Microscopic skin laceration segmentation and classification: A framework of statistical normal distribution and optimal feature selection

Among precision medical techniques, medical image processing is rapidly growing as a successful tool for cancer detection. Skin cancer is one of the crucial cancer types. It is identified through computer vision (CV) techniques using dermoscopic images. The early diagnosis skin cancer from dermoscopic images can be decrease the mortality rate. We propose an automated system for skin lesion detection and classification based on statistical normal distribution and optimal feature selection. Local contrast is controlled using a brighter channel enhancement technique, and segmentation is performed through a statistical normal distribution approach. The multiplication law of probability is implemented for the fusion of segmented images. In the feature extraction phase, optimized histogram, optimized color, and gray level co‐occurrences matrices features are extracted and covariance‐based fusion is performed. Subsequently, optimal features are selected through a binary grasshopper optimization algorithm. The selected optimal features are finally fed to a classifier and evaluated on the ISBI 2016 and ISBI 2017 data sets. Classification accuracy is computed using different Support Vector Machine (SVM) kernel functions, and the best accuracy is obtained for the cubic function. The average accuracies of the proposed segmentation on the PH2 and ISBI 2016 data sets are 93.79 and 96.04%, respectively, for an image size 512 × 512. The accuracies of the proposed classification on the ISBI 2016 and ISBI 2017 data sets are 93.80 and 93.70%, respectively. The proposed system outperforms existing methods on selected data sets.     

Nuclear spatial and spectral features based evolutionary method for meningioma subtypes classification in histopathology

Meningioma subtypes classification is a real‐world multiclass problem from the realm of neuropathology. The major challenge in solving this problem is the inherent complexity due to high intra‐class variability and low inter‐class variation in tissue samples. The development of computational methods to assist pathologists in characterization of these tissue samples would have great diagnostic and prognostic value. In this article, we proposed an optimized evolutionary framework for the classification of benign meningioma into four subtypes. This framework investigates the imperative role of RGB color channels for discrimination of tumor subtypes and compute structural, statistical and spectral phenotypes. An evolutionary technique, Genetic Algorithm, in combination with Support Vector Machine is applied to tune classifier parameters and to select the best possible combination of extracted phenotypes that improved the classification accuracy (94.88%) on meningioma histology dataset, provided by the Institute of Neuropathology, Bielefeld. These statistics show that computational framework can robustly discriminate four subtypes of benign meningioma and may aid pathologists in the diagnosis and classification of these lesions.     

Authentication of bee pollen grains in bright‐field microscopy by combining one‐class classification techniques and image processing

A novel method for authenticating pollen grains in bright‐field microscopic images is presented in this work. The usage of this new method is clear in many application fields such as bee‐keeping sector, where laboratory experts need to identify fraudulent bee pollen samples against local known pollen types. Our system is based on image processing and one‐class classification to reject unknown pollen grain objects. The latter classification technique allows us to tackle the major difficulty of the problem, the existence of many possible fraudulent pollen types, and the impossibility of modeling all of them. Different one‐class classification paradigms are compared to study the most suitable technique for solving the problem. In addition, feature selection algorithms are applied to reduce the complexity and increase the accuracy of the models. For each local pollen type, a one‐class classifier is trained and aggregated into a multiclassifier model. This multiclassification scheme combines the output of all the one‐class classifiers in a unique final response. The proposed method is validated by authenticating pollen grains belonging to different Spanish bee pollen types. The overall accuracy of the system on classifying fraudulent microscopic pollen grain objects is 92.3%. The system is able to rapidly reject pollen grains, which belong to nonlocal pollen types, reducing the laboratory work and effort. The number of possible applications of this authentication method in the microscopy research field is unlimited. Microsc. Res. Tech. 2012. © 2012 Wiley Periodicals, Inc.     

发动机异常检测多目标优化方法

针对柴油发动机异常检测中的特征选择和分类器参数与检测精度之间的耦合关系,提出了一种基于非支配排序粒子群优化的柴油发动机异常检测封装式多目标同步优化方法.利用双树复小波包的分解与重构,对发动机振动信号进行时域、频域和时频域多角度特征提取,构建了较完备的特征参数集,分析了故障诊断中特征选择与分类器参数优化对检测精度的影响,运用非支配排序粒子群优化算法对多个优化目标进行协调和折衷处理,同时追求特征参数子集维数最小化和分类正确率最大化.实验数据分析表明,该方法能够寻找出最优的特征子集和分类器参数,提高柴油发动机异常检测的精度和效率.     

