Fus2Net: A Novel Convolutional Neural Network for Classification of Benign and Malignant Breast Tumor in Ultrasound Images
preprint
OA: closed
AI-generated summary
This study proposes Fus2Net, a novel convolutional neural network for classifying benign and malignant breast ultrasound tumor images, achieving 92% accuracy, 95.65% sensitivity, 88.89% specificity, and a 0.97 AUC.
One-sentence paraphrase of the abstract; not a substitute for reading it. No clinical advice. How this works
AI-generated deep summary
The paper studied deep-learning classification of benign versus malignant breast tumors using breast ultrasound images, developing and testing a novel convolutional neural network architecture called Fus2Net. Using 1052 hospital-collected ultrasound images (696 benign, 356 malignant) with data augmentation to address class imbalance and 10-fold cross-validation for generalization assessment, the model was evaluated on accuracy, sensitivity, specificity, and AUC without additional training. Fus2Net, which includes two self-designed feature extraction modules and uses hyperparameter fine-tuning and regularization for convergence, achieved 92% accuracy, 95.65% sensitivity, 88.89% specificity, and an AUC of 0.97, and the authors report improved comprehensive performance versus existing CNN architectures. The paper is a preprint (not peer reviewed), and it is limited to a single relatively small dataset collected from one local hospital. The paper does not explicitly discuss endometriosis or adenomyosis; it was included in the corpus via a keyword match in the upstream search index.
Read from the paper's body, not the abstract. Not a substitute for reading the paper. No clinical advice. How this works
Abstract
Abstract Background: The rapid development of artificial intelligence technology has improved the capability of automatic breast cancer diagnosis, compared to traditional machine learning methods. Convolutional Neural Network (CNN) can automatically select high-efficiency features, which helps to improve the level of computer-aided diagnosis (CAD). It can improve the performance of distinguishing benign and malignant breast ultrasound (BUS) tumor images and makes rapid breast tumor screening possible. Results: The classification model was evaluated by using BUS tumor images without training. Evaluation indicators include accuracy, sensitivity, specificity, and Area Under Curve (AUC) value. The results in the Fus2Net model had an accuracy of 92%, the sensitivity reached 95.65%, the specificity reached 88.89%, and the AUC value reached 0.97 for classifying BUS tumor images. Conclusions: The experiment compared the existing CNN categorized architecture, and the Fus2Net architecture we customed has more advantages in a comprehensive performance. The obtained results demonstrated that the Fus2Net classification method we proposed can better assist radiologists in the diagnosis of benign and malignant BUS tumor images. Methods: The existing public datasets are small and the amount of data suffer from the balance issue. In this paper, we provide a relatively larger dataset with a total of 1052 ultrasound images, including 696 benign images and 356 malignant images, which were collected from a local hospital. We proposed a novel CNN named Fus2Net for the benign and malignant classification of BUS tumor images and it contains two self-designed feature extraction modules. To evaluate how the classifier generalizes on the experimental dataset, 10-fold cross validation was employed. Meanwhile, to solve the balance of the dataset, the training data was augmented before being fed into the Fus2Net. In the experiment, we used hyperparameter fine-tuning and regularization technology to make the Fus2Net convergence.
