• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

基于卷积神经网络（CNN）和手工特征混合模型的血液涂片图像分析，用于对白细胞类型进行分类以预测血液学情况。

Blood Slide Image Analysis to Classify WBC Types for Prediction Haematology Based on a Hybrid Model of CNN and Handcrafted Features.

作者信息

Olayah Fekry, Senan Ebrahim Mohammed, Ahmed Ibrahim Abdulrab, Awaji Bakri

机构信息

Department of Information System, Faculty Computer Science and information System, Najran University, Najran 66462, Saudi Arabia.

Department of Artificial Intelligence, Faculty of Computer Science and Information Technology, Alrazi University, Sana'a, Yemen.

出版信息

Diagnostics (Basel). 2023 May 29;13(11):1899. doi: 10.3390/diagnostics13111899.

DOI:10.3390/diagnostics13111899
PMID:37296753
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC10252914/
Abstract

White blood cells (WBCs) are one of the main components of blood produced by the bone marrow. WBCs are part of the immune system that protects the body from infectious diseases and an increase or decrease in the amount of any type that causes a particular disease. Thus, recognizing the WBC types is essential for diagnosing the patient's health and identifying the disease. Analyzing blood samples to determine the amount and WBC types requires experienced doctors. Artificial intelligence techniques were applied to analyze blood samples and classify their types to help doctors distinguish between types of infectious diseases due to increased or decreased WBC amounts. This study developed strategies for analyzing blood slide images to classify WBC types. The first strategy is to classify WBC types by the SVM-CNN technique. The second strategy for classifying WBC types is by SVM based on hybrid CNN features, which are called VGG19-ResNet101-SVM, ResNet101-MobileNet-SVM, and VGG19-ResNet101-MobileNet-SVM techniques. The third strategy for classifying WBC types by FFNN is based on a hybrid model of CNN and handcrafted features. With MobileNet and handcrafted features, FFNN achieved an AUC of 99.43%, accuracy of 99.80%, precision of 99.75%, specificity of 99.75%, and sensitivity of 99.68%.

摘要

白细胞（WBCs）是骨髓产生的血液主要成分之一。白细胞是免疫系统的一部分，可保护身体免受传染病侵害，且任何一种特定疾病相关白细胞数量的增减都会引发该疾病。因此，识别白细胞类型对于诊断患者健康状况和确定疾病至关重要。分析血样以确定白细胞数量和类型需要经验丰富的医生。人工智能技术被应用于分析血样并对其类型进行分类，以帮助医生因白细胞数量增减区分传染病类型。本研究开发了分析血涂片图像以对白细胞类型进行分类的策略。第一种策略是通过支持向量机-卷积神经网络（SVM-CNN）技术对白细胞类型进行分类。第二种对白细胞类型进行分类的策略是基于混合卷积神经网络特征的支持向量机，即VGG19-ResNet101-SVM、ResNet101-MobileNet-SVM和VGG19-ResNet101-MobileNet-SVM技术。第三种通过前馈神经网络（FFNN）对白细胞类型进行分类的策略基于卷积神经网络和手工特征的混合模型。借助MobileNet和手工特征，前馈神经网络实现了99.43%的曲线下面积（AUC）、99.80%的准确率、99.75%的精确率、99.75%的特异性和99.68%的灵敏度。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/3ff5bd84bf22/diagnostics-13-01899-g017.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/ba4ab9423512/diagnostics-13-01899-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/d03ac8b78abc/diagnostics-13-01899-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/3aa64fec7b1e/diagnostics-13-01899-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/51509430e4d6/diagnostics-13-01899-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/cb2231cee07e/diagnostics-13-01899-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/527ae19502ab/diagnostics-13-01899-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/e53a150ceace/diagnostics-13-01899-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/d08c99cbba22/diagnostics-13-01899-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/a156158f5231/diagnostics-13-01899-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/8853d3ddeaa5/diagnostics-13-01899-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/4269511948ce/diagnostics-13-01899-g011.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/b03d764d19ee/diagnostics-13-01899-g012.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/b376dfcd7a3f/diagnostics-13-01899-g013.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/5205a98711c2/diagnostics-13-01899-g014.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/d482e96ec6e7/diagnostics-13-01899-g015.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/a48014d073a6/diagnostics-13-01899-g016.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/3ff5bd84bf22/diagnostics-13-01899-g017.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/ba4ab9423512/diagnostics-13-01899-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/d03ac8b78abc/diagnostics-13-01899-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/3aa64fec7b1e/diagnostics-13-01899-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/51509430e4d6/diagnostics-13-01899-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/cb2231cee07e/diagnostics-13-01899-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/527ae19502ab/diagnostics-13-01899-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/e53a150ceace/diagnostics-13-01899-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/d08c99cbba22/diagnostics-13-01899-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/a156158f5231/diagnostics-13-01899-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/8853d3ddeaa5/diagnostics-13-01899-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/4269511948ce/diagnostics-13-01899-g011.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/b03d764d19ee/diagnostics-13-01899-g012.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/b376dfcd7a3f/diagnostics-13-01899-g013.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/5205a98711c2/diagnostics-13-01899-g014.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/d482e96ec6e7/diagnostics-13-01899-g015.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/a48014d073a6/diagnostics-13-01899-g016.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/dc6f/10252914/3ff5bd84bf22/diagnostics-13-01899-g017.jpg