Detection of microscopic glaucoma through fundus images using deep transfer learning approach

Glaucoma disease in humans can lead to blindness if it progresses to the point where it affects the oculus' optic nerve head. It is not easily detected since there are no symptoms, but it can be detected using tonometry, ophthalmoscopy, and perimeter. However, advances in artificial intelligence approaches have permitted machine learning techniques to diagnose at an early stage. Numerous methods have been proposed using Machine Learning to diagnose glaucoma with different data sets and techniques but these are complex methods. Although, medical imaging instruments are used as glaucoma screening methods, fundus imaging specifically is the most used screening technique for glaucoma detection. This study presents a novel DenseNet and DarkNet combination to classify normal and glaucoma affected fundus image. These frameworks have been trained and tested on three data sets of high-resolution fundus (HRF), RIM 1, and ACRIMA. A total of 658 images have been used for healthy eyes and 612 images for glaucoma-affected eyes classification. It has also been observed that the fusion of DenseNet and DarkNet outperforms the two CNN networks and achieved 99.7% accuracy, 98.9% sensitivity, 100% specificity for the HRF database. In contrast, for the RIM1 database, 89.3% accuracy, 93.3% sensitivity, 88.46% specificity has been attained. Moreover, for the ACRIMA database, 99% accuracy, 100% sensitivity, 99% specificity has been achieved. Therefore, the proposed method is robust and efficient with less computational time and complexity compared to the literature available.     

Lungs nodule detection framework from computed tomography images using support vector machine

The emergence of cloud infrastructure has the potential to provide significant benefits in a variety of areas in the medical imaging field. The driving force behind the extensive use of cloud infrastructure for medical image processing is the exponential increase in the size of computed tomography (CT) and magnetic resonance imaging (MRI) data. The size of a single CT/MRI image has increased manifold since the inception of these imagery techniques. This demand for the introduction of effective and efficient frameworks for extracting relevant and most suitable information (features) from these sizeable images. As early detection of lungs cancer can significantly increase the chances of survival of a lung scanner patient, an effective and efficient nodule detection system can play a vital role. In this article, we have proposed a novel classification framework for lungs nodule classification with less false positive rates (FPRs), high accuracy, sensitivity rate, less computationally expensive and uses a small set of features while preserving edge and texture information. The proposed framework comprises multiple phases that include image contrast enhancement, segmentation, feature extraction, followed by an employment of these features for training and testing of a selected classifier. Image preprocessing and feature selection being the primary steps—playing their vital role in achieving improved classification accuracy. We have empirically tested the efficacy of our technique by utilizing the well‐known Lungs Image Consortium Database dataset. The results prove that the technique is highly effective for reducing FPRs with an impressive sensitivity rate of 97.45%.     

基于迁移学习与特征融合的眼底图像分类

针对大量眼底图片难以收集和标注、有经验的眼科医生地区分配不均匀等,导致眼底疾病患者检查准确度低、花费时间较长等问题,本文基于迁移学习提出一种图像分类方法:首先修改EfficientNet-B0和EfficientNet-B7模型并进行参数微调,将微调后的模型作为特征提取器用于提取眼底图像的特征,再对提取的特征进行特征融...     

Classification of stomach infections: A paradigm of convolutional neural network along with classical features fusion and selection

Automated detection and classification of gastric infections (i.e., ulcer, polyp, esophagitis, and bleeding) through wireless capsule endoscopy (WCE) is still a key challenge. Doctors can identify these endoscopic diseases by using the computer‐aided diagnostic (CAD) systems. In this article, a new fully automated system is proposed for the recognition of gastric infections through multi‐type features extraction, fusion, and robust features selection. Five key steps are performed—database creation, handcrafted and convolutional neural network (CNN) deep features extraction, a fusion of extracted features, selection of best features using a genetic algorithm (GA), and recognition. In the features extraction step, discrete cosine transform, discrete wavelet transform strong color feature, and VGG16‐based CNN features are extracted. Later, these features are fused by simple array concatenation and GA is performed through which best features are selected based on K‐Nearest Neighbor fitness function. In the last, best selected features are provided to Ensemble classifier for recognition of gastric diseases. A database is prepared using four datasets—Kvasir, CVC‐ClinicDB, Private, and ETIS‐LaribPolypDB with four types of gastric infections such as ulcer, polyp, esophagitis, and bleeding. Using this database, proposed technique performs better as compared to existing methods and achieves an accuracy of 96.5%.     

