Medical Image Retrieval with Compact Binary Codes Generated in Frequency Domain Using Highly Reactive Convolutional Features

Efficient retrieval of relevant medical cases using semantically similar medical images from large scale repositories can assist medical experts in timely decision making and diagnosis. However, the ever-increasing volume of images hinder performance of image retrieval systems. Recently, features from deep convolutional neural networks (CNN) have yielded state-of-the-art performance in image retrieval. Further, locality sensitive hashing based approaches have become popular for their ability to allow efficient retrieval in large scale datasets. In this paper, we present a highly efficient method to compress selective convolutional features into sequence of bits using Fast Fourier Transform (FFT). Firstly, highly reactive convolutional feature maps from a pre-trained CNN are identified for medical images based on their neuronal responses using optimal subset selection algorithm. Then, layer-wise global mean activations of the selected feature maps are transformed into compact binary codes using binarization of its Fourier spectrum. The acquired hash codes are highly discriminative and can be obtained efficiently from the original feature vectors without any training. The proposed framework has been evaluated on two large datasets of radiology and endoscopy images. Experimental evaluations reveal that the proposed method significantly outperforms other features extraction and hashing schemes in both effectiveness and efficiency.     

手掌静脉识别：基于端到端卷积神经网络方法

目的提出一种基于端到端卷积神经网络的手掌静脉识别方法。方法在构建的手掌静脉识别网络模型中,卷积层和池化 层交替级联提取图像特征,同时通过神经网络分类器进行分类识别,采用包含动量项的随机梯度下降法最小化识别误差,在误 差减小的方向上不断优化模型。采用训练集数据扩展、批归一化、Dropout、L2参数正则化四种方法提升网络的泛化能力。结果 对公共的PolyU库(图像在高约束条件下获取)和自建库（图像在自然条件下获取）中全部500个对象的识别,正确识别率分别达 到99.90%和98.05%,单个样本的识别时间均小于9 ms。结论与传统算法相比,本文方法能够有效提升掌静脉识别在实际应用 中的准确率,为掌静脉识别提供一种新思路。     

基于多维度特征融合的深度学习骨龄评估模型

目的 通过基于特征提取的深度卷积神经网络,结合关键区域特征和人口学信息,评估儿童骨龄。方法 自动识别左手X线图像数据,对图像进行预处理,使用基于深度神经网络的X线图像分析方法,实现左手关节骨龄17个关键区域特征的自动提取,再将骨龄影像特征与临床大数据（人口统计、性别）融合训练骨龄评估模型,测试模型的评估效能。结果 使用基于深度学习的特征区域提取方法比传统图像分析方法可以更好地提取特征信息,结合临床信息从另一维度补充了骨龄发育信息。基于多维度数据特征融合的骨龄评估模型检测得到的骨龄平均绝对误差为0.455,优于传统方法和仅端到端的深度学习方法。结论 相较传统的机器学习特征提取方法,基于特征提取的深度卷积神经网络在骨龄回归模型上有更好的表现,结合人口和性别信息可进一步提升基于图像的骨龄预测准确率。     

Automated Diagnosis of Alzheimer Disease using the Scale-Invariant Feature Transforms in Magnetic Resonance Images

In this paper we present an automated method for diagnosing Alzheimer disease (AD) from brain MR images. The approach uses the scale-invariant feature transforms (SIFT) extracted from different slices in MR images for both healthy subjects and subjects with Alzheimer disease. These features are then clustered in a group of features which they can be used to transform a full 3-dimensional image from a subject to a histogram of these features. A feature selection strategy was used to select those bins from these histograms that contribute most in classifying the two groups. This was done by ranking the features using the Fisher’s discriminant ratio and a feature subset selection strategy using the genetic algorithm. These selected bins of the histograms are then used for the classification of healthy/patient subjects from MR images. Support vector machines with different kernels were applied to the data for the discrimination of the two groups, namely healthy subjects and patients diagnosed by AD. The results indicate that the proposed method can be used for diagnose of AD from MR images with the accuracy of %86 for the subjects aged from 60 to 80 years old and with mild AD.     

