Control experiments on the BUSI dataset.

<div><p>Breast cancer is the most common cancer in women. Breast masses are one of the distinctive signs for diagnosing breast cancer, and ultrasound is widely used for screening as a non-invasive and effective method for breast examination. In this study, we used the Mendeley and BUSI d...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Hao Zhang (15339) (author)
مؤلفون آخرون: He Liang (5151095) (author), Guo Wenjia (20036991) (author), Ma Jing (17687604) (author), Sun Gang (20036994) (author), Ma Hongbing (20036997) (author)
منشور في: 2024
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1852025473887895552
author Hao Zhang (15339)
author2 He Liang (5151095)
Guo Wenjia (20036991)
Ma Jing (17687604)
Sun Gang (20036994)
Ma Hongbing (20036997)
author2_role author
author
author
author
author
author_facet Hao Zhang (15339)
He Liang (5151095)
Guo Wenjia (20036991)
Ma Jing (17687604)
Sun Gang (20036994)
Ma Hongbing (20036997)
author_role author
dc.creator.none.fl_str_mv Hao Zhang (15339)
He Liang (5151095)
Guo Wenjia (20036991)
Ma Jing (17687604)
Sun Gang (20036994)
Ma Hongbing (20036997)
dc.date.none.fl_str_mv 2024-11-01T17:32:14Z
dc.identifier.none.fl_str_mv 10.1371/journal.pone.0307916.t004
dc.relation.none.fl_str_mv https://figshare.com/articles/dataset/Control_experiments_on_the_BUSI_dataset_/27471939
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Cell Biology
Cancer
Space Science
Biological Sciences not elsewhere classified
Mathematical Sciences not elsewhere classified
Information Systems not elsewhere classified
guiding clinical decision
feature extraction capabilities
experimental results demonstrate
dice similarity coefficient
deep learning approaches
comprising 250 images
three attention modules
scale attention modules
channel attention modules
densely connected network
deep learning method
div >< p
diagnosing breast cancer
breast ultrasound images
breast cancer
780 images
multiple attention
attention gates
network ’
breast masses
breast examination
method achieves
effective method
common cancer
source code
relevant features
mass segmentation
encoding stage
distinctive signs
decoding stage
crucial role
better capture
accurate measurement
487 benign
150 malignant
133 normal
100 benign
dc.title.none.fl_str_mv Control experiments on the BUSI dataset.
dc.type.none.fl_str_mv Dataset
info:eu-repo/semantics/publishedVersion
dataset
description <div><p>Breast cancer is the most common cancer in women. Breast masses are one of the distinctive signs for diagnosing breast cancer, and ultrasound is widely used for screening as a non-invasive and effective method for breast examination. In this study, we used the Mendeley and BUSI datasets, comprising 250 images (100 benign, 150 malignant) and 780 images (133 normal, 487 benign, 210 malignant), respectively. The datasets were split into 80% for training and 20% for validation. The accurate measurement and characterization of different breast tumors play a crucial role in guiding clinical decision-making. The area and shape of the different breast tumors detected are critical for clinicians to make accurate diagnostic decisions. In this study, a deep learning method for mass segmentation in breast ultrasound images is proposed, which uses densely connected U-net with attention gates (AGs) as well as channel attention modules and scale attention modules for accurate breast tumor segmentation.The densely connected network is employed in the encoding stage to enhance the network’s feature extraction capabilities. Three attention modules are integrated in the decoding stage to better capture the most relevant features. After validation on the Mendeley and BUSI datasets, the experimental results demonstrate that our method achieves a Dice Similarity Coefficient (DSC) of 0.8764 and 0.8313, respectively, outperforming other deep learning approaches. The source code is located at <a href="http://github.com/zhanghaoCV/plos-one" target="_blank">github.com/zhanghaoCV/plos-one</a>.</p></div>
eu_rights_str_mv openAccess
id Manara_de3ff3d693eced5b4cf850d795d233bb
identifier_str_mv 10.1371/journal.pone.0307916.t004
network_acronym_str Manara
network_name_str ManaraRepo
oai_identifier_str oai:figshare.com:article/27471939
publishDate 2024
