JSCDS: A Core Data Selection Method with Jason-Shannon Divergence for Caries RGB Images-Efficient Learning

Deep learning-based RGB caries detection improves the efficiency of caries identification and is crucial for preventing oral diseases. The performance of deep learning models depends on high-quality data and requires substantial training resources, making efficient deployment challenging. Core data...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhang, Peiliang, Tong, Yujia, Du, Chenghu, Che, Chao, Zhu, Yongjun
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Zhang, Peiliang
Tong, Yujia
Du, Chenghu
Che, Chao
Zhu, Yongjun
description Deep learning-based RGB caries detection improves the efficiency of caries identification and is crucial for preventing oral diseases. The performance of deep learning models depends on high-quality data and requires substantial training resources, making efficient deployment challenging. Core data selection, by eliminating low-quality and confusing data, aims to enhance training efficiency without significantly compromising model performance. However, distance-based data selection methods struggle to distinguish dependencies among high-dimensional caries data. To address this issue, we propose a Core Data Selection Method with Jensen-Shannon Divergence (JSCDS) for efficient caries image learning and caries classification. We describe the core data selection criterion as the distribution of samples in different classes. JSCDS calculates the cluster centers by sample embedding representation in the caries classification network and utilizes Jensen-Shannon Divergence to compute the mutual information between data samples and cluster centers, capturing nonlinear dependencies among high-dimensional data. The average mutual information is calculated to fit the above distribution, serving as the criterion for constructing the core set for model training. Extensive experiments on RGB caries datasets show that JSCDS outperforms other data selection methods in prediction performance and time consumption. Notably, JSCDS exceeds the performance of the full dataset model with only 50% of the core data, with its performance advantage becoming more pronounced in the 70% of core data.
doi_str_mv 10.48550/arxiv.2407.00362
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2407_00362</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2407_00362</sourcerecordid><originalsourceid>FETCH-arxiv_primary_2407_003623</originalsourceid><addsrcrecordid>eNqFzrsOgkAQheFtLIz6AFbOC4DrBTV2Cngh2og9meAAY2DXLBvUt_cSe6tT_Kf4hOiPpDtdeJ4conlw446ncu5KOZmN2-IaxX4QL2EFvjYEAVqEmEpKLWsFR7KFvsCdbQER1lo5cYFKvUvADZmcVEqQaQM-GqYaTts17CvMqXbCLOOUSVk4EBrFKu-KVoZlTb3fdsRgE579nfNVJTfDFZpn8tElX93k_-MF7UZD8A</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>JSCDS: A Core Data Selection Method with Jason-Shannon Divergence for Caries RGB Images-Efficient Learning</title><source>arXiv.org</source><creator>Zhang, Peiliang ; Tong, Yujia ; Du, Chenghu ; Che, Chao ; Zhu, Yongjun</creator><creatorcontrib>Zhang, Peiliang ; Tong, Yujia ; Du, Chenghu ; Che, Chao ; Zhu, Yongjun</creatorcontrib><description>Deep learning-based RGB caries detection improves the efficiency of caries identification and is crucial for preventing oral diseases. The performance of deep learning models depends on high-quality data and requires substantial training resources, making efficient deployment challenging. Core data selection, by eliminating low-quality and confusing data, aims to enhance training efficiency without significantly compromising model performance. However, distance-based data selection methods struggle to distinguish dependencies among high-dimensional caries data. To address this issue, we propose a Core Data Selection Method with Jensen-Shannon Divergence (JSCDS) for efficient caries image learning and caries classification. We describe the core data selection criterion as the distribution of samples in different classes. JSCDS calculates the cluster centers by sample embedding representation in the caries classification network and utilizes Jensen-Shannon Divergence to compute the mutual information between data samples and cluster centers, capturing nonlinear dependencies among high-dimensional data. The average mutual information is calculated to fit the above distribution, serving as the criterion for constructing the core set for model training. Extensive experiments on RGB caries datasets show that JSCDS outperforms other data selection methods in prediction performance and time consumption. Notably, JSCDS exceeds the performance of the full dataset model with only 50% of the core data, with its performance advantage becoming more pronounced in the 70% of core data.