基于耦合的卷积-图卷积神经网络的阿尔茨海默病的磁共振诊断方法

目的 针对已有方法未利用大脑拓扑信息的问题,提出基于耦合的卷积-图卷积神经网络的疾病诊断模型,以实现对阿尔茨海默病及其前驱症状的精确诊断,为临床提供可靠的辅助诊断信息。方法 根据ADNI数据库提供的信息,将MMSE评分在20~26分、同时CDR评分为0.5或1的被试的疾病标签标记为AD组;将MMSE评分在24~30分且CDR评分为0、无抑郁症状、无认知障碍、无焦虑症状的被试疾病标签标记为NC组。本文提出一种耦合的卷积-图卷积神经网络（CCGCN）模型,以组间比较获取的疾病相关区域作为输入,利用卷积神经网络,从大脑磁共振图像的不同区域提取疾病相关的特征,再使用图卷积网络,结合提取到的特征,对区域间拓扑结构进行建模,并在图卷积网络中嵌入图池化操作,从而自适应地学习大脑拓扑结构与疾病诊断任务之间的内在联系。利用ADNI数据集,获得CCGCN模型对阿尔茨海默病及其前驱症状的疾病诊断准确率、灵敏度和特异度,并进行模型结构的消融实验。结果 该模型在阿尔茨海默病的诊断任务上取得了92.5%的准确率、88.1%的灵敏度和96.0%的特异度,诊断精度优于目前最先进的方法;同时在区分进行型轻度认知障碍患者和稳定型轻度认知障碍患者的任务上取得了79.8%的准确率、55.3%的灵敏度和83.7%的特异度;消融实验的结果显示了CCGCN模型各组成成分的有效性。结论 基于耦合的卷积-图卷积神经网络的疾病诊断模型利用了原始图像的结构和拓扑信息,相比现有方法可以提供更加精确的阿尔茨海默病诊断结果,有望将其应用于临床的辅助诊断中。     

An Improved Pathological Brain Detection System Based on Two-Dimensional PCA and Evolutionary Extreme Learning Machine

Pathological brain detection has made notable stride in the past years, as a consequence many pathological brain detection systems (PBDSs) have been proposed. But, the accuracy of these systems still needs significant improvement in order to meet the necessity of real world diagnostic situations. In this paper, an efficient PBDS based on MR images is proposed that markedly improves the recent results. The proposed system makes use of contrast limited adaptive histogram equalization (CLAHE) to enhance the quality of the input MR images. Thereafter, two-dimensional PCA (2DPCA) strategy is employed to extract the features and subsequently, a PCA+LDA approach is used to generate a compact and discriminative feature set. Finally, a new learning algorithm called MDE-ELM is suggested that combines modified differential evolution (MDE) and extreme learning machine (ELM) for segregation of MR images as pathological or healthy. The MDE is utilized to optimize the input weights and hidden biases of single-hidden-layer feed-forward neural networks (SLFN), whereas an analytical method is used for determining the output weights. The proposed algorithm performs optimization based on both the root mean squared error (RMSE) and norm of the output weights of SLFNs. The suggested scheme is benchmarked on three standard datasets and the results are compared against other competent schemes. The experimental outcomes show that the proposed scheme offers superior results compared to its counterparts. Further, it has been noticed that the proposed MDE-ELM classifier obtains better accuracy with compact network architecture than conventional algorithms.     

Study on Contribution of Biological Interpretable and Computer-Aided Features Towards the Classification of Childhood Medulloblastoma Cells

Diagnosis and Prognosis of brain tumour in children is always a critical case. Medulloblastoma is that subtype of brain tumour which occurs most frequently amongst children. Post-operation, the classification of its subtype is most vital for further clinical management. In this paper a novel approach of pathological subtype classification using biological interpretable and computer-aided textural features is forwarded. The classifier for accurate features prediction is built purely on the feature set obtained by segmentation of the ground truth cells from the original histological tissue images, marked by an experienced pathologist. The work is divided into five stages: marking of ground truth, segmentation of ground truth images, feature extraction, feature reduction and finally classification. Kmeans colour segmentation is used to segment out the ground truth cells from histological images. For feature extraction we used morphological, colour and textural features of the cells followed by feature reduction using Principal Component Analysis. Finally both binary and multiclass classification is done using Support Vector Method (SVM). The classification was compared using six different classifiers and performance was evaluated employing five-fold cross-validation technique. The accuracy achieved for binary and multiclass classification before applying PCA were 95.4 and 62.1% and after applying PCA were 100 and 84.9% respectively. The run-time analysis are also shown. Results reveal that this technique of cell level classification can be successfully adopted as architectural view can be confusing. Moreover it conforms substantially to the pathologist’s point of view regarding morphological and colour features, with the addition of computer assisted texture feature.     

