检索结果-内蒙古大学图书馆

Deep Learning Methods for Screening Pulmonary Tuberculosis Using Chest X-rays

COMPUTER METHODS IN BIOMECHANICS AND BIOMEDICAL ENGINEERING-IMAGING AND VISUALIZATION 2021年第1期9卷 39-49页

作者： Dasanayaka, Chirath Dissanayake, Maheshi Buddhinee Univ Peradeniya Fac Engn Dept Elect & Elect Engn Kandy Sri Lanka

Tuberculosis (TB) is a contagious bacterial airborne disease, and is one of the top 10 causes of death worldwide. According to the World Health Organisation, around 1.8 billion people are infected with TB and 1.6 million deaths were reported in 2018. More importantly, 95% of cases and deaths were from developing countries. Yet, TB is a completely curable disease through early diagnosis. To achieve this goal one of the key requirements is efficient utilisation of existing diagnostic technologies, among which chest X-ray is the first line of diagnostic tool used for screening for active TB. The presented deep learning pipeline consists of three different modern deep learning architectures, to generate, segment, and classify lung X-rays. Apart from this, image preprocessing, image augmentation, genetic algorithm based hyper parameter tuning, and model ensembling were used to improve the diagnostic process. We were able to achieve classification accuracy of 97.1% (Youden's index-0.941, sensitivity of 97.9%, specificity of 96.2%) which is a considerable improvement compared to the existing work in the literature. In our work, we present a highly accurate, automated TB screening system using chest X-rays, which would be helpful especially for low income countries with low access to qualified medical professionals.

关键词： Machine Learning automated Tuberculosis Screening Chest/Lung X-ray analysis Deep Convolutional Generative Adversarial Networks segmentation and classification

来源：评论

学校读者我要写书评

暂无评论

Research on Intelligent Identification and Grading of Nonmetallic Inclusions in Steels Based on Deep Learning

引用

MICROMACHINES 2023年第2期14卷 482-482页

作者： Zhu, Xiaolin Wan, Wenhai Qian, Ling Cai, Yu Chen, Xiang Zhang, Pingze Huang, Guanxi Liu, Bo Yao, Qiang Li, Shaoyuan Yao, Zhengjun Nanjing Univ Aeronaut & Astronaut Coll Mat Sci & Technol Nanjing 211106 Peoples R China Jiangsu Prod Qual Testing & Inspection Inst Nanjing 210007 Peoples R China Jiangsu Changbao Precis Steel Tube Co Ltd Changzhou 213200 Peoples R China Jiangsu Zhongxin Pipe Scitec Co Ltd Nanjing 211100 Peoples R China

Non-metallic inclusions are unavoidable defects in steel, and their type, quantity, size, and distribution have a great impact on the quality of steel. At present, non-metallic inclusions are mainly detected manually, which features high work intensity, low efficiency, proneness to misjudgment, and low consistency of results. In this paper, based on deep neural network algorithm, a small number of manually labeled, low-resolution metallographic images collected by optical microscopes are used as the dataset for intelligent boundary extraction, classification, and rating of non-metallic inclusions. The training datasets are cropped into those containing only a single non-metallic inclusion to reduce the interference of background information and improve the accuracy. To deal with the unbalanced distribution of each category of inclusions, the reweighting cross entropy loss and focal loss are respectively used as the category prediction loss and boundary prediction loss of the DeepLabv3+ semantic segmentation model. Finally, the length and width of the minimum enclosing rectangle of the segmented inclusions are measured to calculate the grade of inclusions. The resulting accuracy is 90.34% in segmentation and 90.35% in classification. As is verified, the model-based rating results are consistent with those of manual labeling. For a single sample, the detection time is reduced from 30 min to 15 s, significantly improving the detection efficiency.

关键词： non-metallic inclusions segmentation and classification intelligent rating deep learning

来源：评论

学校读者我要写书评

暂无评论

Automatic Brain Tumor Detection and classification Based on IoT and Machine Learning Techniques

引用

FLUCTUATION AND NOISE LETTERS 2022年第3期21卷 2250030-2250030页

作者： Sundarasekar, Revathi Appathurai, Ahilan Anna Univ Informat & Commun Engn Chennai Tamil Nadu India Anna Univ Infant Jesus Coll Engn Dept ECE Chennai Tamil Nadu India