Full text
17,406 characters
· extracted from
preprint-html
· click to expand
Fus2Net: A Novel Convolutional Neural Network for Classification of Benign and Malignant Breast Tumor in Ultrasound Images | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Fus2Net: A Novel Convolutional Neural Network for Classification of Benign and Malignant Breast Tumor in Ultrasound Images He Ma, Ronghui Tian, Hong Li, Hang Sun, Guoxiu Lu, Ruibo Liu, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-853246/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 18 Nov, 2021 Read the published version in BioMedical Engineering OnLine → Version 1 posted 15 You are reading this latest preprint version Abstract Background : The rapid development of artificial intelligence technology has improved the capability of automatic breast cancer diagnosis, compared to traditional machine learning methods. Convolutional Neural Network (CNN) can automatically select high-efficiency features, which helps to improve the level of computer-aided diagnosis (CAD). It can improve the performance of distinguishing benign and malignant breast ultrasound (BUS) tumor images and makes rapid breast tumor screening possible. Results : The classification model was evaluated by using BUS tumor images without training. Evaluation indicators include accuracy, sensitivity, specificity, and Area Under Curve (AUC) value. The results in the Fus2Net model had an accuracy of 92%, the sensitivity reached 95.65%, the specificity reached 88.89%, and the AUC value reached 0.97 for classifying BUS tumor images. Conclusions : The experiment compared the existing CNN categorized architecture, and the Fus2Net architecture we customed has more advantages in a comprehensive performance. The obtained results demonstrated that the Fus2Net classification method we proposed can better assist radiologists in the diagnosis of benign and malignant BUS tumor images. Methods : The existing public datasets are small and the amount of data suffer from the balance issue. In this paper, we provide a relatively larger dataset with a total of 1052 ultrasound images, including 696 benign images and 356 malignant images, which were collected from a local hospital. We proposed a novel CNN named Fus2Net for the benign and malignant classification of BUS tumor images and it contains two self-designed feature extraction modules. To evaluate how the classifier generalizes on the experimental dataset, 10-fold cross validation was employed. Meanwhile, to solve the balance of the dataset, the training data was augmented before being fed into the Fus2Net. In the experiment, we used hyperparameter fine-tuning and regularization technology to make the Fus2Net convergence. Biomedical Engineering Convolutional neural network Deep learning Data augmentation Breast ultrasound tumor images Classification Full Text Cite Share Download PDF Status: Published Journal Publication published 18 Nov, 2021 Read the published version in BioMedical Engineering OnLine → Version 1 posted Review # 3 received at journal 23 Sep, 2021 Review # 2 received at journal 14 Sep, 2021 Reviewer # 5 agreed at journal 12 Sep, 2021 Reviewer # 4 agreed at journal 12 Sep, 2021 Reviewer # 3 agreed at journal 10 Sep, 2021 Reviewer # 2 agreed at journal 10 Sep, 2021 Review # 1 received at journal 09 Sep, 2021 Reviews received at journal 09 Sep, 2021 Reviewer # 1 agreed at journal 08 Sep, 2021 Editor assigned by journal 08 Sep, 2021 Reviewers invited by journal 07 Sep, 2021 Submission checks completed at journal 07 Sep, 2021 Editor invited by journal 07 Sep, 2021 First submitted to journal 02 Sep, 2021 Editorial decision: Major revision 28 Aug, 2021 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-853246","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":51209449,"identity":"19b5fbe9-e373-4766-9908-07394240fbbb","order_by":0,"name":"He Ma","email":"","orcid":"","institution":"Northeastern University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"He","middleName":"","lastName":"Ma","suffix":""},{"id":51209450,"identity":"76c73144-34ae-464f-9480-3db5debd675f","order_by":1,"name":"Ronghui Tian","email":"","orcid":"https://orcid.org/0000-0002-2051-5680","institution":"Northeastern University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ronghui","middleName":"","lastName":"Tian","suffix":""},{"id":51209451,"identity":"e9ec767d-e707-47de-805d-6f200f2ba7ee","order_by":2,"name":"Hong Li","email":"","orcid":"","institution":"Northeastern University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Hong","middleName":"","lastName":"Li","suffix":""},{"id":51209452,"identity":"54bbf1a8-cdf3-4622-96f0-9db93c87a950","order_by":3,"name":"Hang Sun","email":"","orcid":"","institution":"Northeastern