</description><identifier>DOI: 10.48550/arxiv.2407.00362</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Computer Vision and Pattern Recognition</subject><creationdate>2024-06</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2407.00362$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2407.00362$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Zhang, Peiliang</creatorcontrib><creatorcontrib>Tong, Yujia</creatorcontrib><creatorcontrib>Du, Chenghu</creatorcontrib><creatorcontrib>Che, Chao</creatorcontrib><creatorcontrib>Zhu, Yongjun</creatorcontrib><title>JSCDS: A Core Data Selection Method with Jason-Shannon Divergence for Caries RGB Images-Efficient Learning</title><description>Deep learning-based RGB caries detection improves the efficiency of caries identification and is crucial for preventing oral diseases. The performance of deep learning models depends on high-quality data and requires substantial training resources, making efficient deployment challenging. Core data selection, by eliminating low-quality and confusing data, aims to enhance training efficiency without significantly compromising model performance. However, distance-based data selection methods struggle to distinguish dependencies among high-dimensional caries data. To address this issue, we propose a Core Data Selection Method with Jensen-Shannon Divergence (JSCDS) for efficient caries image learning and caries classification. We describe the core data selection criterion as the distribution of samples in different classes. JSCDS calculates the cluster centers by sample embedding representation in the caries classification network and utilizes Jensen-Shannon Divergence to compute the mutual information between data samples and cluster centers, capturing nonlinear dependencies among high-dimensional data. The average mutual information is calculated to fit the above distribution, serving as the criterion for constructing the core set for model training. Extensive experiments on RGB caries datasets show that JSCDS outperforms other data selection methods in prediction performance and time consumption. Notably, JSCDS exceeds the performance of the full dataset model with only 50% of the core data, with its performance advantage becoming more pronounced in the 70% of core data.</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Computer Vision and Pattern Recognition</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNqFzrsOgkAQheFtLIz6AFbOC4DrBTV2Cngh2og9meAAY2DXLBvUt_cSe6tT_Kf4hOiPpDtdeJ4conlw446ncu5KOZmN2-IaxX4QL2EFvjYEAVqEmEpKLWsFR7KFvsCdbQER1lo5cYFKvUvADZmcVEqQaQM-GqYaTts17CvMqXbCLOOUSVk4EBrFKu-KVoZlTb3fdsRgE579nfNVJTfDFZpn8tElX93k_-MF7UZD8A</recordid><startdate>20240629</startdate><enddate>20240629</enddate><creator>Zhang, Peiliang</creator><creator>Tong, Yujia</creator><creator>Du, Chenghu</creator><creator>Che, Chao</creator><creator>Zhu, Yongjun</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20240629</creationdate><title>JSCDS: A Core Data Selection Method with Jason-Shannon Divergence for Caries RGB Images-Efficient Learning</title><author>Zhang, Peiliang ; Tong, Yujia ; Du, Chenghu ; Che, Chao ; Zhu, Yongjun</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-arxiv_primary_2407_003623</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Computer Vision and Pattern Recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>Zhang, Peiliang</creatorcontrib><creatorcontrib>Tong, Yujia</creatorcontrib><creatorcontrib>Du, Chenghu</creatorcontrib><creatorcontrib>Che, Chao</creatorcontrib><creatorcontrib>Zhu, Yongjun</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Zhang, Peiliang</au><au>Tong, Yujia</au><au>Du, Chenghu</au><au>Che, Chao</au><au>Zhu, Yongjun</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>JSCDS: A Core Data Selection Method with Jason-Shannon Divergence for Caries RGB Images-Efficient Learning</atitle><date>2024-06-29</date><risdate>2024</risdate><abstract>Deep learning-based RGB caries detection improves the efficiency of caries identification and is crucial for preventing oral diseases. The performance of deep learning models depends on high-quality data and requires substantial training resources, making efficient deployment challenging. Core data selection, by eliminating low-quality and confusing data, aims to enhance training efficiency without significantly compromising model performance. However, distance-based data selection methods struggle to distinguish dependencies among high-dimensional caries data. To address this issue, we propose a Core Data Selection Method with Jensen-Shannon Divergence (JSCDS) for efficient caries image learning and caries classification. We describe the core data selection criterion as the distribution of samples in different classes. JSCDS calculates the cluster centers by sample embedding representation in the caries classification network and utilizes Jensen-Shannon Divergence to compute the mutual information between data samples and cluster centers, capturing nonlinear dependencies among high-dimensional data. The average mutual information is calculated to fit the above distribution, serving as the criterion for constructing the core set for model training. Extensive experiments on RGB caries datasets show that JSCDS outperforms other data selection methods in prediction performance and time consumption. Notably, JSCDS exceeds the performance of the full dataset model with only 50% of the core data, with its performance advantage becoming more pronounced in the 70% of core data.</abstract><doi>10.48550/arxiv.2407.00362</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2407.00362
ispartof
issn
language eng
recordid cdi_arxiv_primary_2407_00362
source arXiv.org
subjects Computer Science - Artificial Intelligence
Computer Science - Computer Vision and Pattern Recognition
title JSCDS: A Core Data Selection Method with Jason-Shannon Divergence for Caries RGB Images-Efficient Learning
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-05T03%3A07%3A38IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=JSCDS:%20A%20Core%20Data%20Selection%20Method%20with%20Jason-Shannon%20Divergence%20for%20Caries%20RGB%20Images-Efficient%20Learning&rft.au=Zhang,%20Peiliang&rft.date=2024-06-29&rft_id=info:doi/10.48550/arxiv.2407.00362&rft_dat=%3Carxiv_GOX%3E2407_00362%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true