Brain tumor detection, segmentation, and classification are essential in clinical diagnosis and efficient treatment. Researchers have recently shown a greater interest in attaining accurate brain tumor categorization using the Internet of Things (IoT) and machine learning. The rigidity of tumor classification and segmentation in magnetic resonance imaging is due to large data and indistinct boundaries. Hence, in this study, Machine Learning assisted Automatic Brain Tumor Detection Framework (MLABTDF) has been proposed using IoT. Our study includes establishing a deep convolutional neural network (DCNN) for spotting brain tumors from magnetic resonance imageries. This article accommodated technologies of the IoT for helping brain treatment specialists in identifying the need to make surgeries contingent on MR images. The standard medical image dataset has been gathered and experimentally examined to validate the accuracy, efficiency, specificity, sensitivity, optimum automatic recognition for non-tumor and tumor regions, and the model's error rate utilizing statistical construction. This study pays its ability in brain irregularity recognition and analysis in the healthcare sector without humanoid intermediation. Compared to other systems, the experimental results show that the recommended MLABTDF model improves efficiency by 95.7%, segmentation and classification accuracy by 99.9%, specificity by 97.3%, sensitivity by 96.4%, optimal automatic detection by 93.4%, Matthews correlation coefficient ratio by 97.1% and error rate by 10.2%.

关键词： Brain tumor detection segmentation and classification MRI machine learning IoT

来源：评论

学校读者我要写书评

暂无评论

An Accurate Neural Network for Cytologic Whole-Slide Image Analysis 20

An Accurate Neural Network for Cytologic Whole-Slide Image A...

引用

Australasian Computer Science Week Multiconference (ACSW)

作者： Deng, Junwei Lu, Yizhou Ke, Jing Shanghai Jiao Tong Univ Univ Michigan Shanghai Jiao Tong Univ Joint Inst Shanghai Peoples R China Shanghai Jiao Tong Univ Dept Comp Sci & Engn Shanghai Peoples R China

ISBN: (纸本)9781450376976

Typically, high accuracy in deep learning is achieved by large dataset in pixel-wise labeling for segmentation or image-level labeling for classification. However, in biomedical domain, the challenge is not only the availability of image data itself, but also the acquisition of relevant annotations for these images from clinicians. In this work, we propose a novel two-stage architecture to jointly perform the tasks of detection, segmentation and classification of abnormal cells and cancer. Compared with one-step detection for all the catalogues, we combine the advantage of image-level and pixel-level labeling in our deep learning based framework. We use the detection of lesions in cervical clinical dataset as a case study for performance evaluation. In the first stage, a hybrid ResNet and U-Net architecture is designed to predict three catalogues of nuclei, cytoplasm and background with pixel-wise labeled segmentation map. In the second stage, a residual learning based model is applied to the identified nuclei for subtype classification. Confirmed with cytotechnologist, the proposed model is estimated to efficiently deduct more than 90% annotation burden compared with pixel-wise labeling approach. Moreover, the proposed two-stage approach model outperforms one-stage neural network in segmentation and classification for objects with high similarities in appearance. Our collected real-life clinical cytology images and the source code in the experiments are provided in https ://***/SJTU-AI-GPU/TwoStageCellsegmentation.(1)

关键词： cervical cancer detection segmentation and classification two-stage deep learning pixel-wise and image-level labeling

来源：评论

学校读者我要写书评

暂无评论

An automatic machine learning approach for ischemic stroke onset time identification based on DWI and FLAIR imaging

引用

NEUROIMAGE-CLINICAL 2021年 31卷 102744-102744页

作者： Zhu, Haichen Jiang, Liang Zhang, Hong Luo, Limin Chen, Yang Chen, Yuchen Southeast Univ Sch Comp Sci & Engn Lab Image Sci & Technol Key Lab Comp Network & Informat IntegratMinist E Nanjing 210096 Peoples R China Nanjing Med Univ Nanjing Hosp 1 Dept Radiol Nanjing 210006 Peoples R China Nanjing Med Univ Affiliated Jiangning Hosp Dept Radiol Nanjing 210000 Peoples R China Southeast Univ Sch Cyber Sci & Engn Nanjing 210096 Peoples R China

Current thrombolysis for acute ischemic stroke (AIS) treatment strictly relies on the time since stroke (TSS) less than 4.5 h. However, some patients are excluded from thrombolytic treatment because of the unknown TSS. The diffusion-weighted imaging (DWI) and fluid-attenuated inversion recovery (FLAIR) mismatch can simply identify TSS since lesion intensities are not identical at different onset time. In this paper, we propose an automatic machine learning method to classify the TSS less than or more than 4.5 h. First, we develop a cross-modal convolutional neural network to accurately segment the stroke lesions from DWI and FLAIR images. Second, the features are extracted from DWI and FLAIR according to the segmentation regions of interest (ROI). Finally, the features are fed to machine learning models to identify TSS. In DWI and FLAIR ROI segmentation, the networks obtain high Dice coefficients with 0.803 and 0.647. The classification test results show that our model achieves an accuracy of 0.805, with a sensitivity of 0.769 and a specificity of 0.840. Our approach outperforms human reading DWI-FLAIR mismatch model, illustrating the potential for automatic and fast TSS identification.

