged. In the third approach, a set of images in the three-dimensional scene is compressed into a single two-dimensional image. Four different compression methods are used. Features are then extracted from the resulting two-dimensional image. The third approach is employed on both the raw and transformed images.Each approach has its advantages and disadvantages. The first approach is fast and produces reasonable results. The second approach is more computationally expensive but produces the best results. The last approach overcomes the memory storage problem of the first two approaches since the set of images is compressed into one. The method of compression using the highest gradient pixel produces better results overall than other data reduction techniques and produces results comparable to the first approach.     

Interval-coded texture features for artifact rejection in automated cervical cytology

In order to improve the separation between abnormal cells and noncellular artifacts in the CERVIFIP automated cervical cytology prescreening system, 22 different object texture features were investigated. The features were all statistical parameters of the pixel density histograms or one-dimensional filtered values of central and border regions of the object images. The features were calculated for 231 images (100 cells and 131 artifacts) detected as Suspect Cells by the current CERVIFIP and were then tested in hierarchical and linear discriminant classifiers. After selecting the two best features for use in a hierarchical classifier, 83% correct classification was achieved. One of these features was specifically designed to remove poorly focused objects. With maximum likelihood discrimination using all 22 features, an overall correct classification rate of 90% was obtained.     

Automatic diagnosis of mammographic abnormalities based on hybrid features with learning classifier

Breast cancer screening is currently performed by mammography, which is limited by overlying anatomy and dense breast tissue. Computer-aided detection (CAD) systems can serve as a double reader to improve radiologist performance. In this paper, we have applied a novel approach to segmentation of suspicious region by mammogram and classification based on hybrid features with learning classifier. We formulated differentiation of lesion from normal tissue as a supervised learning problem, and applied this learning method to develop the classification algorithm. The algorithm has been verified with 164 mammograms in the mini Mammographic Image Analysis Society database. The experimental results show that the detection method has a sensitivity of 94.5% at 0.26 false positives per image. The efficiency of algorithm is measured using free receiver operating characteristics curve and the results are highlighted. We conclude that CAD technology with learning classifier has the potential to help radiologists with the task of discriminating between lesion and normal tissues.     

基于复合叶片特征的计算机植物识别方法

该文探讨如何根据植物的叶片特征,利用图像处理和机器学习的方法对植物进行分类。鉴于现有的叶片分类系统多采用单一的特征,如几何和纹理等,仅能在小规模数据库上得到较好的结果。然而,随着样本种类的增多,单一特征在不同种类叶片之间的相似性非常明显,致使分类正确率降低。该研究使用多种复合特征,并提出了原创的预处理方法以及宽度、叶缘频率特征,较传统的几何特征更为详尽。研究结果显示,复合特征可以有效避免算法过拟合问题,使之适用于更大的数据库。通过提取21类植物的叶片宽度、颜色、叶缘和纹理共292维特征,对1 915张数字图像进行了分类,正确率达到93%,并分析了各类特征对分类结果的影响。研究结果表明,在不影响分类正确率前提下,可将特征减少到约100维。     

Application of LANDSAT MSS Digital Image Processing Techniques in Investigation of Aquatic Plant Distribution in Honghu Lake

In this paper an experiment in the application of LANDSAT MSS digital image proces. sing technique to classify the aquatic plants in the Honghu Lake, China, as well as the necessary ground feature spectrum measurement and plant chlorophyll content determination techniques are introduced. In order to obtain the expected computer aided classification result different methods of digital image feature extraction have been tried, namely the ratio transformation, the biomass index transformation and the linear stretching of image intensity. The data from spectrum measurement and chlorophyll content determination were used to distinguish the attributions of different classes of aquatic plant associations from one another by comparing with their spectrum response intensity values on different LANDSAT band images. The original LANDSAT MSS images were digitally rectified by control points to a certain map projection system using a general polynomial approach scheme before the classifying activities, so that the classification result may be transfered to an existing map. In this paper the resulting colourassigued image and the thematic map of Honghu Lakes aquatic plant (association) distribution are included, the areas of different classes of aquatic plants are listed and the reliability of the resulting classification and pattern recognition are analysed. These results will provide useful informations for the investigation of the present situation as well as the historical succession of the Honghu Lake and its aquatic plant distribution, and also for the research works of the lake management.     