相似文献

1
Blood Slide Image Analysis to Classify WBC Types for Prediction Haematology Based on a Hybrid Model of CNN and Handcrafted Features.基于卷积神经网络（CNN）和手工特征混合模型的血液涂片图像分析，用于对白细胞类型进行分类以预测血液学情况。
Diagnostics (Basel). 2023 May 29;13(11):1899. doi: 10.3390/diagnostics13111899.
2
Hybrid Models Based on Fusion Features of a CNN and Handcrafted Features for Accurate Histopathological Image Analysis for Diagnosing Malignant Lymphomas.基于卷积神经网络融合特征与手工特征的混合模型用于准确的组织病理学图像分析以诊断恶性淋巴瘤
Diagnostics (Basel). 2023 Jul 4;13(13):2258. doi: 10.3390/diagnostics13132258.
3
Hybrid Techniques for the Diagnosis of Acute Lymphoblastic Leukemia Based on Fusion of CNN Features.基于卷积神经网络特征融合的急性淋巴细胞白血病诊断混合技术
Diagnostics (Basel). 2023 Mar 8;13(6):1026. doi: 10.3390/diagnostics13061026.
4
Hybrid Techniques of X-ray Analysis to Predict Knee Osteoarthritis Grades Based on Fusion Features of CNN and Handcrafted.基于卷积神经网络（CNN）与手工特征融合的X射线分析混合技术预测膝关节骨关节炎等级
Diagnostics (Basel). 2023 May 2;13(9):1609. doi: 10.3390/diagnostics13091609.
5
Analyzing Histological Images Using Hybrid Techniques for Early Detection of Multi-Class Breast Cancer Based on Fusion Features of CNN and Handcrafted.基于卷积神经网络（CNN）和手工特征融合的混合技术分析组织学图像以早期检测多类乳腺癌
Diagnostics (Basel). 2023 May 17;13(10):1753. doi: 10.3390/diagnostics13101753.
6
Automatic Analysis of MRI Images for Early Prediction of Alzheimer's Disease Stages Based on Hybrid Features of CNN and Handcrafted Features.基于卷积神经网络（CNN）混合特征和手工特征的MRI图像自动分析用于阿尔茨海默病阶段的早期预测
Diagnostics (Basel). 2023 May 8;13(9):1654. doi: 10.3390/diagnostics13091654.
7
Hybrid Techniques of Facial Feature Image Analysis for Early Detection of Autism Spectrum Disorder Based on Combined CNN Features.基于组合卷积神经网络特征的面部特征图像分析混合技术用于自闭症谱系障碍的早期检测
Diagnostics (Basel). 2023 Sep 14;13(18):2948. doi: 10.3390/diagnostics13182948.
8
Automatic Classification of Colour Fundus Images for Prediction Eye Disease Types Based on Hybrid Features.基于混合特征的眼底彩色图像自动分类以预测眼病类型
Diagnostics (Basel). 2023 May 11;13(10):1706. doi: 10.3390/diagnostics13101706.
9
Hybrid Methods for Fundus Image Analysis for Diagnosis of Diabetic Retinopathy Development Stages Based on Fusion Features.基于融合特征的糖尿病视网膜病变发展阶段诊断的眼底图像分析混合方法
Diagnostics (Basel). 2023 Aug 28;13(17):2783. doi: 10.3390/diagnostics13172783.
10
Hybrid Models for Endoscopy Image Analysis for Early Detection of Gastrointestinal Diseases Based on Fused Features.基于融合特征的用于早期检测胃肠道疾病的内镜图像分析混合模型
Diagnostics (Basel). 2023 May 16;13(10):1758. doi: 10.3390/diagnostics13101758.