基于3D深度卷积神经网络依据MRI生成伪CT的研究

目的:研究一种依据MRI生成伪CT的方法，从而减少放疗过程中额外CT的使用，降低患者辐射剂量，提高放疗精准度。方法:提出一种基于3D深度卷积神经网络（DCNN）的预测算法，利用单张图像的解剖特征以及相邻图像层之间的关联信息，从而提高了图像特征提取的准确性。采用U-net网络结构，通过编码部分的卷积层、池化层和解码部分的上采样、卷积层，对MRI和对应的CT进行端到端转换的学习。采集13例患者图像数据，应用留一交叉验证的方法，分别对3D DCNN和2D DCNN的伪CT结果与原始CT进行对照比较。结果:提出的3D DCNN算法的平均绝对误差（MAE）为86 HU，远小于2D DCNN的136 HU。结论:3D DCNN算法能更准确的生成伪CT，明显改善了骨骼、空气与软组织之间的误转化。     

生成对抗网络的血管内超声图像超分辨率重建

针对超声图像分辨率低导致视觉效果差的问题,本文以超分辨率重建为基础,结合生成对抗网络的方法,生成相对原图更 加清晰的血管内超声图像,用于辅助医生诊断与治疗。本方法应用生成对抗网络,生成器生成图像,判别器判断图像真伪。其 过程：低分辨率图像经过亚像素卷积层r2个特征通道,产生尺寸大小相同的r2个特征图,对每个特征图中相对应的同一像素重新 排列成一个r×r的子块,其对应高分辨率图像中的某一个子块,经过放大,产生r2倍的高分辨率图像。生成对抗网络经过不断优 化,获得更优质清晰的图像。将本方法（SRGAN）得出的结果与双立方插值（Bicubic）、超分辨率卷积网络（SRCNN）和亚像素卷 积网络（ESPCN）等方法比较,其峰值信噪比（PSNR）和结构相似性（SSIM）分别提高2.369dB和1.79%。因此,我们得知：结合生 成对抗网络的图像超分辨率重建能获得很好的血管内超声图像诊断视觉效果。     

A Machine Learning Ensemble Classifier for Early Prediction of Diabetic Retinopathy

The main complication of diabetes is Diabetic retinopathy (DR), retinal vascular disease and it leads to the blindness. Regular screening for early DR disease detection is considered as an intensive labor and resource oriented task. Therefore, automatic detection of DR diseases is performed only by using the computational technique is the great solution. An automatic method is more reliable to determine the presence of an abnormality in Fundus images (FI) but, the classification process is poorly performed. Recently, few research works have been designed for analyzing texture discrimination capacity in FI to distinguish the healthy images. However, the feature extraction (FE) process was not performed well, due to the high dimensionality. Therefore, to identify retinal features for DR disease diagnosis and early detection using Machine Learning and Ensemble Classification method, called, Machine Learning Bagging Ensemble Classifier (ML-BEC) is designed. The ML-BEC method comprises of two stages. The first stage in ML-BEC method comprises extraction of the candidate objects from Retinal Images (RI). The candidate objects or the features for DR disease diagnosis include blood vessels, optic nerve, neural tissue, neuroretinal rim, optic disc size, thickness and variance. These features are initially extracted by applying Machine Learning technique called, t-distributed Stochastic Neighbor Embedding (t-SNE). Besides, t-SNE generates a probability distribution across high-dimensional images where the images are separated into similar and dissimilar pairs. Then, t-SNE describes a similar probability distribution across the points in the low-dimensional map. This lessens the Kullback–Leibler divergence among two distributions regarding the locations of the points on the map. The second stage comprises of application of ensemble classifiers to the extracted features for providing accurate analysis of digital FI using machine learning. In this stage, an automatic detection of DR screening system using Bagging Ensemble Classifier (BEC) is investigated. With the help of voting the process in ML-BEC, bagging minimizes the error due to variance of the base classifier. With the publicly available retinal image databases, our classifier is trained with 25% of RI. Results show that the ensemble classifier can achieve better classification accuracy (CA) than single classification models. Empirical experiments suggest that the machine learning-based ensemble classifier is efficient for further reducing DR classification time (CT).     

