PXseg: automatic tooth segmentation, numbering and abnormal morphology detection based on CBCT and panoramic radiographs

Abstract Objective PXseg, a novel approach for tooth segmentation, numbering and abnormal morphology detection in panoramic X-ray (PX), was designed and promoted through optimizing annotation and applying pre-training. Methods Derived from multicenter, ctPXs generated from cone beam computed tomogra...

Full description

Saved in:
Bibliographic Details
Main Authors: Raokaijuan Wang, Fangyuan Cheng, Guangsheng Dai, Jiayu Zhang, Chengmin Fan, Jinghong Yu, Juan Li, Fulin Jiang
Format: Article
Language:English
Published: BMC 2025-07-01
Series:BMC Oral Health
Subjects:
Online Access:https://doi.org/10.1186/s12903-025-06356-w
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849389078404923392
author Raokaijuan Wang
Fangyuan Cheng
Guangsheng Dai
Jiayu Zhang
Chengmin Fan
Jinghong Yu
Juan Li
Fulin Jiang
author_facet Raokaijuan Wang
Fangyuan Cheng
Guangsheng Dai
Jiayu Zhang
Chengmin Fan
Jinghong Yu
Juan Li
Fulin Jiang
author_sort Raokaijuan Wang
collection DOAJ
description Abstract Objective PXseg, a novel approach for tooth segmentation, numbering and abnormal morphology detection in panoramic X-ray (PX), was designed and promoted through optimizing annotation and applying pre-training. Methods Derived from multicenter, ctPXs generated from cone beam computed tomography (CBCT) with accurate 3D labels were utilized for pre-training, while conventional PXs (cPXs) with 2D labels were input for training. Visual and statistical analyses were conducted using the internal dataset to assess segmentation and numbering performances of PXseg and compared with the model without ctPX pre-training, while the accuracy of PXseg detecting abnormal teeth was evaluated using the external dataset consisting of cPXs with complex dental diseases. Besides, a diagnostic testing was performed to contrast diagnostic efficiency with and without PXseg’s assistance. Results The DSC and F1-score of PXseg in tooth segmentation reached 0.882 and 0.902, which increased by 4.6% and 4.0% compared to the model without pre-training. For tooth numbering, the F1-score of PXseg reached 0.943 and increased by 2.2%. Based on the promotion in segmentation, the accuracy of abnormal tooth morphology detection exceeded 0.957 and was 4.3% higher. A website was constructed to assist in PX interpretation, and the diagnostic efficiency was greatly enhanced with the assistance of PXseg. Conclusions The application of accurate labels in ctPX increased the pre-training weight of PXseg and improved the training effect, achieving promotions in tooth segmentation, numbering and abnormal morphology detection. Rapid and accurate results provided by PXseg streamlined the workflow of PX diagnosis, possessing significant clinical application prospect.
format Article
id doaj-art-c54e0a46e4d14ee29e152b7c3ec96360
institution Kabale University
issn 1472-6831
language English
publishDate 2025-07-01
publisher BMC
record_format Article
series BMC Oral Health
spelling doaj-art-c54e0a46e4d14ee29e152b7c3ec963602025-08-20T03:42:03ZengBMCBMC Oral Health1472-68312025-07-0125111210.1186/s12903-025-06356-wPXseg: automatic tooth segmentation, numbering and abnormal morphology detection based on CBCT and panoramic radiographsRaokaijuan Wang0Fangyuan Cheng1Guangsheng Dai2Jiayu Zhang3Chengmin Fan4Jinghong Yu5Juan Li6Fulin Jiang7Department of Orthodontics, West China School of Stomatology, Sichuan UniversityChengdu Boltzmann Intelligence Technology Co., LtdChengdu Boltzmann Intelligence Technology Co., LtdChengdu Boltzmann Intelligence Technology Co., LtdChengdu Boltzmann Intelligence Technology Co., LtdChongqing University Three Gorges HospitalDepartment of Orthodontics, West China School of Stomatology, Sichuan UniversityDepartment of Orthodontics, Chongqing University Three Gorges HospitalAbstract Objective PXseg, a novel approach for tooth segmentation, numbering and abnormal morphology detection in panoramic X-ray (PX), was designed and promoted through optimizing annotation and applying pre-training. Methods Derived from multicenter, ctPXs generated from cone beam computed tomography (CBCT) with accurate 3D labels were utilized for pre-training, while conventional PXs (cPXs) with 2D labels were input for training. Visual and statistical analyses were conducted using the internal dataset to assess segmentation and numbering performances of PXseg and compared with the model without ctPX pre-training, while the accuracy of PXseg detecting abnormal teeth was evaluated using the external dataset consisting of cPXs with complex dental diseases. Besides, a diagnostic testing was performed to contrast diagnostic efficiency with and without PXseg’s assistance. Results The DSC and F1-score of PXseg in tooth segmentation reached 0.882 and 0.902, which increased by 4.6% and 4.0% compared to the model without pre-training. For tooth numbering, the F1-score of PXseg reached 0.943 and increased by 2.2%. Based on the promotion in segmentation, the accuracy of abnormal tooth morphology detection exceeded 0.957 and was 4.3% higher. A website was constructed to assist in PX interpretation, and the diagnostic efficiency was greatly enhanced with the assistance of PXseg. Conclusions The application of accurate labels in ctPX increased the pre-training weight of PXseg and improved the training effect, achieving promotions in tooth segmentation, numbering and abnormal morphology detection. Rapid and accurate results provided by PXseg streamlined the workflow of PX diagnosis, possessing significant clinical application prospect.https://doi.org/10.1186/s12903-025-06356-wTooth segmentationPanoramic X-rayCone beam computed tomographyConvolutional neural network
spellingShingle Raokaijuan Wang
Fangyuan Cheng
Guangsheng Dai
Jiayu Zhang
Chengmin Fan
Jinghong Yu
Juan Li
Fulin Jiang
PXseg: automatic tooth segmentation, numbering and abnormal morphology detection based on CBCT and panoramic radiographs
BMC Oral Health
Tooth segmentation
Panoramic X-ray
Cone beam computed tomography
Convolutional neural network
title PXseg: automatic tooth segmentation, numbering and abnormal morphology detection based on CBCT and panoramic radiographs
title_full PXseg: automatic tooth segmentation, numbering and abnormal morphology detection based on CBCT and panoramic radiographs
title_fullStr PXseg: automatic tooth segmentation, numbering and abnormal morphology detection based on CBCT and panoramic radiographs
title_full_unstemmed PXseg: automatic tooth segmentation, numbering and abnormal morphology detection based on CBCT and panoramic radiographs
title_short PXseg: automatic tooth segmentation, numbering and abnormal morphology detection based on CBCT and panoramic radiographs
title_sort pxseg automatic tooth segmentation numbering and abnormal morphology detection based on cbct and panoramic radiographs
topic Tooth segmentation
Panoramic X-ray
Cone beam computed tomography
Convolutional neural network
url https://doi.org/10.1186/s12903-025-06356-w
work_keys_str_mv AT raokaijuanwang pxsegautomatictoothsegmentationnumberingandabnormalmorphologydetectionbasedoncbctandpanoramicradiographs
AT fangyuancheng pxsegautomatictoothsegmentationnumberingandabnormalmorphologydetectionbasedoncbctandpanoramicradiographs
AT guangshengdai pxsegautomatictoothsegmentationnumberingandabnormalmorphologydetectionbasedoncbctandpanoramicradiographs
AT jiayuzhang pxsegautomatictoothsegmentationnumberingandabnormalmorphologydetectionbasedoncbctandpanoramicradiographs
AT chengminfan pxsegautomatictoothsegmentationnumberingandabnormalmorphologydetectionbasedoncbctandpanoramicradiographs
AT jinghongyu pxsegautomatictoothsegmentationnumberingandabnormalmorphologydetectionbasedoncbctandpanoramicradiographs
AT juanli pxsegautomatictoothsegmentationnumberingandabnormalmorphologydetectionbasedoncbctandpanoramicradiographs
AT fulinjiang pxsegautomatictoothsegmentationnumberingandabnormalmorphologydetectionbasedoncbctandpanoramicradiographs