A new texture and shape based technique for improving meningioma classification

Over the past decade, computer‐aided diagnosis is rapidly growing due to the availability of patient data, sophisticated image acquisition tools and advancement in image processing and machine learning algorithms. Meningiomas are the tumors of brain and spinal cord. They account for 20% of all the brain tumors. Meningioma subtype classification involves the classification of benign meningioma into four major subtypes: meningothelial, fibroblastic, transitional, and psammomatous. Under the microscope, the histology images of these four subtypes show a variety of textural and structural characteristics. High intraclass and low interclass variabilities in meningioma subtypes make it an extremely complex classification problem. A number of techniques have been proposed for meningioma subtype classification with varying performances on different subtypes. Most of these techniques employed wavelet packet transforms for textural features extraction and analysis of meningioma histology images. In this article, a hybrid classification technique based on texture and shape characteristics is proposed for the classification of meningioma subtypes. Meningothelial and fibroblastic subtypes are classified on the basis of nuclei shapes while grey‐level co‐occurrence matrix textural features are used to train a multilayer perceptron for the classification of transitional and psammomatous subtypes. On the whole, average classification accuracy of 92.50% is achieved through the proposed hybrid classifier; which to the best of our knowledge is the highest. Microsc. Res. Tech. 77:862–873, 2014. © 2014 Wiley Periodicals, Inc.     

Automated identification of normoblast cell from human peripheral blood smear images

In this paper, we have presented a new computer‐aided technique for automatic detection of nucleated red blood cells (NRBCs) or normoblast cell from peripheral blood smear image. The proposed methodology initiates with the localization of the nucleated cells by adopting multilevel thresholding approach in smear images. A novel colour space transformation technique has been introduced to differentiate nucleated blood cells [white blood cells (WBCs) and NRBC] from red blood cells (RBCs) by enhancing the contrast between them. Subsequently, special fuzzy c‐means (SFCM) clustering algorithm is applied on enhanced image to segment out the nucleated cell. Finally, nucleated RBC and WBC are discriminated by the random forest tree classifier based on first‐order statistical‐based features. Experimentally, we observed that the proposed technique achieved 99.42% accuracy in automatic detection of NRBC from blood smear images. Further, the technique could be used to assist the clinicians to diagnose a different anaemic condition.     

Cloud‐based decision support system for the detection and classification of malignant cells in breast cancer using breast cytology images

The advancement of computer‐ and internet‐based technologies has transformed the nature of services in healthcare by using mobile devices in conjunction with cloud computing. The classical phenomenon of patient–doctor diagnostics is extended to a more robust advanced concept of E‐health, where remote online/offline treatment and diagnostics can be performed. In this article, we propose a framework which incorporates a cloud‐based decision support system for the detection and classification of malignant cells in breast cancer, while using breast cytology images. In the proposed approach, shape‐based features are used for the detection of tumor cells. Furthermore, these features are used for the classification of cells into malignant and benign categories using Naive Bayesian and Artificial Neural Network. Moreover, an important phase addressed in the proposed framework is the grading of the affected cells, which could help in grade level necessary medical procedures for patients during the diagnostic process. For demonstrating the e effectiveness of the proposed approach, experiments are performed on real data sets comprising of patients data, which has been collected from the pathology department of Lady Reading Hospital of Pakistan. Moreover, a cross‐validation technique has been performed for the evaluation of the classification accuracy, which shows performance accuracy of 98% as compared to physical methods used by a pathologist for the detection and classification of the malignant cell. Experimental results show that the proposed approach has significantly improved the detection and classification of the malignant cells in breast cytology images.     

彩色眼底图像视盘自动定位与分割

针对彩色眼底图像视盘定位时图像边缘高亮环对定位准确率的影响,提出了一种有效的图像预处理方法。针对已有的视盘分割算法中存在的问题,提出了一种结合形态学、椭圆拟合及梯度矢量流(GVF)Snake模型的分割算法。提出的预处理方法首先利用最小二乘法拟合出眼底图像的边界,然后裁剪掉边界的一部分高亮像素点,最后进行视盘定位。视盘分割算法则首先进行血管擦除,然后用椭圆拟合提取初始轮廓,最后使用GVF Snake精确调整视盘边界。用提出的方法对Messidor眼底图像数据库1 200幅图像上进行了实验,结果显示:视盘定位准确率由原来没经过预处理的95.4%提升到了98.7%;视盘分割错误率与当前已知最好的算法相比由12.5%降低到了9.39%。结果表明:提出的眼底图像视盘自动定位与分割方法准确率高、实用性强,可以用于眼科疾病的计算机辅助诊断。     

