Development of a Deep Learning Algorithm for Periapical Disease Detection in Dental Radiographs
Periapical radiolucencies, which can be detected on panoramic radiographs, are one of the most common radiographic findings in dentistry and have a differential diagnosis including infections, granuloma, cysts and tumors. In this study, we seek to investigate the ability with which 24 oral and maxil...
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-06-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/10/6/430 |
| id |
doaj-c5c2eca4b1d54875b15e36e73f38e454 |
|---|---|
| record_format |
Article |
| collection |
DOAJ |
| language |
English |
| format |
Article |
| sources |
DOAJ |
| author |
Michael G. Endres Florian Hillen Marios Salloumis Ahmad R. Sedaghat Stefan M. Niehues Olivia Quatela Henning Hanken Ralf Smeets Benedicta Beck-Broichsitter Carsten Rendenbach Karim Lakhani Max Heiland Robert A. Gaudin |
| spellingShingle |
Michael G. Endres Florian Hillen Marios Salloumis Ahmad R. Sedaghat Stefan M. Niehues Olivia Quatela Henning Hanken Ralf Smeets Benedicta Beck-Broichsitter Carsten Rendenbach Karim Lakhani Max Heiland Robert A. Gaudin Development of a Deep Learning Algorithm for Periapical Disease Detection in Dental Radiographs Diagnostics artificial intelligence diagnosis computer-assisted image interpretation computer-assisted machine learning |
| author_facet |
Michael G. Endres Florian Hillen Marios Salloumis Ahmad R. Sedaghat Stefan M. Niehues Olivia Quatela Henning Hanken Ralf Smeets Benedicta Beck-Broichsitter Carsten Rendenbach Karim Lakhani Max Heiland Robert A. Gaudin |
| author_sort |
Michael G. Endres |
| title |
Development of a Deep Learning Algorithm for Periapical Disease Detection in Dental Radiographs |
| title_short |
Development of a Deep Learning Algorithm for Periapical Disease Detection in Dental Radiographs |
| title_full |
Development of a Deep Learning Algorithm for Periapical Disease Detection in Dental Radiographs |
| title_fullStr |
Development of a Deep Learning Algorithm for Periapical Disease Detection in Dental Radiographs |
| title_full_unstemmed |
Development of a Deep Learning Algorithm for Periapical Disease Detection in Dental Radiographs |
| title_sort |
development of a deep learning algorithm for periapical disease detection in dental radiographs |
| publisher |
MDPI AG |
| series |
Diagnostics |
| issn |
2075-4418 |
| publishDate |
2020-06-01 |
| description |
Periapical radiolucencies, which can be detected on panoramic radiographs, are one of the most common radiographic findings in dentistry and have a differential diagnosis including infections, granuloma, cysts and tumors. In this study, we seek to investigate the ability with which 24 oral and maxillofacial (OMF) surgeons assess the presence of periapical lucencies on panoramic radiographs, and we compare these findings to the performance of a predictive deep learning algorithm that we have developed using a curated data set of 2902 de-identified panoramic radiographs. The mean diagnostic positive predictive value (PPV) of OMF surgeons based on their assessment of panoramic radiographic images was 0.69(± 0.13), indicating that dentists on average falsely diagnose 31% of cases as radiolucencies. However, the mean diagnostic true positive rate (TPR) was 0.51(± 0.14), indicating that on average 49% of all radiolucencies were missed. We demonstrate that the deep learning algorithm achieves a better performance than 14 of 24 OMF surgeons within the cohort, exhibiting an average precision of 0.60(± 0.04), and an F<sub>1 </sub>score of 0.58(± 0.04) corresponding to a PPV of 0.67(± 0.05) and TPR of 0.51(± 0.05). The algorithm, trained on limited data and evaluated on clinically validated ground truth, has potential to assist OMF surgeons in detecting periapical lucencies on panoramic radiographs. |
| topic |
artificial intelligence diagnosis computer-assisted image interpretation computer-assisted machine learning |
| url |
https://www.mdpi.com/2075-4418/10/6/430 |
| work_keys_str_mv |
