Unsupervised Fuzzy Based Vessel Segmentation In Pathological Digital Fundus Images

Performing the segmentation of vasculature in the retinal images having pathology is a challenging problem. This paper presents a novel approach for automated segmentation of the vasculature in retinal images. The approach uses the intensity information from red and green channels of the same retinal image to correct non-uniform illumination in color fundus images. Matched filtering is utilized to enhance the contrast of blood vessels against the background. The enhanced blood vessels are then segmented by employing spatially weighted fuzzy c-means clustering based thresholding which can well maintain the spatial structure of the vascular tree segments. The proposed method’s performance is evaluated on publicly available DRIVE and STARE databases of manually labeled images. On the DRIVE and STARE databases, it achieves an area under the receiver operating characteristic curve of 0.9518 and 0.9602 respectively, being superior to those presented by state-of-the-art unsupervised approaches and comparable to those obtained with the supervised methods.     

Medical Image Analysis using Convolutional Neural Networks: A Review

The science of solving clinical problems by analyzing images generated in clinical practice is known as medical image analysis. The aim is to extract information in an affective and efficient manner for improved clinical diagnosis. The recent advances in the field of biomedical engineering have made medical image analysis one of the top research and development area. One of the reasons for this advancement is the application of machine learning techniques for the analysis of medical images. Deep learning is successfully used as a tool for machine learning, where a neural network is capable of automatically learning features. This is in contrast to those methods where traditionally hand crafted features are used. The selection and calculation of these features is a challenging task. Among deep learning techniques, deep convolutional networks are actively used for the purpose of medical image analysis. This includes application areas such as segmentation, abnormality detection, disease classification, computer aided diagnosis and retrieval. In this study, a comprehensive review of the current state-of-the-art in medical image analysis using deep convolutional networks is presented. The challenges and potential of these techniques are also highlighted.     

An Ensemble Rule Learning Approach for Automated Morphological Classification of Erythrocytes

The analysis of pathophysiological change to erythrocytes is important for early diagnosis of anaemia. The manual assessment of pathology slides is time-consuming and complicated regarding various types of cell identification. This paper proposes an ensemble rule-based decision-making approach for morphological classification of erythrocytes. Firstly, the digital microscopic blood smear images are pre-processed for removal of spurious regions followed by colour normalisation and thresholding. The erythrocytes are segmented from background image using the watershed algorithm. The shape features are then extracted from the segmented image to detect shape abnormality present in microscopic blood smear images. The decision about the abnormality is taken using proposed multiple rule-based expert systems. The deciding factor is majority ensemble voting for abnormally shaped erythrocytes. Here, shape-based features are considered for nine different types of abnormal erythrocytes including normal erythrocytes. Further, the adaptive boosting algorithm is used to generate multiple decision tree models where each model tree generates an individual rule set. The supervised classification method is followed to generate rules using a C4.5 decision tree. The proposed ensemble approach is precise in detecting eight types of abnormal erythrocytes with an overall accuracy of 97.81% and weighted sensitivity of 97.33%, weighted specificity of 99.7%, and weighted precision of 98%. This approach shows the robustness of proposed strategy for erythrocytes classification into abnormal and normal class. The article also clarifies its latent quality to be incorporated in point of care technology solution targeting a rapid clinical assistance.     

基于狄利克雷分布和参数分析的高斯混合模型图像分割算法

传统的高斯混合模型对于含有噪声的图像不能进行有效的分割。针对有噪声图像的分割问题,提出了一种基于狄利克雷分布和参数分析的高斯混合模型图像分割算法。首先采用高斯函数对像素计算先验概率值,然后采用狄利克雷分布和定律关联像素间的邻域信息,并利用梯度下降法优化参数。实验结果表明,本文算法对无噪声和有噪声图像的分割结果比传统方法更有效,误分率更低。     