引用本文的文献

1
Toward Digital Periodontal Health: Recent Advances and Future Perspectives.迈向数字化牙周健康：最新进展与未来展望。
Bioengineering (Basel). 2024 Sep 18;11(9):937. doi: 10.3390/bioengineering11090937.
2
Efficient Malaria Parasite Detection From Diverse Images of Thick Blood Smears for Cross-Regional Model Accuracy.从厚血涂片的多样图像中高效检测疟原虫以实现跨区域模型准确性。
IEEE Open J Eng Med Biol. 2023 Oct 30;4:226-233. doi: 10.1109/OJEMB.2023.3328435. eCollection 2023.

本文引用的文献

1
Multi-Techniques for Analyzing X-ray Images for Early Detection and Differentiation of Pneumonia and Tuberculosis Based on Hybrid Features.基于混合特征的用于早期检测及区分肺炎和肺结核的X射线图像分析多技术
Diagnostics (Basel). 2023 Feb 20;13(4):814. doi: 10.3390/diagnostics13040814.
2
ETECADx: Ensemble Self-Attention Transformer Encoder for Breast Cancer Diagnosis Using Full-Field Digital X-ray Breast Images.ETECADx：用于使用全场数字化乳腺X线摄影图像进行乳腺癌诊断的集成自注意力Transformer编码器
Diagnostics (Basel). 2022 Dec 28;13(1):89. doi: 10.3390/diagnostics13010089.
3
BCNet: A Deep Learning Computer-Aided Diagnosis Framework for Human Peripheral Blood Cell Identification.
BCNet：一种用于人类外周血细胞识别的深度学习计算机辅助诊断框架。
Diagnostics (Basel). 2022 Nov 16;12(11):2815. doi: 10.3390/diagnostics12112815.
4
A Hybrid Workflow of Residual Convolutional Transformer Encoder for Breast Cancer Classification Using Digital X-ray Mammograms.一种基于数字乳腺X线摄影的用于乳腺癌分类的残差卷积变压器编码器混合工作流程。
Biomedicines. 2022 Nov 18;10(11):2971. doi: 10.3390/biomedicines10112971.
5
Multi-Modal Brain Tumor Detection Using Deep Neural Network and Multiclass SVM.基于深度神经网络和多类支持向量机的多模态脑肿瘤检测
Medicina (Kaunas). 2022 Aug 12;58(8):1090. doi: 10.3390/medicina58081090.
6
Early Diagnosis of Brain Tumour MRI Images Using Hybrid Techniques between Deep and Machine Learning.基于深度学习和机器学习的混合技术对脑瘤 MRI 图像的早期诊断
Comput Math Methods Med. 2022 May 18;2022:8330833. doi: 10.1155/2022/8330833. eCollection 2022.
7
Recombinant HBsAg of the Wild-Type and the G145R Escape Mutant, included in the New Multivalent Vaccine against Hepatitis B Virus, Dramatically Differ in their Effects on Leukocytes from Healthy Donors In Vitro.新型乙肝病毒多价疫苗中包含的野生型和G145R逃逸突变体的重组乙肝表面抗原（HBsAg），在体外对健康供体白细胞的影响上存在显著差异。
Vaccines (Basel). 2022 Feb 3;10(2):235. doi: 10.3390/vaccines10020235.
8
Multi-Method Diagnosis of Blood Microscopic Sample for Early Detection of Acute Lymphoblastic Leukemia Based on Deep Learning and Hybrid Techniques.基于深度学习和混合技术的血液显微镜样本的多方法诊断，用于急性淋巴细胞白血病的早期检测。
Sensors (Basel). 2022 Feb 19;22(4):1629. doi: 10.3390/s22041629.
9
Cross-Sectional and Time-Dependent Analyses on Inflammatory Markers following Natural Killer Cell Activity.自然杀伤细胞活性后炎症标志物的横断面分析和时间依赖性分析
Diagnostics (Basel). 2022 Feb 9;12(2):448. doi: 10.3390/diagnostics12020448.
10
Score and Correlation Coefficient-Based Feature Selection for Predicting Heart Failure Diagnosis by Using Machine Learning Algorithms.基于评分和相关系数的特征选择在使用机器学习算法预测心力衰竭诊断中的应用。
Comput Math Methods Med. 2021 Dec 20;2021:8500314. doi: 10.1155/2021/8500314. eCollection 2021.