Deep-Learning-Based Detection of Cranio-Spinal Differences between Skeletal Classification Using Cephalometric Radiography
The aim of this study was to reveal cranio-spinal differences between skeletal classification using convolutional neural networks (CNNs). Transverse and longitudinal cephalometric images of 832 patients were used for training and testing of CNNs (365 males and 467 females). Labeling was performed su...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-03-01
|
Series: | Diagnostics |
Subjects: | |
Online Access: | https://www.mdpi.com/2075-4418/11/4/591 |
id |
doaj-5273b33d547b40f6bfac1b9adb8acfc5 |
---|---|
record_format |
Article |
spelling |
doaj-5273b33d547b40f6bfac1b9adb8acfc52021-03-26T00:05:32ZengMDPI AGDiagnostics2075-44182021-03-011159159110.3390/diagnostics11040591Deep-Learning-Based Detection of Cranio-Spinal Differences between Skeletal Classification Using Cephalometric RadiographySeung Hyun Jeong0Jong Pil Yun1Han-Gyeol Yeom2Hwi Kang Kim3Bong Chul Kim4Safety System Research Group, Korea Institute of Industrial Technology (KITECH), Gyeongsan 38408, KoreaSafety System Research Group, Korea Institute of Industrial Technology (KITECH), Gyeongsan 38408, KoreaDepartment of Oral and Maxillofacial Radiology, Daejeon Dental Hospital, Wonkwang University College of Dentistry, Daejeon 35233, KoreaDepartment of Oral and Maxillofacial Surgery, Daejeon Dental Hospital, Wonkwang University College of Dentistry, Daejeon 35233, KoreaDepartment of Oral and Maxillofacial Surgery, Daejeon Dental Hospital, Wonkwang University College of Dentistry, Daejeon 35233, KoreaThe aim of this study was to reveal cranio-spinal differences between skeletal classification using convolutional neural networks (CNNs). Transverse and longitudinal cephalometric images of 832 patients were used for training and testing of CNNs (365 males and 467 females). Labeling was performed such that the jawbone was sufficiently masked, while the parts other than the jawbone were minimally masked. DenseNet was used as the feature extractor. Five random sampling crossvalidations were performed for two datasets. The average and maximum accuracy of the five crossvalidations were 90.43% and 92.54% for test 1 (evaluation of the entire posterior–anterior (PA) and lateral cephalometric images) and 88.17% and 88.70% for test 2 (evaluation of the PA and lateral cephalometric images obscuring the mandible). In this study, we found that even when jawbones of class I (normal mandible), class II (retrognathism), and class III (prognathism) are masked, their identification is possible through deep learning applied only in the cranio-spinal area. This suggests that cranio-spinal differences between each class exist.https://www.mdpi.com/2075-4418/11/4/591machine learningartificial intelligencemalocclusiondiagnostic imaging |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Seung Hyun Jeong Jong Pil Yun Han-Gyeol Yeom Hwi Kang Kim Bong Chul Kim |
spellingShingle |
Seung Hyun Jeong Jong Pil Yun Han-Gyeol Yeom Hwi Kang Kim Bong Chul Kim Deep-Learning-Based Detection of Cranio-Spinal Differences between Skeletal Classification Using Cephalometric Radiography Diagnostics machine learning artificial intelligence malocclusion diagnostic imaging |
author_facet |
Seung Hyun Jeong Jong Pil Yun Han-Gyeol Yeom Hwi Kang Kim Bong Chul Kim |
author_sort |
Seung Hyun Jeong |
title |
Deep-Learning-Based Detection of Cranio-Spinal Differences between Skeletal Classification Using Cephalometric Radiography |
title_short |
Deep-Learning-Based Detection of Cranio-Spinal Differences between Skeletal Classification Using Cephalometric Radiography |
title_full |
Deep-Learning-Based Detection of Cranio-Spinal Differences between Skeletal Classification Using Cephalometric Radiography |
title_fullStr |
Deep-Learning-Based Detection of Cranio-Spinal Differences between Skeletal Classification Using Cephalometric Radiography |
title_full_unstemmed |
Deep-Learning-Based Detection of Cranio-Spinal Differences between Skeletal Classification Using Cephalometric Radiography |
title_sort |
deep-learning-based detection of cranio-spinal differences between skeletal classification using cephalometric radiography |
publisher |
MDPI AG |
series |
Diagnostics |
issn |
2075-4418 |
publishDate |
2021-03-01 |
description |
The aim of this study was to reveal cranio-spinal differences between skeletal classification using convolutional neural networks (CNNs). Transverse and longitudinal cephalometric images of 832 patients were used for training and testing of CNNs (365 males and 467 females). Labeling was performed such that the jawbone was sufficiently masked, while the parts other than the jawbone were minimally masked. DenseNet was used as the feature extractor. Five random sampling crossvalidations were performed for two datasets. The average and maximum accuracy of the five crossvalidations were 90.43% and 92.54% for test 1 (evaluation of the entire posterior–anterior (PA) and lateral cephalometric images) and 88.17% and 88.70% for test 2 (evaluation of the PA and lateral cephalometric images obscuring the mandible). In this study, we found that even when jawbones of class I (normal mandible), class II (retrognathism), and class III (prognathism) are masked, their identification is possible through deep learning applied only in the cranio-spinal area. This suggests that cranio-spinal differences between each class exist. |
topic |
machine learning artificial intelligence malocclusion diagnostic imaging |
url |
https://www.mdpi.com/2075-4418/11/4/591 |
work_keys_str_mv |
AT seunghyunjeong deeplearningbaseddetectionofcraniospinaldifferencesbetweenskeletalclassificationusingcephalometricradiography AT jongpilyun deeplearningbaseddetectionofcraniospinaldifferencesbetweenskeletalclassificationusingcephalometricradiography AT hangyeolyeom deeplearningbaseddetectionofcraniospinaldifferencesbetweenskeletalclassificationusingcephalometricradiography AT hwikangkim deeplearningbaseddetectionofcraniospinaldifferencesbetweenskeletalclassificationusingcephalometricradiography AT bongchulkim deeplearningbaseddetectionofcraniospinaldifferencesbetweenskeletalclassificationusingcephalometricradiography |
_version_ |
1724203090389762048 |