关键词： Machine learning segmentation and classification Time since stroke Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

DCG-Net: Dynamic Capsule Graph Convolutional Network for Point Clouds

引用

IEEE ACCESS 2020年 8卷 188056-188067页

作者： Bazazian, Dena Nahata, Dhananjay Ctr Tecnol Telecomunicac Catalunya CTTC Geomat Dept Barcelona 08860 Spain Univ Autonoma Barcelona UAB Comp Sci Dept Barcelona 08193 Spain

This article introduces DCG-Net (Dynamic Capsule Graph Network) to analyze point clouds for the tasks of classification and segmentation. DCG-Net aggregates point cloud features to build and update the graphs based on the dynamic routing mechanism of capsule networks at each layer of a convolutional network. The first layer of DGC-Net exploits the geometrical attributes of the point cloud to build a graph by neighborhood aggregation while the deeper layers of the network dynamically update the graph based on the feature space of convolutions. We conduct extensive experiments on public datasets, ModelNet40, ShapeNet-Part. Our experimental results demonstrate that DCG-Net achieves state-of-the-art performance on public datasets, 93.4% accuracy on ModelNet40, and 85.4% instance mIoU (mean Intersection over Union) on ShapeNet-Part.

关键词： Three-dimensional displays Routing Task analysis Eigenvalues and eigenfunctions Deep learning Solid modeling Sensors Convolutional networks dynamic capsule graph point clouds segmentation and classification

来源：评论

学校读者我要写书评

暂无评论

classification of acute lymphoblastic leukemia using deep learning

引用

MICROSCOPY RESEARCH AND TECHNIQUE 2018年第11期81卷 1310-1317页

作者： Rehman, Amjad Abbas, Naveed Saba, Tanzila Rahman, Syed Ijaz ur Mehmood, Zahid Kolivand, Hoshang Al Yamamah Univ Coll Comp & Informat Syst Riyadh Saudi Arabia Islamia Coll Univ Peshawar Dept Comp Sci Peshawar Pakistan Prince Sultan Univ Coll Comp & Informat Sci Riyadh Saudi Arabia Univ Engn & Technol Dept Software Engn Taxila Pakistan Liverpool John Moores Univ Dept Comp Sci Liverpool Merseyside England

Acute Leukemia is a life-threatening disease common both in children and adults that can lead to death if left untreated. Acute Lymphoblastic Leukemia (ALL) spreads out in children's bodies rapidly and takes the life within a few weeks. To diagnose ALL, the hematologists perform blood and bone marrow examination. Manual blood testing techniques that have been used since long time are often slow and come out with the less accurate diagnosis. This work improves the diagnosis of ALL with a computer-aided system, which yields accurate result by using image processing and deep learning techniques. This research proposed a method for the classification of ALL into its subtypes and reactive bone marrow (normal) in stained bone marrow images. A robust segmentation and deep learning techniques with the convolutional neural network are used to train the model on the bone marrow images to achieve accurate classification results. Experimental results thus obtained and compared with the results of other classifiers Naive Bayesian, KNN, and SVM. Experimental results reveal that the proposed method achieved 97.78% accuracy. The obtained results exhibit that the proposed approach could be used as a tool to diagnose Acute Lymphoblastic Leukemia and its sub-types that will definitely assist pathologists.

关键词： acute lymphoblastic leukemia bone marrow deep learning segmentation and classification

来源：评论

学校读者我要写书评

暂无评论

State-of-the-Art: DTM Generation Using Airborne LIDAR Data

引用

SENSORS 2017年第1期17卷 150-150页

作者： Chen, Ziyue Gao, Bingbo Devereux, Bernard Beijing Normal Univ Coll Global Change & Earth Syst Sci 19 Xinjiekouwai St Beijing 100875 Peoples R China Beijing Acad Agr & Forestry Sci Beijing Res Ctr Informat Technol Agr Beijing 100097 Peoples R China Univ Cambridge UK Dept Geog Cambridge CB2 3EN England