基于非局部能谱相似特征的基物质分解方法用于双能CT图像去噪

目的 提出一种非局部能谱相似特征引导的双能CT基物质分解方法（NSSD-Net）以抑制低剂量能谱CT基物质图像的相关性噪声。方法 首先构建模型驱动的双能CT迭代分解模型,采用迭代软阈值算法（ISTA）优化分解模型目标函数的求解过程,并利用深度学习技术将此过程展开为迭代分解网络的形式。然后构建非局部能谱相似特征引导的代价函数,约束网络的训练过程。利用双能CT真实病人数据所建立的基物质分解数据集进行评估。将NSSD-Net与2种传统模型驱动的基物质分解方法、1种基于数据驱动的基物质分解方法以及1种基于数据-模型耦合驱动的监督分解方法进行对比实验。结果 与传统模型驱动的基物质分解方法以及数据驱动的基物质分解方法相比,NSSD-Net方法在水和骨基物质分解结果中均获得最高的PNSR指标（31.383和31.444）、最高的SSIM指标（0.970和0.963）以及最低的RMSE指标（2.901和1.633）;与数据-模型耦合驱动的监督分解方法相比,NSSD-Net方法在水和骨基物质分解结果中均获得最高的SSIM指标;临床影像专家的主观图像质量评估结果显示,NSSD-Net方法在水和骨基物质分解结果中图像质量评分均最高（8.625和8.250）,与其他4种对比方法分解性能之间的差异具有统计学意义（P<0.001）。结论 本方法可以获得高质量的基物质分解结果,有效避免训练数据质量问题和模型不可解释问题。     

An Automated Self-Learning Quantification System to Identify Visible Areas in Capsule Endoscopy Images

Visibility in capsule endoscopic images is presently evaluated through intermittent analysis of frames selected by a physician. It is thus subjective and not quantitative. A method to automatically quantify the visibility on capsule endoscopic images has not been reported. Generally, when designing automated image recognition programs, physicians must provide a training image; this process is called supervised learning. We aimed to develop a novel automated self-learning quantification system to identify visible areas on capsule endoscopic images. The technique was developed using 200 capsule endoscopic images retrospectively selected from each of three patients. The rate of detection of visible areas on capsule endoscopic images between a supervised learning program, using training images labeled by a physician, and our novel automated self-learning program, using unlabeled training images without intervention by a physician, was compared. The rate of detection of visible areas was equivalent for the supervised learning program and for our automatic self-learning program. The visible areas automatically identified by self-learning program correlated to the areas identified by an experienced physician. We developed a novel self-learning automated program to identify visible areas in capsule endoscopic images.     

An End-to-End System for Automatic Urinary Particle Recognition with Convolutional Neural Network

The urine sediment analysis of particles in microscopic images can assist physicians in evaluating patients with renal and urinary tract diseases. Manual urine sediment examination is labor-intensive, subjective and time-consuming, and the traditional automatic algorithms often extract the hand-crafted features for recognition. Instead of using the hand-crafted features, in this paper we propose to exploit convolutional neural network (CNN) to learn features in an end-to-end manner to recognize the urinary particle. We treat the urinary particle recognition as object detection and exploit two state-of-the-art CNN-based object detection methods, Faster R-CNN and single shot multibox detector (SSD), along with their variants for urinary particle recognition. We further investigate different factors involving these CNN-based methods to improve the performance of urinary particle recognition. We comprehensively evaluate these methods on a dataset consisting of 5,376 annotated images corresponding to 7 categories of urinary particle, i.e., erythrocyte, leukocyte, epithelial cell, crystal, cast, mycete, epithelial nuclei, and obtain a best mean average precision (mAP) of 84.1% while taking only 72 ms per image on a NVIDIA Titan X GPU.     

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.     

Active learning for clinical text classification: is it better than random sampling?

Objective

This study explores active learning algorithms as a way to reduce the requirements for large training sets in medical text classification tasks.