Automated system for characterization and classification of malaria‐infected stages using light microscopic images of thin blood smears

In this paper, we propose a comprehensive image characterization cum classification framework for malaria‐infected stage detection using microscopic images of thin blood smears. The methodology mainly includes microscopic imaging of Leishman stained blood slides, noise reduction and illumination correction, erythrocyte segmentation, feature selection followed by machine classification. Amongst three‐image segmentation algorithms (namely, rule‐based, Chan–Vese‐based and marker‐controlled watershed methods), marker‐controlled watershed technique provides better boundary detection of erythrocytes specially in overlapping situations. Microscopic features at intensity, texture and morphology levels are extracted to discriminate infected and noninfected erythrocytes. In order to achieve subgroup of potential features, feature selection techniques, namely, F‐statistic and information gain criteria are considered here for ranking. Finally, five different classifiers, namely, Naive Bayes, multilayer perceptron neural network, logistic regression, classification and regression tree (CART), RBF neural network have been trained and tested by 888 erythrocytes (infected and noninfected) for each features’ subset. Performance evaluation of the proposed methodology shows that multilayer perceptron network provides higher accuracy for malaria‐infected erythrocytes recognition and infected stage classification. Results show that top 90 features ranked by F‐statistic (specificity: 98.64%, sensitivity: 100%, PPV: 99.73% and overall accuracy: 96.84%) and top 60 features ranked by information gain provides better results (specificity: 97.29%, sensitivity: 100%, PPV: 99.46% and overall accuracy: 96.73%) for malaria‐infected stage classification.     

Fault diagnosis of rotary kiln using SVM and binary ACO

This paper proposes a novel hybrid algorithm for fault diagnosis of rotary kiln based on a binary ant colony (BACO) and support vector machine (SVM). The algorithm can find a subset selection which is attained through the elimination of the features that produce noise or are strictly correlated with other already selected features. The BACO algorithm can improve classification accuracy with an appropriate feature subset and optimal parameters of SVM. The proposed algorithm is easily implemented and because of use of a simple filter in that, its computational complexity is very low. The performance of the proposed algorithm is evaluated through two real Rotary Cement kiln datasets. The results show that our algorithm outperforms existing algorithms.     

Vehicle Color Recognition with Vehicle-Color Saliency Detection and Dual-Orientational Dimensionality Reduction of CNN Deep Features

Color is one of the most stable attributes of vehicles and often used as a valuable cue in some important applications. Various complex environmental factors, such as illumination, weather, noise and etc., result in the visual characteristics of the vehicle color being obvious diversity. Vehicle color recognition in complex environments has been a challenging task. The state-of-the-arts methods roughly take the whole image for color recognition, but many parts of the images such as car windows; wheels and background contain no color information, which will have negative impact on the recognition accuracy. In this paper, a novel vehicle color recognition method using local vehicle-color saliency detection and dual-orientational dimensionality reduction of convolutional neural network (CNN) deep features has been proposed. The novelty of the proposed method includes two parts: (1) a local vehicle-color saliency detection method has been proposed to determine the vehicle color region of the vehicle image and exclude the influence of non-color regions on the recognition accuracy; (2) dual-orientational dimensionality reduction strategy has been designed to greatly reduce the dimensionality of deep features that are learnt from CNN, which will greatly mitigate the storage and computational burden of the subsequent processing, while improving the recognition accuracy. Furthermore, linear support vector machine is adopted as the classifier to train the dimensionality reduced features to obtain the recognition model. The experimental results on public dataset demonstrate that the proposed method can achieve superior recognition performance over the state-of-the-arts methods.     

彩色视网膜眼底图像血管自动检测方法

为了给视网膜图像配准、光照校正及视网膜内部病理学检测等问题提供有效依据,本文提出一种有效检测及识别彩色视网膜眼底图像血管的全自动方法。针对视网膜可见血管呈长条型管状、局部具有较好直线型结构的形态特点,本文采取适用于条状结构的组合移位滤波响应模型进行特征提取。针对血管和血管末端特征的不同,分别配置对称和非对称的两种滤波模型进行跟踪,利用组合移位滤波模型(对称和非对称)获取到的响应及G通道像素灰度值共同构建特征向量库,采用AdaBoost分类器对各个像素点进行分类判定。基于国际公共数据库DRIVE与STARE的实验结果表明,该方法针对两个标准数据库的分割结果(DRIVE:Accuracy=0.948 9,Sensitivity=0.765 7,Specificity=0.980 9;STARE:Accuracy=0.956 7,Sensitivity=0.771 7,Specificity=0.976 6)均优于已有方法,适用于彩色视网膜眼底图像的计算机辅助定量分析,可作为临床借鉴。     