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Control experiments on the BUSI dataset.Hao Zhang (15339)He Liang (5151095)Guo Wenjia (20036991)Ma Jing (17687604)Sun Gang (20036994)Ma Hongbing (20036997)Cell BiologyCancerSpace ScienceBiological Sciences not elsewhere classifiedMathematical Sciences not elsewhere classifiedInformation Systems not elsewhere classifiedguiding clinical decisionfeature extraction capabilitiesexperimental results demonstratedice similarity coefficientdeep learning approachescomprising 250 imagesthree attention modulesscale attention moduleschannel attention modulesdensely connected networkdeep learning methoddiv >< pdiagnosing breast cancerbreast ultrasound imagesbreast cancer780 imagesmultiple attentionattention gatesnetwork ’breast massesbreast examinationmethod achieveseffective methodcommon cancersource coderelevant featuresmass segmentationencoding stagedistinctive signsdecoding stagecrucial rolebetter captureaccurate measurement487 benign150 malignant133 normal100 benign<div><p>Breast cancer is the most common cancer in women. Breast masses are one of the distinctive signs for diagnosing breast cancer, and ultrasound is widely used for screening as a non-invasive and effective method for breast examination. In this study, we used the Mendeley and BUSI datasets, comprising 250 images (100 benign, 150 malignant) and 780 images (133 normal, 487 benign, 210 malignant), respectively. The datasets were split into 80% for training and 20% for validation. The accurate measurement and characterization of different breast tumors play a crucial role in guiding clinical decision-making. The area and shape of the different breast tumors detected are critical for clinicians to make accurate diagnostic decisions. In this study, a deep learning method for mass segmentation in breast ultrasound images is proposed, which uses densely connected U-net with attention gates (AGs) as well as channel attention modules and scale attention modules for accurate breast tumor segmentation.The densely connected network is employed in the encoding stage to enhance the network’s feature extraction capabilities. Three attention modules are integrated in the decoding stage to better capture the most relevant features. After validation on the Mendeley and BUSI datasets, the experimental results demonstrate that our method achieves a Dice Similarity Coefficient (DSC) of 0.8764 and 0.8313, respectively, outperforming other deep learning approaches. The source code is located at <a href="http://github.com/zhanghaoCV/plos-one" target="_blank">github.com/zhanghaoCV/plos-one</a>.</p></div>2024-11-01T17:32:14ZDatasetinfo:eu-repo/semantics/publishedVersiondataset10.1371/journal.pone.0307916.t004https://figshare.com/articles/dataset/Control_experiments_on_the_BUSI_dataset_/27471939CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/274719392024-11-01T17:32:14Z
spellingShingle Control experiments on the BUSI dataset.
Hao Zhang (15339)
Cell Biology
Cancer
Space Science
Biological Sciences not elsewhere classified
Mathematical Sciences not elsewhere classified
Information Systems not elsewhere classified
guiding clinical decision
feature extraction capabilities
experimental results demonstrate
dice similarity coefficient
deep learning approaches
comprising 250 images
three attention modules
scale attention modules
channel attention modules
densely connected network
deep learning method
div >< p
diagnosing breast cancer
breast ultrasound images
breast cancer
780 images
multiple attention
attention gates
network ’
breast masses
breast examination
method achieves
effective method
common cancer
source code
relevant features
mass segmentation
encoding stage
distinctive signs
decoding stage
crucial role
better capture
accurate measurement
487 benign
150 malignant
133 normal
100 benign
status_str publishedVersion
title Control experiments on the BUSI dataset.
title_full Control experiments on the BUSI dataset.
title_fullStr Control experiments on the BUSI dataset.
title_full_unstemmed Control experiments on the BUSI dataset.
title_short Control experiments on the BUSI dataset.
title_sort Control experiments on the BUSI dataset.
topic Cell Biology
Cancer
Space Science
Biological Sciences not elsewhere classified
Mathematical Sciences not elsewhere classified
Information Systems not elsewhere classified
guiding clinical decision
feature extraction capabilities
experimental results demonstrate
dice similarity coefficient
deep learning approaches
comprising 250 images
three attention modules
scale attention modules
channel attention modules
densely connected network
deep learning method
div >< p
diagnosing breast cancer
breast ultrasound images
breast cancer
780 images
multiple attention
attention gates
network ’
breast masses
breast examination
method achieves
effective method
common cancer
source code
relevant features
mass segmentation
encoding stage
distinctive signs
decoding stage
crucial role
better capture
accurate measurement
487 benign
150 malignant
133 normal
100 benign