University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Hang","middleName":"","lastName":"Sun","suffix":""},{"id":51209453,"identity":"bc86330f-f0e5-4202-86e6-6b208b73d727","order_by":4,"name":"Guoxiu Lu","email":"","orcid":"","institution":"Northeastern University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Guoxiu","middleName":"","lastName":"Lu","suffix":""},{"id":51209454,"identity":"e0059eb8-2b04-4019-b6b5-6731072c0b27","order_by":5,"name":"Ruibo Liu","email":"","orcid":"","institution":"Northeastern University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ruibo","middleName":"","lastName":"Liu","suffix":""},{"id":51209455,"identity":"d434fa8a-248a-4d9e-b6eb-3950d8867adf","order_by":6,"name":"Zhiguo Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAElEQVRIie3RsWoCMRjA8RwBp9isCfQhPhGuFuTuVSIHNx2dbyopB9lKV/sWTuKYQ/CWVNcDFzcpONyN0g6NCHYy51ho/kMI4ftBQhDy+f5qgWR2pfuv73xMKJU3E4IYMek9n+qbCDqTvlqOQQr3NFQfy8/jYhTNqlcNzGwIIB00beYg5ikdccOSmVkLAfmWPGCJ+fv8Ogl1FsJAsQRqAVqYLXmUuof7LrI5hDA5k4Es1ZpY2EHqbLgrFYvAboIXpbtJXB9CO8kEN6sUI5MQPi0L51v4WzZsj+o5vquKFUZ5FFNalE3rILbe6R8n8vcgkFcmL+HmdMGuKZ/P5/vH/QB1MVl2K1/7gAAAAABJRU5ErkJggg==","orcid":"","institution":"General Hospital of Northern Theatre command","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Zhiguo","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2021-08-28 10:55:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-853246/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-853246/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12938-021-00950-z","type":"published","date":"2021-11-18T11:54:43+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":15662805,"identity":"cf322496-0e7c-4c96-8ba8-8a5b9d866fa0","added_by":"auto","created_at":"2021-11-18 11:54:54","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1820147,"visible":true,"origin":"","legend":"","description":"","filename":"BMEOD2100596R1reviewer.pdf","url":"https://assets-eu.researchsquare.com/files/rs-853246/v1_covered.pdf"},{"id":13781175,"identity":"9ea098e6-8c19-47a9-8950-5db8de2f9881","added_by":"auto","created_at":"2021-09-20 15:26:36","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1815088,"visible":true,"origin":"","legend":"","description":"","filename":"BMEOD2100596R1reviewer.pdf","url":"https://assets-eu.researchsquare.com/files/rs-853246/v1_covered.pdf"}],"financialInterests":"","formattedTitle":"Fus2Net: A Novel Convolutional Neural Network for Classification of Benign and Malignant Breast Tumor in Ultrasound Images","fulltext":[{"header":"Full Text","content":"This preprint is available for \u003ca href='/article/rs-853246/latest.pdf' target='_blank'\u003edownload as a PDF\u003c/a\u003e."}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"biomedical-engineering-online","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bmeo","sideBox":"Learn more about [BioMedical Engineering OnLine](http://biomedical-engineering-online.biomedcentral.com/)","snPcode":"12938","submissionUrl":"https://submission.nature.com/new-submission/12938/3","title":"BioMedical Engineering OnLine","twitterHandle":"@BioMedCentral","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Convolutional neural network, Deep learning, Data augmentation, Breast ultrasound tumor images, Classification","lastPublishedDoi":"10.21203/rs.3.rs-853246/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-853246/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e: The rapid development of artificial intelligence technology has improved the capability of automatic breast cancer diagnosis, compared to traditional machine learning methods. Convolutional Neural Network (CNN) can automatically select high-efficiency features, which helps to improve the level of computer-aided diagnosis (CAD). It can improve the performance of distinguishing benign and malignant breast ultrasound (BUS) tumor images and makes rapid breast tumor screening possible. \u003c/p\u003e\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e: The classification model was evaluated by using BUS tumor images without training. Evaluation indicators include accuracy, sensitivity, specificity, and Area Under Curve (AUC) value. The results in the Fus2Net model had an accuracy of 92%, the sensitivity reached 95.65%, the specificity reached 88.89%, and the AUC value reached 0.97 for classifying BUS tumor images. \u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConclusions\u003c/strong\u003e: The experiment compared the existing CNN categorized architecture, and the Fus2Net architecture we customed has more advantages in a comprehensive performance. The obtained results demonstrated that the Fus2Net classification method we proposed can better assist radiologists in the diagnosis of benign and malignant