Evaluating the accuracy of automated cephalometric analysis based on artificial intelligence
Abstract Background The purpose of this study was to evaluate the accuracy of automatic cephalometric landmark localization and measurements using cephalometric analysis via artificial intelligence (AI) compared with computer-assisted manual analysis. Methods Reconstructed lateral cephalograms (RLCs...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
BMC
2023-04-01
|
Series: | BMC Oral Health |
Subjects: | |
Online Access: | https://doi.org/10.1186/s12903-023-02881-8 |
_version_ | 1797853520804708352 |
---|---|
author | Han Bao Kejia Zhang Chenhao Yu Hu Li Dan Cao Huazhong Shu Luwei Liu Bin Yan |
author_facet | Han Bao Kejia Zhang Chenhao Yu Hu Li Dan Cao Huazhong Shu Luwei Liu Bin Yan |
author_sort | Han Bao |
collection | DOAJ |
description | Abstract Background The purpose of this study was to evaluate the accuracy of automatic cephalometric landmark localization and measurements using cephalometric analysis via artificial intelligence (AI) compared with computer-assisted manual analysis. Methods Reconstructed lateral cephalograms (RLCs) from cone-beam computed tomography (CBCT) in 85 patients were selected. Computer-assisted manual analysis (Dolphin Imaging 11.9) and AI automatic analysis (Planmeca Romexis 6.2) were used to locate 19 landmarks and obtain 23 measurements. Mean radial error (MRE) and successful detection rate (SDR) values were calculated to assess the accuracy of automatic landmark digitization. Paired t tests and Bland‒Altman plots were used to compare the differences and consistencies in cephalometric measurements between manual and automatic analysis programs. Results The MRE for 19 cephalometric landmarks was 2.07 ± 1.35 mm with the automatic program. The average SDR within 1 mm, 2 mm, 2.5 mm, 3 and 4 mm were 18.82%, 58.58%, 71.70%, 82.04% and 91.39%, respectively. Soft tissue landmarks (1.54 ± 0.85 mm) had the most consistency, while dental landmarks (2.37 ± 1.55 mm) had the most variation. In total, 15 out of 23 measurements were within the clinically acceptable level of accuracy, 2 mm or 2°. The rates of consistency within the 95% limits of agreement were all above 90% for all measurement parameters. Conclusion Automatic analysis software collects cephalometric measurements almost effectively enough to be acceptable in clinical work. Nevertheless, automatic cephalometry is not capable of completely replacing manual tracing. Additional manual supervision and adjustment for automatic programs can increase accuracy and efficiency. |
first_indexed | 2024-04-09T19:51:58Z |
format | Article |
id | doaj.art-72fbbb156bee4ee4a3c98241c0643a9e |
institution | Directory Open Access Journal |
issn | 1472-6831 |
language | English |
last_indexed | 2024-04-09T19:51:58Z |
publishDate | 2023-04-01 |
publisher | BMC |
record_format | Article |
series | BMC Oral Health |
spelling | doaj.art-72fbbb156bee4ee4a3c98241c0643a9e2023-04-03T05:42:45ZengBMCBMC Oral Health1472-68312023-04-0123111010.1186/s12903-023-02881-8Evaluating the accuracy of automated cephalometric analysis based on artificial intelligenceHan Bao0Kejia Zhang1Chenhao Yu2Hu Li3Dan Cao4Huazhong Shu5Luwei Liu6Bin Yan7Department of Orthodontics, The Affiliated Stomatological Hospital of Nanjing Medical UniversityDepartment of Orthodontics, The Affiliated Stomatological Hospital of Nanjing Medical UniversityJiangsu Province Key Laboratory of Oral Diseases, Nanjing Medical UniversityDepartment of Orthodontics, The Affiliated Stomatological Hospital of Nanjing Medical UniversityDepartment of Orthodontics, The Affiliated Stomatological Hospital of Nanjing Medical UniversityLaboratory of Image Science and Technology, Southeast UniversityDepartment of Orthodontics, The Affiliated Stomatological Hospital of Nanjing