Interpretable AI Explores Effective Components of CAD/CAM Resin Composites

Li, H.; Sakai, T.; Tanaka, A.; Ogura, M.; Lee, C.; Yamaguchi, S.; Imazato, S.

doi:https://doi.org/10.1177/00220345221089251

Interpretable AI Explores Effective Components of CAD/CAM Resin Composites

Li, H.; Sakai, T.; Tanaka, A. et al.

Journal of Dental Research, 2022, 101(11), 1363-1371

Number of Access:815（2025-08-15 06:46 Counts）

Identifier to cite or link to this item: https://hdl.handle.net/11094/88501

Link to primary information

File	Format	Terms of use	Size	Views	Date.Available
JDentRes_13_1_002203452210892	pdf	None	377 KB	286	2023.04.15
JDentRes_13_1_002203452210892_Appendix	pdf	None	886 KB	55	2023.04.15
JDentRes_13_1_002203452210892_Figures	pdf	None	3.59 MB	46	2023.04.15
JDentRes_13_1_002203452210892_Tables	pdf	None	139 KB	64	2023.04.15

Item Information

Output File Export EndNote Basic Export Mendeley

Title	Interpretable AI Explores Effective Components of CAD/CAM Resin Composites
Creator	CreatorName Li, H.
	NRID 1000090379082 1000090379082 ORCID 0000-0003-4348-4896 0000-0003-4348-4896 researchmap permalink read0210126 read0210126 Scopus Author ID 55177911100 55177911100 CreatorName Sakai, T.
	CreatorName Tanaka, A.
	CreatorName Ogura, M.
	CreatorName Lee, C.
	NRID 1000030397773 1000030397773 ORCID 0000-0002-4661-8821 0000-0002-4661-8821 researchmap permalink YamaguchiS_30397773 YamaguchiS_30397773 Scopus Author ID 7403873463 7403873463 CreatorName Yamaguchi, S.
	NRID 1000080243244 1000080243244 ORCID 0000-0003-3655-4809 0000-0003-3655-4809 researchmap permalink read0014210 read0014210 Scopus Author ID 55994423600 55994423600 CreatorName Imazato, S.
Description	Li, H., T. Sakai, A. Tanaka, M. Ogura, C. Lee, S. Yamaguchi, and S. Imazato. “Interpretable AI Explores Effective Components of CAD/CAM Resin Composites.” Journal of Dental Research, (April 2022). https://doi.org/10.1177/00220345221089251.
Abstract	High flexural strength of computer-aided manufacturing resin composite blocks (CAD/CAM RCBs) are required in clinical scenarios. However, the conventional in vitro approach of modifying materials’ composition by trial and error was not efficient to explore the effective components that contribute to the flexural strength. Machine learning (ML) is a powerful tool to achieve the above goals. Therefore, the aim of this study was to develop ML models to predict the flexural strength of CAD/CAM RCBs and explore the components that affect flexural strength as the first step. The composition of 12 commercially available products and flexural strength were collected from the manufacturers and literature. The initial data consisted of 16 attributes and 12 samples. Considering that the input data for each sample were recognized as a multidimensional vector, a fluctuation range of 0.1 was proposed for each vector and the number of samples was augmented to 120. Regression algorithms—that is, random forest (RF), extra trees, gradient boosting decision tree, light gradient boosting machine, and extreme gradient boosting—were used to develop 5 ML models to predict flexural strength. An exhaustive search and feature importance analysis were conducted to analyze the effective components that affected flexural strength. The R2 values for each model were 0.947, 0.997, 0.998, 0.983, and 0.927, respectively. The relative errors of all the algorithms were within 15%. Among the high predicted flexural strength group in the exhaustive search, urethane dimethacrylate was contained in all compositions. Filler content and triethylene glycol dimethacrylate were the top 2 features predicted by all models in the feature importance analysis. ZrSiO4 was the third important feature for all models, except the RF model. The ML models established in this study successfully predicted the flexural strength of CAD/CAM RCBs and identified the effective components that affected flexural strength based on the available data set.
Publisher	SAGE Publications
Source Title	Journal of Dental Research
Volume (Issue)	101(11)
Page	1363-1371
Date of Issued	2022-04-15
Language	English
Handle URL	https://hdl.handle.net/11094/88501
PISSN	00220345
eISSN	15440591
NCID	AA00696534
Relation.isVersionOf	DOI (Publisher Version) https://doi.org/10.1177/00220345221089251
Access Rights	open access
CopyRight	The final version of this paper has been published in Journal of Dental Research, April 15, 2022 by SAGE Publications Ltd, All rights reserved. © Authors, 2022. It is available at: http://journals.sagepub.com/home/JDR.
oaire:version	Accepted Manuscript
Category	学術雑誌論文 Journal Article