Feature Extraction of Chinese Materia Medica Fingerprint Based on Star Plot Representation of Multivariate Data

Objective To study a novel feature extraction method of Chinese materia medica (CMM) fingerprint. Methods On the basis of the radar graphical presentation theory of multivariate, the radar map was used to figure the non-map parameters of the CMM fingerprint, then to extract the map features and to propose the feature fusion. Results Better performance was achieved when using this method to test data. Conclusion This shows that the feature extraction based on radar chart presentation can mine the valuable features that facilitate the identification of Chinese medicine.     

基于深度卷积神经网络的脑部图像疾病类别特征库构建

介绍深度卷积神经网络基本理论,阐述基于深度卷积神经网络的脑部图像视觉特征提取,设计一种适用于脑部疾病图像的分类器,进而实现脑部图像疾病类别特征库构建,为基于脑部图像疾病类别特征库开展临床辅助决策等应用提供可能。     

Enhanced Cancer Recognition System Based on Random Forests Feature Elimination Algorithm

Accurate classifiers are vital to design precise computer aided diagnosis (CADx) systems. Classification performances of machine learning algorithms are sensitive to the characteristics of data. In this aspect, determining the relevant and discriminative features is a key step to improve performance of CADx. There are various feature extraction methods in the literature. However, there is no universal variable selection algorithm that performs well in every data analysis scheme. Random Forests (RF), an ensemble of trees, is used in classification studies successfully. The success of RF algorithm makes it eligible to be used as kernel of a wrapper feature subset evaluator. We used best first search RF wrapper algorithm to select optimal features of four medical datasets: colon cancer, leukemia cancer, breast cancer and lung cancer. We compared accuracies of 15 widely used classifiers trained with all features versus to extracted features of each dataset. The experimental results demonstrated the efficiency of proposed feature extraction strategy with the increase in most of the classification accuracies of the algorithms.     

Computer Aided Diagnosis System for Breast Cancer Based on Color Doppler Flow Imaging

Color Doppler flow imaging takes a great value in diagnosing and classifying benign and malignant breast lesions. However, scanning of color Doppler sonography is operator-dependent and ineffective. In this paper, a novel breast classification system based on B-Mode ultrasound and color Doppler flow imaging is proposed. First, different feature extraction methods were used to obtain the texture and geometric features from B-Mode ultrasound images. In color Doppler feature extraction stage, several spectrum features are extracted by applying blood flow velocity analysis to Doppler signals. Moreover, a velocity coherent vector method is proposed based on color coherence vector, which is helpful for designing to the optimize detection of flow indices from different blood flow velocity fields automatically. Finally, a support vector machine classifier with selected feature vectors is used to classify breast tumors into benign and malignant. The experimental results demonstrate that the proposed computer-aided diagnosis system is useful for reducing the unnecessary biopsy and death rate.     

基于改进的GHSOM聚类算法的图像检索

传统的图像检索需要顺序比较图像库中的图像与请求图像的相似度,检索速度和检索准确度都很低。针对此问题,提出了一种基于改进的增长型分层自组织映射网络（GHSOM）的图像检索方法。先将图像特征库用改进的GHSOM算法进行聚类,在图像检索时先在GHSOM网络模型上找到相似的类,然后在相似的类上继续进行检索,大大提高了检索效率。并且在搜索相似的类时充分利用GHSOM网络的分层结构,更进一步地提高了检索效率。改进的GHSOM网络根据算法的特点构建了赤迟信息量（AIC）准则,用该准则来选择每个独立的SOM网络的生长参数,使得每个网络都能正确地表达映射到它的数据集的结构,提高GHSOM网络的聚类效果,从而提高检索的准确性。实验结果表明,改进的GHSOM算法得到了更好的聚类效果,基于它的图像检索方法提高了将近3倍的图像匹配速度,同时图像检索准确率也得到了一定程度的提高。     