Design

Three existing active learning algorithms (distance-based (DIST), diversity-based (DIV), and a combination of both (CMB)) were used to classify text from five datasets. The performance of these algorithms was compared to that of passive learning on the five datasets. We then conducted a novel investigation of the interaction between dataset characteristics and the performance results.

Measurements

Classification accuracy and area under receiver operating characteristics (ROC) curves for each algorithm at different sample sizes were generated. The performance of active learning algorithms was compared with that of passive learning using a weighted mean of paired differences. To determine why the performance varies on different datasets, we measured the diversity and uncertainty of each dataset using relative entropy and correlated the results with the performance differences.

Results

The DIST and CMB algorithms performed better than passive learning. With a statistical significance level set at 0.05, DIST outperformed passive learning in all five datasets, while CMB was found to be better than passive learning in four datasets. We found strong correlations between the dataset diversity and the DIV performance, as well as the dataset uncertainty and the performance of the DIST algorithm.

Conclusion

For medical text classification, appropriate active learning algorithms can yield performance comparable to that of passive learning with considerably smaller training sets. In particular, our results suggest that DIV performs better on data with higher diversity and DIST on data with lower uncertainty.     

Medical Image Analysis by Cognitive Information Systems – a Review

This publication presents a review of medical image analysis systems. The paradigms of cognitive information systems will be presented by examples of medical image analysis systems. The semantic processes present as it is applied to different types of medical images. Cognitive information systems were defined on the basis of methods for the semantic analysis and interpretation of information – medical images – applied to cognitive meaning of medical images contained in analyzed data sets. Semantic analysis was proposed to analyzed the meaning of data. Meaning is included in information, for example in medical images. Medical image analysis will be presented and discussed as they are applied to various types of medical images, presented selected human organs, with different pathologies. Those images were analyzed using different classes of cognitive information systems. Cognitive information systems dedicated to medical image analysis was also defined for the decision supporting tasks. This process is very important for example in diagnostic and therapy processes, in the selection of semantic aspects/features, from analyzed data sets. Those features allow to create a new way of analysis.     

Three-Category Classification of Magnetic Resonance Hearing Loss Images Based on Deep Autoencoder

Hearing loss, a partial or total inability to hear, is known as hearing impairment. Untreated hearing loss can have a bad effect on normal social communication, and it can cause psychological problems in patients. Therefore, we design a three-category classification system to detect the specific category of hearing loss, which is beneficial to be treated in time for patients. Before the training and test stages, we use the technology of data augmentation to produce a balanced dataset. Then we use deep autoencoder neural network to classify the magnetic resonance brain images. In the stage of deep autoencoder, we use stacked sparse autoencoder to generate visual features, and softmax layer to classify the different brain images into three categories of hearing loss. Our method can obtain good experimental results. The overall accuracy of our method is 99.5%, and the time consuming is 0.078 s per brain image. Our proposed method based on stacked sparse autoencoder works well in classification of hearing loss images. The overall accuracy of our method is 4% higher than the best of state-of-the-art approaches.     

Brain MR Image Segmentation with Spatial Constrained K-mean Algorithm and Dual-Tree Complex Wavelet Transform

In brain MR images, the noise and low-contrast significantly deteriorate the segmentation results. In this paper, we propose an automatic unsupervised segmentation method integrating dual-tree complex wavelet transform (DT-CWT) with K-mean algorithm for brain MR image. Firstly, a multi-dimensional feature vector is constructed based on the intensity, the low-frequency subband of DT-CWT and spatial position information. Then, a spatial constrained K-mean algorithm is presented as the segmentation system. The proposed method is validated by extensive experiments using both simulated and real T1-weighted MR images, and compared with the state-of-the-art algorithms.     

