Tongue feature dataset construction and real-time detection.
<h4>Background</h4>Tongue diagnosis in traditional Chinese medicine (TCM) provides clinically important, objective evidence from direct observation of specific features that assist with diagnosis. However, the current interpretation of tongue features requires a significant amount of man...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2024-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0296070&type=printable |
| _version_ | 1797263507021889536 |
|---|---|
| author | Wen-Hsien Chang Chih-Chieh Chen Han-Kuei Wu Po-Chi Hsu Lun-Chien Lo Hsueh-Ting Chu Hen-Hong Chang |
| author_facet | Wen-Hsien Chang Chih-Chieh Chen Han-Kuei Wu Po-Chi Hsu Lun-Chien Lo Hsueh-Ting Chu Hen-Hong Chang |
| author_sort | Wen-Hsien Chang |
| collection | DOAJ |
| description | <h4>Background</h4>Tongue diagnosis in traditional Chinese medicine (TCM) provides clinically important, objective evidence from direct observation of specific features that assist with diagnosis. However, the current interpretation of tongue features requires a significant amount of manpower and time. TCM physicians may have different interpretations of features displayed by the same tongue. An automated interpretation system that interprets tongue features would expedite the interpretation process and yield more consistent results.<h4>Materials and methods</h4>This study applied deep learning visualization to tongue diagnosis. After collecting tongue images and corresponding interpretation reports by TCM physicians in a single teaching hospital, various tongue features such as fissures, tooth marks, and different types of coatings were annotated manually with rectangles. These annotated data and images were used to train a deep learning object detection model. Upon completion of training, the position of each tongue feature was dynamically marked.<h4>Results</h4>A large high-quality manually annotated tongue feature dataset was constructed and analyzed. A detection model was trained with average precision (AP) 47.67%, 58.94%, 71.25% and 59.78% for fissures, tooth marks, thick and yellow coatings, respectively. At over 40 frames per second on a NVIDIA GeForce GTX 1060, the model was capable of detecting tongue features from any viewpoint in real time.<h4>Conclusions/significance</h4>This study constructed a tongue feature dataset and trained a deep learning object detection model to locate tongue features in real time. The model provided interpretability and intuitiveness that are often lacking in general neural network models and implies good feasibility for clinical application. |
| first_indexed | 2024-04-25T00:14:06Z |
| format | Article |
| id | doaj.art-0815fbed4439466c8483f5c502b79b64 |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-04-25T00:14:06Z |
| publishDate | 2024-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-0815fbed4439466c8483f5c502b79b642024-03-13T05:31:26ZengPublic Library of Science (PLoS)PLoS ONE1932-62032024-01-01193e029607010.1371/journal.pone.0296070Tongue feature dataset construction and real-time detection.Wen-Hsien ChangChih-Chieh ChenHan-Kuei WuPo-Chi HsuLun-Chien LoHsueh-Ting ChuHen-Hong Chang<h4>Background</h4>Tongue diagnosis in traditional Chinese medicine (TCM) provides clinically important, objective evidence from direct observation of specific features that assist with diagnosis. However, the current interpretation of tongue features requires a significant amount of manpower and time. TCM physicians may have different interpretations of features displayed by the same tongue. An automated interpretation system that interprets tongue features would expedite the interpretation process and yield more consistent results.<h4>Materials and methods</h4>This study applied deep learning visualization to tongue diagnosis. After collecting tongue images and corresponding interpretation reports by TCM physicians in a single teaching hospital, various tongue features such as fissures, tooth marks, and different types of coatings were annotated manually with rectangles. These annotated data and images were used to train a deep learning object detection model. Upon completion of training, the position of each tongue feature was dynamically marked.<h4>Results</h4>A large high-quality manually annotated tongue feature dataset was constructed and analyzed. A detection model was trained with average precision (AP) 47.67%, 58.94%, 71.25% and 59.78% for fissures, tooth marks, thick and yellow coatings, respectively. At over 40 frames per second on a NVIDIA GeForce GTX 1060, the model was capable of detecting tongue features from any viewpoint in real time.<h4>Conclusions/significance</h4>This study constructed a tongue feature dataset and trained a deep learning object detection model to locate tongue features in real time. The model provided interpretability and intuitiveness that are often lacking in general neural network models and implies good feasibility for clinical application.https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0296070&type=printable |
| spellingShingle | Wen-Hsien Chang Chih-Chieh Chen Han-Kuei Wu Po-Chi Hsu Lun-Chien Lo Hsueh-Ting Chu Hen-Hong Chang Tongue feature dataset construction and real-time detection. PLoS ONE |
| title | Tongue feature dataset construction and real-time detection. |
| title_full | Tongue feature dataset construction and real-time detection. |
| title_fullStr | Tongue feature dataset construction and real-time detection. |
| title_full_unstemmed | Tongue feature dataset construction and real-time detection. |
| title_short | Tongue feature dataset construction and real-time detection. |
| title_sort | tongue feature dataset construction and real time detection |
| url | https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0296070&type=printable |
| work_keys_str_mv | AT wenhsienchang tonguefeaturedatasetconstructionandrealtimedetection AT chihchiehchen tonguefeaturedatasetconstructionandrealtimedetection AT hankueiwu tonguefeaturedatasetconstructionandrealtimedetection AT pochihsu tonguefeaturedatasetconstructionandrealtimedetection AT lunchienlo tonguefeaturedatasetconstructionandrealtimedetection AT hsuehtingchu tonguefeaturedatasetconstructionandrealtimedetection AT henhongchang tonguefeaturedatasetconstructionandrealtimedetection |