東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

FindBook

Google Book

Amazon

博客來

Diagnosing Periodontal Diseases on Radiographs Using a Deep Learning Approach.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Diagnosing Periodontal Diseases on Radiographs Using a Deep Learning Approach./
作者:	Nelson, Jiman.
面頁冊數:	1 online resource (60 pages)
附註:	Source: Masters Abstracts International, Volume: 84-01.
Contained By:	Masters Abstracts International84-01.
標題:	Dentistry. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29257409click for full text (PQDT)
ISBN:	9798837500596

Diagnosing Periodontal Diseases on Radiographs Using a Deep Learning Approach.
Nelson, Jiman.

Diagnosing Periodontal Diseases on Radiographs Using a Deep Learning Approach. - 1 online resource (60 pages)

Source: Masters Abstracts International, Volume: 84-01.

Thesis (M.Sc.D.)--The University of Texas School of Dentistry at Houston, 2022.

Includes bibliographical references

OBJECTIVE: Assessing radiographic bone level is important for periodontal diagnosis. The interpretation of radiographic images is subjective, and accuracy depends on a clinician's experience and knowledge. Artificial intelligence and image analysis can improve reliability.This study aimed to develop and validate a deep learning model that can accurately diagnose periodontal disease, using radiographic bone loss, similar to a trained dental clinician. Also, an online user interface was developed for clinical use and it was validated for the accuracy and usability.MATERIALS AND METHODS: An end-to-end deep learning-based computer aided diagnosis (CAD) was developed to calculate radiographic bone loss (RBL) percentage and distance, assign RBL stages, and suggest a case-level periodontal diagnosis from periapical and bitewing radiographs following 2018 periodontitis classification. This bone assessment model was integrated with an online interface to process uploaded full-mouth series intraoral radiographs (FMS) and generate a comprehensive clinical report.All images were annotated by calibrated examiners and randomly divided to 70%, 10%, and 20% for training, validation, and testing. U-Net with ResNet-34 encoder was used for segmentation models. The segmentation models were validated by Dice Similarity Coefficient (DSC), Jaccard Index (JI), and Pixel Accuracy (PAC). Performance of bone assessment was evaluated by the Area Under the Receiver Operating Characteristic Curve (AUROC), sensitivity and specificity. Comparisons of measurements between the CAD model and experts were analyzed by Student's t-test. The diagnosis accuracy of independent cases was assessed based on the expert-assigned diagnosis. The power analysis was performed to ensure the required image numbers for different models at a significance level of 0.05 and a power of >0.80.RESULTS: In total, 1147 periapical and bitewing radiographic images were analyzed. The performance of segmentation models is highly acceptable. The AUROC values of RBL stage assignment for no bone loss, stage I, stage II, and stage III were 0.98, 0.89, 0.90 and 0.90, respectively. Sensitivity and specificity for different RBL stages are all above 0.8. The time required to generate a comprehensive report for each FMS by the experts (20.40 ± 0.94 minutes) was significantly longer than that by the CAD model (2.16 ± 0.05 minutes) (p<0.01). The accuracy of 100 clinical case diagnosis was 0.84. The accuracy for extent (0.90), stage (0.90) and grade (0.98) was generally high.CONCLUSIONS: The proposed deep learning-based model can reliably assess the radiographic bone level and assign radiographic bone loss stage. The proposed model and interface can assist the clinician in making an accurate periodontal diagnosis. Furthermore, it is useful for the review of a large number of intraoral radiographic images for quality control of clinical diagnosis and research purposes related to periodontal diseases.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2023

Mode of access: World Wide Web

ISBN: 9798837500596Subjects--Topical Terms:

828971
Dentistry.
Subjects--Index Terms:

Deep learningIndex Terms--Genre/Form:

542853
Electronic books.

Diagnosing Periodontal Diseases on Radiographs Using a Deep Learning Approach.
LDR:04258nmm a2200349K 4500 001 2360258
005 20230926101825.5
006 m o d
007 cr mn ---uuuuu
008 241011s2022 xx obm 000 0 eng d
020 $a 9798837500596
035 $a (MiAaPQ)AAI29257409
035 $a AAI29257409
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Nelson, Jiman. $3 3700873
245 1 0 $a Diagnosing Periodontal Diseases on Radiographs Using a Deep Learning Approach.
264 0 $c 2022
300 $a 1 online resource (60 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Masters Abstracts International, Volume: 84-01.
500 $a Advisor: Lee, Chun-Teh.
502 $a Thesis (M.Sc.D.)--The University of Texas School of Dentistry at Houston, 2022.
504 $a Includes bibliographical references
520 $a OBJECTIVE: Assessing radiographic bone level is important for periodontal diagnosis. The interpretation of radiographic images is subjective, and accuracy depends on a clinician's experience and knowledge. Artificial intelligence and image analysis can improve reliability.This study aimed to develop and validate a deep learning model that can accurately diagnose periodontal disease, using radiographic bone loss, similar to a trained dental clinician. Also, an online user interface was developed for clinical use and it was validated for the accuracy and usability.MATERIALS AND METHODS: An end-to-end deep learning-based computer aided diagnosis (CAD) was developed to calculate radiographic bone loss (RBL) percentage and distance, assign RBL stages, and suggest a case-level periodontal diagnosis from periapical and bitewing radiographs following 2018 periodontitis classification. This bone assessment model was integrated with an online interface to process uploaded full-mouth series intraoral radiographs (FMS) and generate a comprehensive clinical report.All images were annotated by calibrated examiners and randomly divided to 70%, 10%, and 20% for training, validation, and testing. U-Net with ResNet-34 encoder was used for segmentation models. The segmentation models were validated by Dice Similarity Coefficient (DSC), Jaccard Index (JI), and Pixel Accuracy (PAC). Performance of bone assessment was evaluated by the Area Under the Receiver Operating Characteristic Curve (AUROC), sensitivity and specificity. Comparisons of measurements between the CAD model and experts were analyzed by Student's t-test. The diagnosis accuracy of independent cases was assessed based on the expert-assigned diagnosis. The power analysis was performed to ensure the required image numbers for different models at a significance level of 0.05 and a power of >0.80.RESULTS: In total, 1147 periapical and bitewing radiographic images were analyzed. The performance of segmentation models is highly acceptable. The AUROC values of RBL stage assignment for no bone loss, stage I, stage II, and stage III were 0.98, 0.89, 0.90 and 0.90, respectively. Sensitivity and specificity for different RBL stages are all above 0.8. The time required to generate a comprehensive report for each FMS by the experts (20.40 ± 0.94 minutes) was significantly longer than that by the CAD model (2.16 ± 0.05 minutes) (p<0.01). The accuracy of 100 clinical case diagnosis was 0.84. The accuracy for extent (0.90), stage (0.90) and grade (0.98) was generally high.CONCLUSIONS: The proposed deep learning-based model can reliably assess the radiographic bone level and assign radiographic bone loss stage. The proposed model and interface can assist the clinician in making an accurate periodontal diagnosis. Furthermore, it is useful for the review of a large number of intraoral radiographic images for quality control of clinical diagnosis and research purposes related to periodontal diseases.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2023
538 $a Mode of access: World Wide Web
650 4 $a Dentistry. $3 828971
653 $a Deep learning
653 $a Dental radiographs
653 $a Periodontal diagnosis
655 7 $a Electronic books. $2 lcsh $3 542853
690 $a 0567
710 2 $a ProQuest Information and Learning Co. $3 783688
710 2 $a The University of Texas School of Dentistry at Houston. $b Periodontics. $3 3685022
773 0 $t Masters Abstracts International $g 84-01.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29257409 $z click for full text (PQDT)