Comparing the accuracy of two machine learning models in detection and classification of periapical lesions using periapical radiographs.
CNN
Deep learning
Periapical index
Periapical lesion
Journal
Oral radiology
ISSN: 1613-9674
Titre abrégé: Oral Radiol
Pays: Japan
ID NLM: 8806621
Informations de publication
Date de publication:
11 Jun 2024
11 Jun 2024
Historique:
received:
17
09
2023
accepted:
31
05
2024
medline:
12
6
2024
pubmed:
12
6
2024
entrez:
11
6
2024
Statut:
aheadofprint
Résumé
Previous deep learning-based studies were mainly conducted on detecting periapical lesions; limited information in classification, such as the periapical index (PAI) scoring system, is available. The study aimed to apply two deep learning models, Faster R-CNN and YOLOv4, in detecting and classifying periapical lesions using the PAI score from periapical radiographs (PR) in three different regions of the dental arch: anterior teeth, premolars, and molars. Out of 2658 PR selected for the study, 2122 PR were used for training, 268 PR were used for validation and 268 PR were used for testing. The diagnosis made by experienced dentists was used as the reference diagnosis. The Faster R-CNN and YOLOv4 models obtained great sensitivity, specificity, accuracy, and precision for detecting periapical lesions. No clear difference in the performance of both models among these three regions was found. The true prediction of Faster R-CNN was 89%, 83.01% and 91.84% for PAI 3, PAI 4 and PAI 5 lesions, respectively. The corresponding values of YOLOv4 were 68.06%, 50.94%, and 65.31%. Our study demonstrated the potential of YOLOv4 and Faster R-CNN models for detecting and classifying periapical lesions based on the PAI scoring system using periapical radiographs.
Sections du résumé
BACKGROUND
BACKGROUND
Previous deep learning-based studies were mainly conducted on detecting periapical lesions; limited information in classification, such as the periapical index (PAI) scoring system, is available. The study aimed to apply two deep learning models, Faster R-CNN and YOLOv4, in detecting and classifying periapical lesions using the PAI score from periapical radiographs (PR) in three different regions of the dental arch: anterior teeth, premolars, and molars.
METHODS
METHODS
Out of 2658 PR selected for the study, 2122 PR were used for training, 268 PR were used for validation and 268 PR were used for testing. The diagnosis made by experienced dentists was used as the reference diagnosis.
RESULTS
RESULTS
The Faster R-CNN and YOLOv4 models obtained great sensitivity, specificity, accuracy, and precision for detecting periapical lesions. No clear difference in the performance of both models among these three regions was found. The true prediction of Faster R-CNN was 89%, 83.01% and 91.84% for PAI 3, PAI 4 and PAI 5 lesions, respectively. The corresponding values of YOLOv4 were 68.06%, 50.94%, and 65.31%.
CONCLUSIONS
CONCLUSIONS
Our study demonstrated the potential of YOLOv4 and Faster R-CNN models for detecting and classifying periapical lesions based on the PAI scoring system using periapical radiographs.
Identifiants
pubmed: 38862834
doi: 10.1007/s11282-024-00759-1
pii: 10.1007/s11282-024-00759-1
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Informations de copyright
© 2024. The Author(s) under exclusive licence to Japanese Society for Oral and Maxillofacial Radiology.
Références
Huumonen S, Suominen A, Vehkalahti M. Prevalence of apical periodontitis in root filled teeth: findings from a nationwide survey in finland. Int Endodontic J. 2017;50(3):229–36.
doi: 10.1111/iej.12625
Nair PR. Pathogenesis of apical periodontitis and the causes of endodontic failures. Crit Rev Oral Biol Med. 2004;15(6):348–81.
doi: 10.1177/154411130401500604
pubmed: 15574679
Antony DP, Thomas T, Nivedhitha M. Two-dimensional periapical, panoramic radiography versus three-dimensional cone-beam computed tomography in the detection of periapical lesion after endodontic treatment: a systematic review. Cureus. 2020. https://doi.org/10.7759/cureus.7736 .
doi: 10.7759/cureus.7736
pubmed: 33354483
pmcid: 7746328
International atomic energy agency, radiation protection in dental radiology No.108 [IAEA Preprint]. Safety reports series. 2021 53:(23) 81.