AT michaelgendres developmentofadeeplearningalgorithmforperiapicaldiseasedetectionindentalradiographs AT florianhillen developmentofadeeplearningalgorithmforperiapicaldiseasedetectionindentalradiographs AT mariossalloumis developmentofadeeplearningalgorithmforperiapicaldiseasedetectionindentalradiographs AT ahmadrsedaghat developmentofadeeplearningalgorithmforperiapicaldiseasedetectionindentalradiographs AT stefanmniehues developmentofadeeplearningalgorithmforperiapicaldiseasedetectionindentalradiographs AT oliviaquatela developmentofadeeplearningalgorithmforperiapicaldiseasedetectionindentalradiographs AT henninghanken developmentofadeeplearningalgorithmforperiapicaldiseasedetectionindentalradiographs AT ralfsmeets developmentofadeeplearningalgorithmforperiapicaldiseasedetectionindentalradiographs AT benedictabeckbroichsitter developmentofadeeplearningalgorithmforperiapicaldiseasedetectionindentalradiographs AT carstenrendenbach developmentofadeeplearningalgorithmforperiapicaldiseasedetectionindentalradiographs AT karimlakhani developmentofadeeplearningalgorithmforperiapicaldiseasedetectionindentalradiographs AT maxheiland developmentofadeeplearningalgorithmforperiapicaldiseasedetectionindentalradiographs AT robertagaudin developmentofadeeplearningalgorithmforperiapicaldiseasedetectionindentalradiographs |
| _version_ |
1725134139609317376 |
| spelling |
doaj-c5c2eca4b1d54875b15e36e73f38e4542020-11-25T01:20:26ZengMDPI AGDiagnostics2075-44182020-06-011043043010.3390/diagnostics10060430Development of a Deep Learning Algorithm for Periapical Disease Detection in Dental RadiographsMichael G. Endres0Florian Hillen1Marios Salloumis2Ahmad R. Sedaghat3Stefan M. Niehues4Olivia Quatela5Henning Hanken6Ralf Smeets7Benedicta Beck-Broichsitter8Carsten Rendenbach9Karim Lakhani10Max Heiland11Robert A. Gaudin12Laboratory for Innovation Science, Harvard University, 175 N. Harvard Street, Suite 1350, Boston, MA 02134, USALaboratory for Innovation Science, Harvard University, 175 N. Harvard Street, Suite 1350, Boston, MA 02134, USADepartment of Oral- and Maxillofacial Surgery, Charité-Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Hindenburgdamm 30, 12203 Berlin, GermanyDepartment of Otolaryngology—Head and Neck Surgery, University of Cincinnati College of Medicine, Medical Sciences Building Room 6410, 231 Albert Sabin Way, Cincinnati, OH 45267, USADepartment of Radiology, Charité-Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Hindenburgdamm 30, 12203 Berlin, GermanyDepartment of Oral- and Maxillofacial Surgery, Universitätsklinikum Hamburg, Eppendorf, Maritnistraße 52, 20246 Hamburg, GermanyDepartment of Oral- and Maxillofacial Surgery, Universitätsklinikum Hamburg, Eppendorf, Maritnistraße 52, 20246 Hamburg, GermanyDepartment of Oral- and Maxillofacial Surgery, Universitätsklinikum Hamburg, Eppendorf, Maritnistraße 52, 20246 Hamburg, GermanyDepartment of Oral- and Maxillofacial Surgery, Charité-Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Hindenburgdamm 30, 12203 Berlin, GermanyDepartment of Oral- and Maxillofacial Surgery, Charité-Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Hindenburgdamm 30, 12203 Berlin, GermanyLaboratory for Innovation Science, Harvard University, 175 N. Harvard Street, Suite 1350, Boston, MA 02134, USADepartment of Oral- and Maxillofacial Surgery, Charité-Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Hindenburgdamm 30, 12203 Berlin, GermanyLaboratory for Innovation Science, Harvard University, 175 N. Harvard Street, Suite 1350, Boston, MA 02134, USAPeriapical radiolucencies, which can be detected on panoramic radiographs, are one of the most common radiographic findings in dentistry and have a differential diagnosis including infections, granuloma, cysts and tumors. In this study, we seek to investigate the ability with which 24 oral and maxillofacial (OMF) surgeons assess the presence of periapical lucencies on panoramic radiographs, and we compare these findings to the performance of a predictive deep learning algorithm that we have developed using a curated data set of 2902 de-identified panoramic radiographs. The mean diagnostic positive predictive value (PPV) of OMF surgeons based on their assessment of panoramic radiographic images was 0.69(± 0.13), indicating that dentists on average falsely diagnose 31% of cases as radiolucencies. However, the mean diagnostic true positive rate (TPR) was 0.51(± 0.14), indicating that on average 49% of all radiolucencies were missed. We demonstrate that the deep learning algorithm achieves a better performance than 14 of 24 OMF surgeons within the cohort, exhibiting an average precision of 0.60(± 0.04), and an F<sub>1 </sub>score of 0.58(± 0.04) corresponding to a PPV of 0.67(± 0.05) and TPR of 0.51(± 0.05). The algorithm, trained on limited data and evaluated on clinically validated ground truth, has potential to assist OMF surgeons in detecting periapical lucencies on panoramic radiographs.https://www.mdpi.com/2075-4418/10/6/430artificial intelligencediagnosiscomputer-assistedimage interpretationcomputer-assistedmachine learning |