Digital terrain model (DTM) generation is the fundamental application of airborne Lidar data. In past decades, a large body of studies has been conducted to present and experiment a variety of DTM generation methods. Although great progress has been made, DTM generation, especially DTM generation in specific terrain situations, remains challenging. This research introduces the general principles of DTM generation and reviews diverse mainstream DTM generation methods. In accordance with the filtering strategy, these methods are classified into six categories: surface-based adjustment;morphology-based filtering, triangulated irregular network (TIN)-based refinement, segmentation and classification, statistical analysis and multi-scale comparison. Typical methods for each category are briefly introduced and the merits and limitations of each category are discussed accordingly. Despite different categories of filtering strategies, these DTM generation methods present similar difficulties when implemented in sharply changing terrain, areas with dense non-ground features and complicated landscapes. This paper suggests that the fusion of multi-sources and integration of different methods can be effective ways for improving the performance of DTM generation.

关键词： DTM generation surface-based morphology-based TIN-based segmentation and classification statistical analysis multi-scale comparison

来源：评论

学校读者我要写书评

暂无评论

Proposed Retinal Abnormality Detection and classification Approach Computer Aided Detection for Diabetic Retinopathy by Machine Learning Approaches 8

Proposed Retinal Abnormality Detection and Classification Ap...

引用

8th IEEE International Conference on Communication Software and Networks (ICCSN)

作者： Raman, Valliappan Then, Patrick Sumari, Putra Swinbume Univ Technol Sarawak Fac Engn Comp & Sci Kuching Malaysia Univ Sains Malaysia Sch Comp Sci George Town Malaysia

ISBN: (纸本)9781509017812

Diabetes occurs when the pancreas fails to secrete enough insulin, slowly affecting the retina of the human eye, leading to diabetic retinopathy. The blood vessels in the retina get altered and have abnormality. Exudates are secreted, micro-aneurysms and haemorrhages occur in the retina. The appearance of these features represents the degree of severity of the disease. Early detection of diabetic retinopathy plays a major role in the success of such disease treatment. The main challenge is to extract exudates which are similar in colour property and size of the optic disk, and then micro-aneurysms are similar in colour and proximity with blood vessels. The main objective of the paper is to develop a computer aided detection system to find the abnormality of retinal imaging and detects the presence of abnormality features from retinal fundus images. There is few existing research works have been undergone by applying machine learning techniques, but existing approaches have not achieved a good accuracy of detection and they have not yielded successful performance in different datasets. The proposed methodology is to enhance the image and filter the noise, detect blood vessel and identify the optic disc, extract the exudates and micro aneurysms, extract the features and classify different stages of diabetic retinopathy into mild, moderate, severe non-proliferative diabetic retinopathy (NPDR) and proliferative Diabetic retinopathy (PDR) by using proposed machine learning methods. The expected output of proposed work in this paper will be a preliminary design and pilot prototype development.

关键词： diabetes exudates microaneurysm segmentation and classification

来源：评论

学校读者我要写书评

暂无评论

Proposed Rentinal Abnormatlty Detection and classification Approach Computer aided detection for diabetic retinopathy by machine learning approaches

Proposed Rentinal Abnormatlty Detection and Classification A...

引用

2016 8th IEEE International Conference on Communication Software and Networks(ICCSN 2016)

作者： Valliappan Raman Patrick Then Putra Sumari Faculty of Engineering Computing and Science Swinburne University of Technology Sarawak School of Computer Science Universiti Sains Malaysia

Diabetes occurs when the pancreas fails to secrete enough insulin,slowly affecting the retina of the human eye,leading to diabetic *** blood vessels in the retina get altered and have *** are secreted,micro-aneurysms and haemorrhages occur in the *** appearance of these features represents the degree of severity of the *** detection of diabetic retinopathy plays a major role in the success of such disease *** main challenge is to extract exudates which are similar in colour property and size of the optic disk,and then micro-aneurysms are similar in colour and proximity with blood *** main objective of the paper is to develop a computer aided detection system to find the abnormality of retinal imaging and detects the presence of abnormality features from retinal fundus *** is few existing research works have been undergone by applying machine learning techniques,but existing approaches have not achieved a good accuracy of detection and they have not yield successful performance in different *** proposed methodology is to enhance the image and filter the noise,detect blood vessel and identify the optic disc,extract the exudates and micro aneurysms extract the features and classify different stages of diabetic retinopathy into mild,moderate,severe non-proliferative diabetic retinopathy(NPDR)and proliferative Diabetic retinopathy(PDR)by using proposed machine learning *** expected output of proposed work in this paper will be a preliminary design and pilot prototype development.

关键词： diabetes exudates microaneurysm segmentation and classification

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：