Mazzi-Chaves JF, et al. Cone-beam computed tomographic–based assessment of filled C-shaped canals: artifact expression of cone-beam computed tomography as opposed to micro–computed tomography and nano–computed tomography. J Endodontic. 2020;46(11):1702–11.
doi: 10.1016/j.joen.2020.07.010
Corbella S, Srinivas S, Cabitza F. Applications of deep learning in dentistry. Oral Surg Oral Med Oral Pathol Oral Radiol. 2021;132(2):225–38.
doi: 10.1016/j.oooo.2020.11.003
pubmed: 33303419
Arsiwala-Scheppach LT, et al. Machine learning in dentistry: a scoping review. J Clin Med. 2023;12(3):937.
doi: 10.3390/jcm12030937
pubmed: 36769585
pmcid: 9918184
Pathak AR, Pandey M, Rautaray S. Application of deep learning for object detection. Procedia computer sci. 2018;132:1706–17.
doi: 10.1016/j.procs.2018.05.144
Thanh MTG, et al. Deep learning application in dental caries detection using intraoral photos taken by smartphones. Appl Sci. 2022;12(11):5504.
doi: 10.3390/app12115504
Krois J, et al. Deep learning for the radiographic detection of periodontal bone loss. Sci Rep. 2019;9(1):8495.
doi: 10.1038/s41598-019-44839-3
pubmed: 31186466
pmcid: 6560098
Moidu NP, et al. Deep learning for categorization of endodontic lesion based on radiographic periapical index scoring system. Clin Oral Investig. 2022;26(1):651–8.
doi: 10.1007/s00784-021-04043-y
pubmed: 34213664
Ørstavik D, Kerekes K, Eriksen HM. The periapical index: a scoring system for radiographic assessment of apical periodontitis. Dent Traumatol. 1986;2(1):20–34.
doi: 10.1111/j.1600-9657.1986.tb00119.x
Matijević J, et al. Prevalence of apical periodontitis and quality of root canal fillings in population of Zagreb Croatia: a cross-sectional study. Croat Med J. 2011;52(6):679–87.
doi: 10.3325/cmj.2011.52.679
pubmed: 22180266
pmcid: 3243319
Sidaravicius B, Aleksejuniene J, Eriksen HJDT. Endodontic treatment and prevalence of apical periodontitis in an adult population of Vilnius. Lithuania. 1999;15(5):210–5.
Tsuneishi M, et al. Radiographic evaluation of periapical status and prevalence of endodontic treatment in an adult Japanese population. Oral Surg, Oral Med, Oral Pathol, Oral Radiol, Endodontol. 2005;100(5):631–5.
doi: 10.1016/j.tripleo.2005.07.029
Correia-Sousa J, et al. Apical periodontitis and related risk factors: Cross-sectional study. Revis Port de Estomatol, Med Dent{\’a}ria e Cir Maxilofac. 2015;56(4):226–32.
Sadr S, et al. Deep learning for detection of periapical radiolucent lesions: a systematic review and meta-analysis of diagnostic test accuracy. J Endod. 2023;49(3):248-261.e3.
doi: 10.1016/j.joen.2022.12.007
pubmed: 36563779
Ekert T, et al. Deep Learning for the Radiographic Detection of Apical Lesions. J Endod. 2019;45(7):917-922.e5.
doi: 10.1016/j.joen.2019.03.016
pubmed: 31160078
Li C-W, et al. Detection of dental apical lesions using CNNs on periapical radiograph. Sensors. 2021;21(21):7049.
doi: 10.3390/s21217049
pubmed: 34770356
pmcid: 8588190
Çelik B, et al. The role of deep learning for periapical lesion detection on panoramic radiographs. Dentomaxillofacial Radiol. 2023. https://doi.org/10.1259/dmfr.20230118 .
doi: 10.1259/dmfr.20230118
Ren S, et al. Faster r-cnn: Towards real-time object detection with region proposal networks. Adv neural inf process syst. 2015;39:1137–49.
Bochkovskiy A, Wang CY, Liao HYM (2020) Yolov4: optimal speed and accuracy of object detection. arXiv preprint. arXiv:2004.10934 .
Peretz B, Gotler M, Kaffe I. Common errors in digital panoramic radiographs of patients with mixed dentition and patients with permanent dentition. Int J Dent. 2012;2012:584138.
doi: 10.1155/2012/584138
pubmed: 22505905
pmcid: 3296161
Brynolf I. A histological and roentgenological study of the periapical region of human upper incisors, vol. 18. Almqvist & Wiksell; 1967.
Zehnder M, Belibasakis GN. A critical analysis of research methods to study clinical molecular biomarkers in endodontic research. Int Endod J. 2022;55:37–45.
doi: 10.1111/iej.13647
pubmed: 34655496
Fatima A, et al. Deep learning-based multiclass instance segmentation for dental lesion detection. Healthcare. 2023. https://doi.org/10.3390/healthcare11030347 .
doi: 10.3390/healthcare11030347
pubmed: 36833098
pmcid: 9956031
Shafi I, et al. Teeth lesion detection using deep learning and the internet of things post-COVID-19. Sensors. 2023;23(15):6837.
doi: 10.3390/s23156837
pubmed: 37571620
pmcid: 10422255