Medical UniversityDepartment of Orthodontics, The Affiliated Stomatological Hospital of Nanjing Medical UniversityAbstract Background The purpose of this study was to evaluate the accuracy of automatic cephalometric landmark localization and measurements using cephalometric analysis via artificial intelligence (AI) compared with computer-assisted manual analysis. Methods Reconstructed lateral cephalograms (RLCs) from cone-beam computed tomography (CBCT) in 85 patients were selected. Computer-assisted manual analysis (Dolphin Imaging 11.9) and AI automatic analysis (Planmeca Romexis 6.2) were used to locate 19 landmarks and obtain 23 measurements. Mean radial error (MRE) and successful detection rate (SDR) values were calculated to assess the accuracy of automatic landmark digitization. Paired t tests and Bland‒Altman plots were used to compare the differences and consistencies in cephalometric measurements between manual and automatic analysis programs. Results The MRE for 19 cephalometric landmarks was 2.07 ± 1.35 mm with the automatic program. The average SDR within 1 mm, 2 mm, 2.5 mm, 3 and 4 mm were 18.82%, 58.58%, 71.70%, 82.04% and 91.39%, respectively. Soft tissue landmarks (1.54 ± 0.85 mm) had the most consistency, while dental landmarks (2.37 ± 1.55 mm) had the most variation. In total, 15 out of 23 measurements were within the clinically acceptable level of accuracy, 2 mm or 2°. The rates of consistency within the 95% limits of agreement were all above 90% for all measurement parameters. Conclusion Automatic analysis software collects cephalometric measurements almost effectively enough to be acceptable in clinical work. Nevertheless, automatic cephalometry is not capable of completely replacing manual tracing. Additional manual supervision and adjustment for automatic programs can increase accuracy and efficiency.https://doi.org/10.1186/s12903-023-02881-8Cephalometric analysisAutomatic identificationCone-beam CTArtificial intelligent |
spellingShingle | Han Bao Kejia Zhang Chenhao Yu Hu Li Dan Cao Huazhong Shu Luwei Liu Bin Yan Evaluating the accuracy of automated cephalometric analysis based on artificial intelligence BMC Oral Health Cephalometric analysis Automatic identification Cone-beam CT Artificial intelligent |
title | Evaluating the accuracy of automated cephalometric analysis based on artificial intelligence |
title_full | Evaluating the accuracy of automated cephalometric analysis based on artificial intelligence |
title_fullStr | Evaluating the accuracy of automated cephalometric analysis based on artificial intelligence |
title_full_unstemmed | Evaluating the accuracy of automated cephalometric analysis based on artificial intelligence |
title_short | Evaluating the accuracy of automated cephalometric analysis based on artificial intelligence |
title_sort | evaluating the accuracy of automated cephalometric analysis based on artificial intelligence |
topic | Cephalometric analysis Automatic identification Cone-beam CT Artificial intelligent |
url | https://doi.org/10.1186/s12903-023-02881-8 |
work_keys_str_mv | AT hanbao evaluatingtheaccuracyofautomatedcephalometricanalysisbasedonartificialintelligence AT kejiazhang evaluatingtheaccuracyofautomatedcephalometricanalysisbasedonartificialintelligence AT chenhaoyu evaluatingtheaccuracyofautomatedcephalometricanalysisbasedonartificialintelligence AT huli evaluatingtheaccuracyofautomatedcephalometricanalysisbasedonartificialintelligence AT dancao evaluatingtheaccuracyofautomatedcephalometricanalysisbasedonartificialintelligence AT huazhongshu evaluatingtheaccuracyofautomatedcephalometricanalysisbasedonartificialintelligence AT luweiliu evaluatingtheaccuracyofautomatedcephalometricanalysisbasedonartificialintelligence AT binyan evaluatingtheaccuracyofautomatedcephalometricanalysisbasedonartificialintelligence |