基于多尺度小波残差网络的稀疏角度CT图像恢复

目的 稀疏角度CT具有加速数据采集和减少辐射剂量的优点。然而,由于采集信息的减少,使用传统滤波反投影算法（FBP）进行重建得到的图像中伴有严重的条形伪影和噪声。针对这一问题,本文提出基于多尺度小波残差网络（MWResNet）对稀疏角度CT图像进行恢复。方法 本网络中将小波网络与残差块相结合,用以增强网络对图像特征的提取能力和加快网络训练效率。实验中使用真实的螺旋几何CT图像数据“Low-dose CT Grand Challenge”数据集训练网络。通过观察图像表征和计算定量参数的方法对结果进行评估,并与其他现有网络进行比较,包括图像恢复迭代残差卷积网络（IRLNet）,残差编码解码卷积神经网络（REDCNN）和FBP卷积神经网络（FBPConvNet）。结果 实验结果表明,本文提出的多尺度小波残差网络优于其余对比方法。结论 本文提出的MWResNet网络能够在保持稀疏角度CT图像边缘细节信息的同时有效抑制噪声和伪影。     

Mobile PHRs Compliance with Android and iOS Usability Guidelines

Recently image fusion has prominent role in medical image processing and is useful to diagnose and treat many diseases. Digital subtraction angiography is one of the most applicable imaging to diagnose brain vascular diseases and radiosurgery of brain. This paper proposes an automatic fuzzy-based multi-temporal fusion algorithm for 2-D digital subtraction angiography images. In this algorithm, for blood vessel map extraction, the valuable frames of brain angiography video are automatically determined to form the digital subtraction angiography images based on a novel definition of vessel dispersion generated by injected contrast material. Our proposed fusion scheme contains different fusion methods for high and low frequency contents based on the coefficient characteristic of wrapping second generation of curvelet transform and a novel content selection strategy. Our proposed content selection strategy is defined based on sample correlation of the curvelet transform coefficients. In our proposed fuzzy-based fusion scheme, the selection of curvelet coefficients are optimized by applying weighted averaging and maximum selection rules for the high frequency coefficients. For low frequency coefficients, the maximum selection rule based on local energy criterion is applied to better visual perception. Our proposed fusion algorithm is evaluated on a perfect brain angiography image dataset consisting of one hundred 2-D internal carotid rotational angiography videos. The obtained results demonstrate the effectiveness and efficiency of our proposed fusion algorithm in comparison with common and basic fusion algorithms.     

决策森林分析大型IBD谱定位酒精依赖症相关基因

ManycomplexhumandiseasessuchasbehaviorsofalcoholisminvestigatedbytheGeneticAnalysisWorkshop14(GAW14,http://www.gaworkshop.org/)arenotsimpleMendeliandisorders.Instead,theymayhavemixedcontributionsofgenes,environ mentsandtheirinteractions.Recentadvancesindi…     

Distantly supervised biomedical relation extraction using piecewise attentive convolutional neural network and reinforcement learning

ObjectiveThere have been various methods to deal with the erroneous training data in distantly supervised relation extraction (RE), however, their performance is still far from satisfaction. We aimed to deal with the insufficient modeling problem on instance-label correlations for predicting biomedical relations using deep learning and reinforcement learning.Materials and MethodsIn this study, a new computational model called piecewise attentive convolutional neural network and reinforcement learning (PACNN+RL) was proposed to perform RE on distantly supervised data generated from Unified Medical Language System with MEDLINE abstracts and benchmark datasets. In PACNN+RL, PACNN was introduced to encode semantic information of biomedical text, and the RL method with memory backtracking mechanism was leveraged to alleviate the erroneous data issue. Extensive experiments were conducted on 4 biomedical RE tasks.ResultsThe proposed PACNN+RL model achieved competitive performance on 8 biomedical corpora, outperforming most baseline systems. Specifically, PACNN+RL outperformed all baseline methods with the F1-score of 0.5592 on the may-prevent dataset, 0.6666 on the may-treat dataset, and 0.3838 on the DDI corpus, 2011. For the protein-protein interaction RE task, we obtained new state-of-the-art performance on 4 out of 5 benchmark datasets.ConclusionsThe performance on many distantly supervised biomedical RE tasks was substantially improved, primarily owing to the denoising effect of the proposed model. It is anticipated that PACNN+RL will become a useful tool for large-scale RE and other downstream tasks to facilitate biomedical knowledge acquisition. We also made the demonstration program and source code publicly available at http://112.74.48.115:9000/.     