Fault diagnosis method of rolling bearing using principal component analysis and support vector machine

To effectively extract the fault feature information of rolling bearings and improve the performance of fault diagnosis, a fault diagnosis method based on principal component analysis and support vector machine was presented, and the rolling bearings signals with different fault states were collected. To address the limitation on effectively dealing with the raw vibration signals by the traditional signal processing technology based on Fourier transform, wavelet packet decomposition was employed to extract the features of bearing faults such as outer ring flaking, inner ring flaking, roller flaking and normal condition. Compared with the previous literature on fault diagnosis using principal component analysis (PCA) and support vector machine (SVM), one-to-one and one-to-many algorithms were taken into account. Additionally, the effect of four kernel functions, such as liner kernel function, polynomial kernel function, radial basis function and hyperbolic tangent kernel function, on the performance of SVM classifier was investigated, and the optimal hype-parameters of SVM classifier model were determined by genetic algorithm optimization. PCA was employed for dimension reduction, so as to reduce the computational complexity. The principal components that reached more than 95 % cumulative contribution rate were extracted by PCA and were input into SVM and BP neural network classifiers for identification. Results show that the fault feature dimensionality of the rolling bearing is reduced from 8-dimensions to 5-dimensions, which can still characterize the bearing status effectively, and the computational complexity is reduced as well. Compared with the raw feature set, PCA has a higher fault diagnosis accuracy (more than 97 %), and a shorter diagnosis time relatively. To better verify the superiority of the proposed method, SVM classification results were compared with the results of BP neural network. It is concluded that SVM classifier achieved a better performance than BP neural network classifier in terms of the classification accuracy and time-cost.     

Super-parameter selection for Gaussian-Kernel SVM based on outlier-resisting

The learning ability and generalizing performance of the support vector machine (SVM) mainly relies on the reasonable selection of super-parameters. When the scale of the training sample set is large and the parameter space is huge, the existing popular super-parameter selection methods are impractical due to high computational complexity. In this paper, a novel super-parameter selection method for SVM with a Gaussian kernel is proposed, which can be divided into the following two stages. The first one is choosing the kernel parameter to ensure a sufficiently large number of potential support vectors retained in the training sample set. The second one is screening out outliers from the training sample set by assigning a special value to the penalty factor, and training out the optimal penalty factor from the remained training sample set without outliers. The whole process of super-parameter selection only needs two train-validate cycles. Therefore, the computational complexity of our method is low. The comparative experimental results concerning 8 benchmark datasets show that our method possesses high classification accuracy and desirable training time.     

Integral DOE and MANOVA techniques for classification feature selection: using solder joint defects as an example

This study presents a novel approach for feature selection using an integrated DOE and MANOVA technique to classify solder joint defects for print circuit boards (PCBs). The main selection procedure includes three stages. The first stage adopts a single feature variable selection algorithm to eliminate poorly discriminated feature variables. The second stage, Plackett-Burman (PB) resolution III design, is then constructed to select the remaining feature variables. The MANOVA technique is then used to calculate the Pillai statistic as the response to the PB design of experiment, and statistical analysis is then executed to obtain the optimal multiple feature variables for multiple groups. The discriminate function classifier is used to evaluate the classification results. The experimental analysis results show that the proposed analysis procedure can acquire an optimum subset of features for classification.     

An effective and accurate identification system of Mycobacterium tuberculosis using convolution neural networks

Tuberculosis (TB) remains the leading cause of morbidity and mortality from infectious disease in developing countries. The sputum smear microscopy remains the primary diagnostic laboratory test. However, microscopic examination is always time‐consuming and tedious. Therefore, an effective computer‐aided image identification system is needed to provide timely assistance in diagnosis. The current identification system usually suffers from complex color variations of the images, resulting in plentiful of false object detection. To overcome the dilemma, we propose a two‐stage Mycobacterium tuberculosis identification system, consisting of candidate detection and classification using convolution neural networks (CNNs). The refined Faster region‐based CNN was used to distinguish candidates of M. tuberculosis and the actual ones were classified by utilizing CNN‐based classifier. We first compared three different CNNs, including ensemble CNN, single‐member CNN, and deep CNN. The experimental results showed that both ensemble and deep CNNs were on par with similar identification performance when analyzing more than 19,000 images. A much better recall value was achieved by using our proposed system in comparison with conventional pixel‐based support vector machine method for M. tuberculosis bacilli detection.