Earlier detection of Alzheimer disease using N-fold cross validation approach

According to the recent study, world-wide 40 million patients are affected by Alzheimer disease (AD) because it is one of the dangerous neurodegenerative disorders. This AD disease has less symptoms such as short term memory loss, mood swings, problem with language understanding and behavioral issues. Due to these low symptoms, AD disease is difficult to recognize in the early stage. So, the automated computer aided system need to be developed for recognizing the AD disease for minimizing the mortality rate. Initially, brain MRI image is collected from patients which are processed by applying different processing steps such as noise removal, segmentation, feature extraction, feature selection and classification. The captured MRI image has noise that is eliminated by applying the Lucy–Richardson approach which examines the each pixel in the image and removes the Gaussian noise which also eliminates the blur image. After eliminating the noise pixel from the image, affected region is segmented by Prolong adaptive exclusive analytical Atlas approach. From the segmented region, different GLCM statistical features are extracted and optimal features subset is selected by applying the hybrid wrapper filtering approach. This selected features are analyzed by N-fold cross validation approach which recognizes the AD related features successfully. Then the efficiency of the system is evaluated with the help of MATLAB based experimental results, in which Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset images are utilized for examining the efficiency in terms of sensitivity, specificity, ROC curve and accuracy.     

基于PCA和KPCA特征抽取的SVM网络入侵检测方法

提出一种新颖的基于特征抽取的异常检测方法,应用主分量分析(PCA)和核主分量分析(KPCA)抽取入侵特征,再应用支持向量机(SVM)检测入侵。其中PCA对输入特征做线性变换,而KPCA通过核函数进行非线性变换。利用KDD 99数据集,将PCA-SVM、KPCA-SVM与SVM、PCR、KPCR进行比较,结果显示:在不降低分类器性能的情况下,特征抽取方法能对输入数据有效降维。在各种方法中,KPCA与SVM的结合能得到最优入侵检测性能。     

Symptomatic vs. Asymptomatic Plaque Classification in Carotid Ultrasound

Quantitative characterization of carotid atherosclerosis and classification into symptomatic or asymptomatic type is crucial in both diagnosis and treatment planning. This paper describes a computer-aided diagnosis (CAD) system which analyzes ultrasound images and classifies them into symptomatic and asymptomatic based on the textural features. The proposed CAD system consists of three modules. The first module is preprocessing, which conditions the images for the subsequent feature extraction. The feature extraction stage uses image texture analysis to calculate Standard deviation, Entropy, Symmetry, and Run Percentage. Finally, classification is performed using AdaBoost and Support Vector Machine for automated decision making. For Adaboost, we compared the performance of five distinct configurations (Least Squares, Maximum- Likelihood, Normal Density Discriminant Function, Pocket, and Stumps) of this algorithm. For Support Vector Machine, we compared the performance using five different configurations (linear kernel, polynomial kernel configurations of different orders and radial basis function kernels). SVM with radial basis function kernel for support vector machine presented the best classification result: classification accuracy of 82.4%, sensitivity of 82.9%, and specificity of 82.1%. We feel that texture features coupled with the Support Vector Machine classifier can be used to identify the plaque tissue type. An Integrated Index, called symptomatic asymptomatic carotid index (SACI), is proposed using texture features to discriminate symptomatic and asymptomatic carotid ultrasound images using just one index or number. We hope this SACI can be used as an adjunct tool by the vascular surgeons for daily screening.     

Highly Sensitive Computer Aided Diagnosis System for Breast Tumor Based on Color Doppler Flow Images

A computer-aided diagnosis (CAD) system for breast tumor based on color Doppler flow images is proposed. Our system consists of automatic segmentation, feature extraction, and classification of breast tumors. First, the B-mode grayscale image containing anatomical information was separated from a color Doppler flow image (CDFI). Second, the boundary of the breast tumor was automatically defined in the B-mode image and then morphologic and gray features were extracted. Third, an optimal feature vector was created using K-means cluster algorithm. Then a back-propagation (BP) artificial neural network (ANN) was used to classify breast tumors as benign, malignant or uncertain. Finally, the blood flow feature was extracted selectively from the CDFI, and was used to classify the uncertain tumor as benign or malignant. Experiments on 500 cases show that the proposed system yields an accuracy of 100% for the malignant and 80.8% for the benign classification. Comparing with other systems, the advantage of our system is that it has a much lower percentage of malignant tumor misdiagnosis.