基于Wasserstein Gan的无监督单模配准方法

本文提出一种基于Wasserstein Gan的无监督单模配准方法。与现有的基于深度学习的单模配准方法不同，本文的方法完成训练不需要Ground truth和预设的相似性度量指标。本文方法的主要结构包括生成网络和判别网络。首先，生成网络输入固定图像（正例图像）和浮动图像并提取图像间潜在的形变场，通过插值方式预测配准图像（负例图像）；然后，判别网络交替输入正例图像和负例图像，判断图像间的相似性，并将判断结果作为损失函数反馈，进而驱动网络参数更新；最后，通过对抗训练，生成网络预测的配准图像能欺骗判别网络，网络收敛。实验中随机选取30例LPBA40脑部数据集、25例EMPIRE10肺部数据集和15例ACDC心脏数据集用作训练数据集，然后将剩下的10例LPBA40脑部数据集、5例EMPIRE10肺部数据集和5例ACDC心脏数据集用作测试数据集。配准结果与Affine算法、Demons算法、SyN算法和VoxelMorph算法对比。实验结果显示，本研究算法的DICE系数（DSC）和归一化相关系数（NCC）评价指标均是最高，表明本文方法的配准精度高于Affine算法、Demons算法、SyN算法和目前无监督的SOTA算法VoxelMorph。     

基于卷积神经网络和超像素的CT图像肝脏分割

肝癌是威胁人类健康的重大疾病之一。从医学影像中将肝脏组织准确地分割出来,是计算机辅助肝脏疾病诊断与手术规划中一个重要环节。由于肝脏的个体差异,周围器官的灰度值相似等因素,从CT图像中精准分割肝脏存在一定困难。提出一种结合卷积神经网络和超像素的CT图像肝脏自动分割方法。首先利用卷积神经网络进行目标检测,自动定位肝脏区域,再利用超像素算法对肝脏进行分割,最后进行腐蚀、膨胀、中值滤波等后处理。本文采用3DIRCADb公开数据集对提出的肝脏自动分割算法进行评估和验证,结果表明肝脏自动分割的DICE指标为0.951,VOE指标为0.0917,RVD指标为-0.018,显示出较好的分割精度。     

基于三维全卷积DenseNet的脑胶质瘤MRI分割

从多模态MRI中对多个脑胶质瘤区域进行精确分割是不少精准医疗步骤的前提。为了有效针对脑胶质瘤MRI的特性和 提升其分割精度,本文提出了多Dice损失函数结构,并采用预实验选择良好的超参数（数据维度、图像融合步长、损失函数的实 现形式）构建一个基于三维全卷积DenseNet的图像特征学习网络。本研究包含了脑胶质瘤MRI的274个已分割训练集和110 个未提供分割的测试集。图像进行灰度归一化后提取三维图像块作为网络输入,网络输出利用图像块融合方法得到最终的分 割结果。相比通用的结构,推荐的结构提高了脑胶质瘤的分割精度。在公开的BraTS2015数据集上进行在线的评估中,整个肿 瘤区、肿瘤核心区和增强肿瘤区的Dice值分别为0.85、0.71、0.63。     

基于U型全卷积神经网络的腹部动脉CT图像分割算法研究

目的 为了精确分割腹部动脉血管,提出一种基于深度学习的全自动腹部动脉CT图像分割算法.方法 采用区域不平衡块生成方法 提取CT血管横断面、冠状面和矢状面图像特征,接着采用U型全卷积神经网络对块特征进行训练与分割,最后采用最大体素保留法获得三维血管分割图像.选用120例患者腹部CT血管图像进行网络训练和分割实验,分割结果...