Abstract
Tongue diagnosis is a useful process in traditional Chinese medicine to assess diseases non-invasively by visually inspecting the tongue and its various properties. In this study, we developed an automated tongue diagnosis system with a mobile app for the general public. The image-segmentation component extracts the tongue body image from an input photograph taken by a smartphone. The tongue-coating color classification component predicts the category of the coating color. The segmented image and diagnosis results are returned to the app and shown to the user. Experimental results show that Mask R-CNN is the optimal choice for tongue-image segmentation under various input image conditions based on the mean interaction over union value of 91 % and the Dice score of 95 % . ResNeXt outperformed other baseline tongue-coating color classification models. In addition, when the input image is adjusted with our color-correction modules in advance, the classification accuracy of ResNeXt101 is improved by approximately 12 % .
Original language | English |
---|---|
Pages (from-to) | 21259-21274 |
Number of pages | 16 |
Journal | Neural Computing and Applications |
Volume | 35 |
Issue number | 28 |
DOIs | |
State | Published - Oct 2023 |
Keywords
- Color correction
- Deep learning
- Image segmentation
- Tongue diagnosis
- Tongue-coating color classification
Fingerprint
Dive into the research topics of 'TongueMobile: automated tongue segmentation and diagnosis on smartphones'. Together they form a unique fingerprint.
View full fingerprint
Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver
Huang, Z. H., Huang, W. C., Wu, H. C., & Fang, W. C. (2023). TongueMobile: automated tongue segmentation and diagnosis on smartphones. Neural Computing and Applications, 35(28), 21259-21274. https://doi.org/10.1007/s00521-023-08902-5
Huang, Zih Hao ; Huang, Wei Cheng ; Wu, Hsien Chang et al. / TongueMobile : automated tongue segmentation and diagnosis on smartphones. In: Neural Computing and Applications. 2023 ; Vol. 35, No. 28. pp. 21259-21274.
@article{88b1d24a5aa245f596b59c3d11f449d7,
title = "TongueMobile: automated tongue segmentation and diagnosis on smartphones",
abstract = "Tongue diagnosis is a useful process in traditional Chinese medicine to assess diseases non-invasively by visually inspecting the tongue and its various properties. In this study, we developed an automated tongue diagnosis system with a mobile app for the general public. The image-segmentation component extracts the tongue body image from an input photograph taken by a smartphone. The tongue-coating color classification component predicts the category of the coating color. The segmented image and diagnosis results are returned to the app and shown to the user. Experimental results show that Mask R-CNN is the optimal choice for tongue-image segmentation under various input image conditions based on the mean interaction over union value of 91 % and the Dice score of 95 % . ResNeXt outperformed other baseline tongue-coating color classification models. In addition, when the input image is adjusted with our color-correction modules in advance, the classification accuracy of ResNeXt101 is improved by approximately 12 % .",
keywords = "Color correction, Deep learning, Image segmentation, Tongue diagnosis, Tongue-coating color classification",
author = "Huang, {Zih Hao} and Huang, {Wei Cheng} and Wu, {Hsien Chang} and Fang, {Wen Chieh}",
note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature.",
year = "2023",
month = oct,
doi = "10.1007/s00521-023-08902-5",
language = "英语",
volume = "35",
pages = "21259--21274",
journal = "Neural Computing and Applications",
issn = "0941-0643",
publisher = "Springer London",
number = "28",
}
Huang, ZH, Huang, WC, Wu, HC & Fang, WC 2023, 'TongueMobile: automated tongue segmentation and diagnosis on smartphones', Neural Computing and Applications, vol. 35, no. 28, pp. 21259-21274. https://doi.org/10.1007/s00521-023-08902-5
TongueMobile: automated tongue segmentation and diagnosis on smartphones. / Huang, Zih Hao; Huang, Wei Cheng; Wu, Hsien Chang et al.
In: Neural Computing and Applications, Vol. 35, No. 28, 10.2023, p. 21259-21274.
Research output: Contribution to journal › Article › peer-review
TY - JOUR
T1 - TongueMobile
T2 - automated tongue segmentation and diagnosis on smartphones
AU - Huang, Zih Hao
AU - Huang, Wei Cheng
AU - Wu, Hsien Chang
AU - Fang, Wen Chieh
N1 - Publisher Copyright:© 2023, The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature.
PY - 2023/10
Y1 - 2023/10
N2 - Tongue diagnosis is a useful process in traditional Chinese medicine to assess diseases non-invasively by visually inspecting the tongue and its various properties. In this study, we developed an automated tongue diagnosis system with a mobile app for the general public. The image-segmentation component extracts the tongue body image from an input photograph taken by a smartphone. The tongue-coating color classification component predicts the category of the coating color. The segmented image and diagnosis results are returned to the app and shown to the user. Experimental results show that Mask R-CNN is the optimal choice for tongue-image segmentation under various input image conditions based on the mean interaction over union value of 91 % and the Dice score of 95 % . ResNeXt outperformed other baseline tongue-coating color classification models. In addition, when the input image is adjusted with our color-correction modules in advance, the classification accuracy of ResNeXt101 is improved by approximately 12 % .
AB - Tongue diagnosis is a useful process in traditional Chinese medicine to assess diseases non-invasively by visually inspecting the tongue and its various properties. In this study, we developed an automated tongue diagnosis system with a mobile app for the general public. The image-segmentation component extracts the tongue body image from an input photograph taken by a smartphone. The tongue-coating color classification component predicts the category of the coating color. The segmented image and diagnosis results are returned to the app and shown to the user. Experimental results show that Mask R-CNN is the optimal choice for tongue-image segmentation under various input image conditions based on the mean interaction over union value of 91 % and the Dice score of 95 % . ResNeXt outperformed other baseline tongue-coating color classification models. In addition, when the input image is adjusted with our color-correction modules in advance, the classification accuracy of ResNeXt101 is improved by approximately 12 % .
KW - Color correction
KW - Deep learning
KW - Image segmentation
KW - Tongue diagnosis
KW - Tongue-coating color classification
UR - http://www.scopus.com/inward/record.url?scp=85166781752&partnerID=8YFLogxK
U2 - 10.1007/s00521-023-08902-5
DO - 10.1007/s00521-023-08902-5
M3 - 文章
AN - SCOPUS:85166781752
SN - 0941-0643
VL - 35
SP - 21259
EP - 21274
JO - Neural Computing and Applications
JF - Neural Computing and Applications
IS - 28
ER -
Huang ZH, Huang WC, Wu HC, Fang WC. TongueMobile: automated tongue segmentation and diagnosis on smartphones. Neural Computing and Applications. 2023 Oct;35(28):21259-21274. doi: 10.1007/s00521-023-08902-5