Resource Type	journal article
Local Type	Journal Article
DCMI Type	Text
DCTERMS.bibliographicCitation	Journal of Dental Research. 2022, 101(11), p. 1363-1371
DC.title	Interpretable AI Explores Effective Components of CAD/CAM Resin Composites
DC.creator	Li, H.
	Sakai, T.
	Tanaka, A.
	Ogura, M.
	Lee, C.
	Yamaguchi, S.
	Imazato, S.
DC.publisher	SAGE Publications
DC.language' scheme='DCTERMS.RFC1766	English
DC.type' scheme='DCTERMS.DCMIType	Text
DCTERMS.issued' scheme='DCTERMS.W3CDTF	2022-04-15
DC.identifier	https://doi.org/10.1177/00220345221089251
DC.description	Li, H., T. Sakai, A. Tanaka, M. Ogura, C. Lee, S. Yamaguchi, and S. Imazato. “Interpretable AI Explores Effective Components of CAD/CAM Resin Composites.” Journal of Dental Research, (April 2022). https://doi.org/10.1177/00220345221089251.
DCTERMS.abstract	High flexural strength of computer-aided manufacturing resin composite blocks (CAD/CAM RCBs) are required in clinical scenarios. However, the conventional in vitro approach of modifying materials’ composition by trial and error was not efficient to explore the effective components that contribute to the flexural strength. Machine learning (ML) is a powerful tool to achieve the above goals. Therefore, the aim of this study was to develop ML models to predict the flexural strength of CAD/CAM RCBs and explore the components that affect flexural strength as the first step. The composition of 12 commercially available products and flexural strength were collected from the manufacturers and literature. The initial data consisted of 16 attributes and 12 samples. Considering that the input data for each sample were recognized as a multidimensional vector, a fluctuation range of 0.1 was proposed for each vector and the number of samples was augmented to 120. Regression algorithms—that is, random forest (RF), extra trees, gradient boosting decision tree, light gradient boosting machine, and extreme gradient boosting—were used to develop 5 ML models to predict flexural strength. An exhaustive search and feature importance analysis were conducted to analyze the effective components that affected flexural strength. The R2 values for each model were 0.947, 0.997, 0.998, 0.983, and 0.927, respectively. The relative errors of all the algorithms were within 15%. Among the high predicted flexural strength group in the exhaustive search, urethane dimethacrylate was contained in all compositions. Filler content and triethylene glycol dimethacrylate were the top 2 features predicted by all models in the feature importance analysis. ZrSiO4 was the third important feature for all models, except the RF model. The ML models established in this study successfully predicted the flexural strength of CAD/CAM RCBs and identified the effective components that affected flexural strength based on the available data set.
DC.rights	The final version of this paper has been published in Journal of Dental Research, April 15, 2022 by SAGE Publications Ltd, All rights reserved. © Authors, 2022. It is available at: http://journals.sagepub.com/home/JDR.
DC.format	88501
citation_title	Interpretable AI Explores Effective Components of CAD/CAM Resin Composites
citation_author	Li, H.
	Sakai, T.
	Tanaka, A.
	Ogura, M.
	Lee, C.
	Yamaguchi, S.
	Imazato, S.
citation_publisher	SAGE Publications
citation_language	English
citation_date	2022-04-15
citation_journal_title	Journal of Dental Research
citation_volume	101
citation_issue	11
citation_firstpage	1363
citation_lastpage	1371
citation_doi	https://doi.org/10.1177/00220345221089251
citation_public_url	https://hdl.handle.net/11094/88501

Interpretable AI Explores Effective Components of CAD/CAM Resin Composites

Link to primary information

Item Information

Show details