基于混合堆叠集成模型和特征选择的宫颈癌检测新方法

Pratiksha D. Nandanwar, Dr. Somnath B. Dhonde
{"title":"基于混合堆叠集成模型和特征选择的宫颈癌检测新方法","authors":"Pratiksha D. Nandanwar, Dr. Somnath B. Dhonde","doi":"10.37391/ijeer.110246","DOIUrl":null,"url":null,"abstract":"Around the world, millions of women are diagnosed with cervical cancer each year. Early detection is very important to produce a better overall quality of life for those diagnosed with the disease and reduce the burden on the healthcare system. In recent years, the field of machine learning (ML) has been developing methods that can improve the accuracy of detecting cervical cancer. This paper presents a new approach to this problem by using a combination of image segmentation and feature extraction techniques. The proposed approach is divided into three phases. The first stage involves image segmentation, which is performed to extract the regions of interest from the input image. The second stage is comprised of extracting the features from the ROI with the help of the Histogram and Hu Moments techniques. The techniques used in this approach, namely the Hu Moments and Histogram techniques, respectively, can capture the shape information in the ROI. In the third stage of the project, we use a hybrid approach to classify the image. The proposed model is composed of several base classifiers, which are trained on varying subsets of the features that were extracted. These resulting classifiers then make a classification decision. We tested the proposed model against a large dataset of images for cervical cancer. The results of the experiments revealed that it performed better than the existing methods in detecting the disease. It was able to achieve an accuracy of 96.5%, an F1 score of 96.9%, and a recall of 96.7%. The proposed model was successful in accomplishing a remarkable accuracy of 96.5%, making it an ideal candidate for use in the detection of cervical cancer. It was also able to perform feature extraction using the Histogram techniques and image segmentation. The proposed method could help medical professionals improve the diagnosis and reduce the burden of this disease on women worldwide.","PeriodicalId":158560,"journal":{"name":"International Journal of Electrical and Electronics Research","volume":"14 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A Novel Approach to Cervical Cancer Detection Using Hybrid Stacked Ensemble Models and Feature Selection\",\"authors\":\"Pratiksha D. Nandanwar, Dr. Somnath B. Dhonde\",\"doi\":\"10.37391/ijeer.110246\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Around the world, millions of women are diagnosed with cervical cancer each year. Early detection is very important to produce a better overall quality of life for those diagnosed with the disease and reduce the burden on the healthcare system. In recent years, the field of machine learning (ML) has been developing methods that can improve the accuracy of detecting cervical cancer. This paper presents a new approach to this problem by using a combination of image segmentation and feature extraction techniques. The proposed approach is divided into three phases. The first stage involves image segmentation, which is performed to extract the regions of interest from the input image. The second stage is comprised of extracting the features from the ROI with the help of the Histogram and Hu Moments techniques. The techniques used in this approach, namely the Hu Moments and Histogram techniques, respectively, can capture the shape information in the ROI. In the third stage of the project, we use a hybrid approach to classify the image. The proposed model is composed of several base classifiers, which are trained on varying subsets of the features that were extracted. These resulting classifiers then make a classification decision. We tested the proposed model against a large dataset of images for cervical cancer. The results of the experiments revealed that it performed better than the existing methods in detecting the disease. It was able to achieve an accuracy of 96.5%, an F1 score of 96.9%, and a recall of 96.7%. The proposed model was successful in accomplishing a remarkable accuracy of 96.5%, making it an ideal candidate for use in the detection of cervical cancer. It was also able to perform feature extraction using the Histogram techniques and image segmentation. The proposed method could help medical professionals improve the diagnosis and reduce the burden of this disease on women worldwide.\",\"PeriodicalId\":158560,\"journal\":{\"name\":\"International Journal of Electrical and Electronics Research\",\"volume\":\"14 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-06-30\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International Journal of Electrical and Electronics Research\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.37391/ijeer.110246\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Electrical and Electronics Research","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.37391/ijeer.110246","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

在世界各地,每年有数百万妇女被诊断患有宫颈癌。早期发现对于提高被诊断患有这种疾病的人的整体生活质量和减轻医疗保健系统的负担非常重要。近年来,机器学习(ML)领域一直在开发可以提高宫颈癌检测准确性的方法。本文提出了一种结合图像分割和特征提取技术的新方法。建议的方法分为三个阶段。第一阶段涉及图像分割,即从输入图像中提取感兴趣的区域。第二阶段是利用直方图和胡矩技术从ROI中提取特征。该方法中使用的技术分别是Hu Moments和Histogram技术,可以捕获ROI中的形状信息。在项目的第三阶段,我们使用混合方法对图像进行分类。该模型由几个基本分类器组成,这些分类器在提取的特征的不同子集上进行训练。这些结果分类器然后做出分类决策。我们针对子宫颈癌图像的大型数据集测试了提出的模型。实验结果表明,它比现有的方法在检测疾病方面表现得更好。它能够达到96.5%的准确率,96.9%的F1分数和96.7%的召回率。所提出的模型成功地实现了96.5%的显着准确性,使其成为宫颈癌检测的理想候选者。利用直方图技术进行特征提取和图像分割。所提议的方法可以帮助医疗专业人员改进诊断并减轻这种疾病对全世界妇女的负担。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
A Novel Approach to Cervical Cancer Detection Using Hybrid Stacked Ensemble Models and Feature Selection
Around the world, millions of women are diagnosed with cervical cancer each year. Early detection is very important to produce a better overall quality of life for those diagnosed with the disease and reduce the burden on the healthcare system. In recent years, the field of machine learning (ML) has been developing methods that can improve the accuracy of detecting cervical cancer. This paper presents a new approach to this problem by using a combination of image segmentation and feature extraction techniques. The proposed approach is divided into three phases. The first stage involves image segmentation, which is performed to extract the regions of interest from the input image. The second stage is comprised of extracting the features from the ROI with the help of the Histogram and Hu Moments techniques. The techniques used in this approach, namely the Hu Moments and Histogram techniques, respectively, can capture the shape information in the ROI. In the third stage of the project, we use a hybrid approach to classify the image. The proposed model is composed of several base classifiers, which are trained on varying subsets of the features that were extracted. These resulting classifiers then make a classification decision. We tested the proposed model against a large dataset of images for cervical cancer. The results of the experiments revealed that it performed better than the existing methods in detecting the disease. It was able to achieve an accuracy of 96.5%, an F1 score of 96.9%, and a recall of 96.7%. The proposed model was successful in accomplishing a remarkable accuracy of 96.5%, making it an ideal candidate for use in the detection of cervical cancer. It was also able to perform feature extraction using the Histogram techniques and image segmentation. The proposed method could help medical professionals improve the diagnosis and reduce the burden of this disease on women worldwide.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
1.70
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信