BUS tumor images. \u003c/p\u003e\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e: The existing public datasets are small and the amount of data suffer from the balance issue. In this paper, we provide a relatively larger dataset with a total of 1052 ultrasound images, including 696 benign images and 356 malignant images, which were collected from a local hospital. We proposed a novel CNN named Fus2Net for the benign and malignant classification of BUS tumor images and it contains two self-designed feature extraction modules. To evaluate how the classifier generalizes on the experimental dataset, 10-fold cross validation was employed. Meanwhile, to solve the balance of the dataset, the training data was augmented before being fed into the Fus2Net. In the experiment, we used hyperparameter fine-tuning and regularization technology to make the Fus2Net convergence.\u003c/p\u003e","manuscriptTitle":"Fus2Net: A Novel Convolutional Neural Network for Classification of Benign and Malignant Breast Tumor in Ultrasound Images","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2021-09-20 15:26:30","doi":"10.21203/rs.3.rs-853246/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2021-09-24T00:00:00+00:00","index":3,"fulltext":"Recommendation: Reviewer's comments unavailable due to the journal's policy.\n"},{"type":"editorInvitedReview","content":"","date":"2021-09-15T00:00:00+00:00","index":2,"fulltext":"Recommendation: Reviewer's comments unavailable due to the journal's policy.\n"},{"type":"reviewerAgreed","content":"","date":"2021-09-13T01:00:00+00:00","index":5,"fulltext":""},{"type":"reviewerAgreed","content":"","date":"2021-09-13T00:00:00+00:00","index":4,"fulltext":""},{"type":"reviewerAgreed","content":"","date":"2021-09-11T01:00:00+00:00","index":3,"fulltext":""},{"type":"reviewerAgreed","content":"","date":"2021-09-11T00:00:00+00:00","index":2,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2021-09-10T00:00:00+00:00","index":1,"fulltext":"Recommendation: Reviewer's comments unavailable due to the journal's policy.\n"},{"type":"editorInvitedReview","content":"","date":"2021-09-09T11:14:49+00:00","index":0,"fulltext":""},{"type":"reviewerAgreed","content":"","date":"2021-09-09T00:00:00+00:00","index":1,"fulltext":""},{"type":"editorAssigned","content":"","date":"2021-09-08T13:33:48+00:00","index":"","fulltext":""},{"type":"reviewersInvited","content":"","date":"2021-09-08T00:00:00+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2021-09-07T23:00:00+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2021-09-07T23:00:00+00:00","index":"","fulltext":""},{"type":"submitted","content":"BioMedical Engineering OnLine","date":"2021-09-03T02:50:12+00:00","index":"","fulltext":""},{"type":"decision","content":"Major revision","date":"2021-08-28T11:41:39+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"biomedical-engineering-online","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bmeo","sideBox":"Learn more about [BioMedical Engineering OnLine](http://biomedical-engineering-online.biomedcentral.com/)","snPcode":"12938","submissionUrl":"https://submission.nature.com/new-submission/12938/3","title":"BioMedical Engineering OnLine","twitterHandle":"@BioMedCentral","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f6e09c3d-a8c5-44e1-abe7-8ff53f9cc6d5","owner":[],"postedDate":"September 20th, 2021","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":7300538,"name":"Biomedical Engineering"}],"tags":[],"updatedAt":"2021-11-18T11:54:43+00:00","versionOfRecord":{"articleIdentity":"rs-853246","link":"https://doi.org/10.1186/s12938-021-00950-z","journal":{"identity":"biomedical-engineering-online","isVorOnly":false,"title":"BioMedical Engineering OnLine"},"publishedOn":"2021-11-18 11:54:43","publishedOnDateReadable":"November 18th, 2021"},"versionCreatedAt":"2021-09-20 15:26:30","video":"","vorDoi":"10.1186/s12938-021-00950-z","vorDoiUrl":"https://doi.org/10.1186/s12938-021-00950-z","workflowStages":[]},"version":"v1","identity":"rs-853246","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-853246","identity":"rs-853246","version":["v1"]},"buildId":"FbvkV6FR0MCFSLy54lSbu","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}
Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.
My notes (saved in your browser only)
Ask this paper
Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works
Citation neighborhood (no data yet)
We don't have any in-corpus citations linked to this paper yet. The paper's references may be in our DB but unresolved to ``paper_id`` (resolution happens at ingest when the cited DOI matches a row we already have). Run the cross-source citation reconcile pass to retry.
Source provenance
- europepmc
- last seen: 2026-05-19T01:45:01.086888+00:00