Small bowel image classification using cross-co-occurrence matrices on wavelet domain

This paper presents a novel system to compute the automated classification of wireless capsule endoscope images. Classification is achieved by a classical statistical approach, but novel features are extracted from the wavelet domain and they contain both color and texture information. First, a shift-invariant discrete wavelet transform (SIDWT) is computed to ensure that the multiresolution feature extraction scheme is robust to shifts. The SIDWT expands the signal (in a shift-invariant way) over the basis functions which maximize information. Then cross-co-occurrence matrices of wavelet subbands are calculated and used to extract both texture and color information. Canonical discriminant analysis is utilized to reduce the feature space and then a simple 1D classifier with the leave one out method is used to automatically classify normal and abnormal small bowel images. A classification rate of 94.7% is achieved with a database of 75 images (41 normal and 34 abnormal cases). The high success rate could be attributed to the robust feature set which combines multiresolutional color and texture features, with shift, scale and semi-rotational invariance. This result is very promising and the method could be used in a computer-aided diagnosis system or a content-based image retrieval scheme.     

Comparison of three hexagonal tessellations through extraction of blood cell geometric features

Although image data are almost universally acquired on rectangular sampling lattices, the regular hexagonal lattice offers important theoretical advantages for tessellation of images, particularly when subsequent processing involves operations on local image neighborhoods. The few systems capable of processing hexagonally tessellated images have approximated this tessellation by using image data acquired on a rectangular sampling lattice, from which six of the eight image samples were selected from each local neighborhood. This paper describes a simple method of directly acquiring image data in hexagonal image tessellations; the method is used to compare at constant sampling density the most common of these approximating image tessellations with both a nonregular and a regular hexagonal image tessellation. The test objects were human blood cells, from which features describing cellular geometry were extracted for each image tessellation. Compared to the approximating tessellation, the nonregular tessellation tended to decrease feature means and increase feature variances. In contrast, the regular tessellation tended to increase feature means and decrease feature variances. Consequently, the extracted features showed subtle but consistent differences, with decreasing anisotropic effects and data dispersion for the regular tessellation. In addition, cells contacting others near the 45 degree diagonals were more readily segmented when the image was tessellated on the regular lattice. Expected to be general, these trends recommend use of the regular tessellation, especially when classification accuracy may depend on small differences in several similar geometric features.     

基于不同决策树的面向对象林区遥感影像分类比较

面向地理对象影像分析技术(GEOBIA)是影像分辨率越来越高的背景下的产物.如何提高高分辨率影像分类精度和分类效率是影像处理的重要议题之一.本研究对QuickBird影像多尺度分割后的对象进行分类,分析了C5.0、C4.5、CART决策树算法在林区面向对象分类中的效率,并与kNN算法的分类精度进行比较.利用eCognition软件对遥感影像进行多尺度分割,分析得到最佳尺度为90和40.在90尺度下分离出植被和非植被后,在40尺度下提取不同类别植被的光谱、纹理、形状等共21个特征,并利用C5.0、C4.5、CART决策树算法分别对其进行知识挖掘,自动建立分类规则.最后利用建立的分类规则分别对植被区域进行分类,并比较分析其精度.结果表明: 基于决策树的分类精度均高于传统的kNN法.其中,C5.0方法的精度最高,其总体分类精度为90.0%,Kappa系数0.87.决策树算法能有效提高林区树种分类精度,且C5.0决策树的Boosting算法对该分类效果具有最明显的提升.     

基于SIFT特征和近似最近邻算法的医学CT图像检索

针对医学X线计算机断层（Computed Tomography,CT）图像,提出了一种基于尺度不变特征变换（Scale InvariantFeature Transform,SIFT）特征和近似最近邻算法的检索方法。首先通过SIFT算法得到图像的特征点和相应的特征向量,再采用近似最近邻算法进行SIFT特征向量的匹配搜索,得到数据库中与参考图像最相似的图像序列。实验结果表明,该法能检索到与目标图像细节相符的结果,大大提高了检索速度。与传统的基于纹理的检索方法相比,查准率和检索结果与目标图像的相似程度方面更佳,符合医学CT图像检索的要求。