Pith. sign in

REVIEW 4 major objections 5 minor 53 references

Dr. Tongue: Sign-Oriented Multi-label Detection for Remote Tongue Diagnosis

T0 review · 4 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper claims that a sign-oriented multi-branch network, combined with adaptive tongue detection, segmentation, and upright orientation, recognizes eight tongue attributes from uncontrolled telehealth images with near-practitioner…

desk verdict The TongueDx dataset is a genuinely useful contribution, but the headline F1 gain rests on unvalidated labels and a mixed evaluation protocol, so treat the quantitative claim as not established. read the letter →

arxiv 2501.03053 v2 pith:NRZEHYHG submitted 2025-01-06 eess.IV cs.CV

Yiliang Chen , Steven SC Ho , Cheng Xu , Yao Jie Xie , Wing-Fai Yeung , Shengfeng He , Jing Qin This is my paper · ORCID
classification eess.IVcs.CV
keywords tonguediagnosismulti-labelclassificationtelemedicinedeeplearningimagedatasetTraditionalChineseMedicinesign-orientednetworksegmentation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper claims that tongue diagnosis, traditionally done by looking at the tongue surface, can be automated for telehealth images that are low-resolution and unevenly lit. It proposes a pipeline that first locates, segments, and rotates each tongue into a standard upright view, then runs a Sign-oriented Network (SignNet) that separates the whole tongue, tongue body, and tongue edge into branches aligned with how practitioners read color, shape, region, and fur signs. The authors report that SignNet raises average accuracy by about 1% and average F1-score by about 4% over ResNet50, and that it outperforms laypersons and approaches registered practitioners on eight binary tongue attributes. A new public dataset of 5109 smartphone and laptop tongue images from 4650 subjects, labeled for eight attributes plus demographics, is introduced to support remote diagnosis research. If right, the framework would give telehealth systems a way to extract clinically meaningful tongue signs from ordinary consumer-camera images.

What carries the argument

The central machinery is SignNet, a three-branch network that mirrors the paper's attribute-sign relationship table: the whole-tongue branch predicts color category and fur presence, the body branch handles body-region attributes such as crack, and the edge branch handles edge attributes such as tooth marks, with self-attention fusing color and fur features into each regional branch. Before this, the Adaptive Tongue Feature Extraction (ATFE) module standardizes input by detecting the tongue with a lightweight YOLOv5-MobileNetV3 network, segmenting it with MobileSAM, and rotating it upright with a contour-based tongue-tip and top alignment algorithm. A tongue body and edge separation algorithm with an adaptive edge width generates the regional inputs. The loss combines weighted binary cross-entropy to counter class imbalance.

What would settle it

Have an independent panel of practitioners re-annotate a random sample of the test-set images and measure inter-annotator agreement against the original labels; if agreement on Pale or Ecchymosis is at or near chance (for example, kappa below 0.4), the claimed improvements over baseline and the comparison with practitioners would be an artifact of label noise rather than a real signal.

Watch

Extended reading notes

Core claim

The paper's central claim is that a sign-oriented, multi-branch network, fed by an adaptive tongue extraction module, can recognize eight tongue attributes (Pale, TipSideRed, RedSpot, Ecchymosis, Crack, ToothMark, FurThick, FurYellow) in uncontrolled telehealth images better than generic deep backbones. The load-bearing numbers are that ResNet50+ATFE gains about 2% average F1 over plain ResNet50, and SignNet adds a further improvement of about 1% average accuracy and about 4% average F1 over ResNet50. The design explicitly encodes the practitioner's sign relationships: color cues are reorganized into four simplified color categories, fur is reduced to presence or absence, and edge-specific versus body-specific attributes are processed in separate branches fused by self-attention. The authors also position the TongueDx dataset as a uniquely telehealth-oriented public resource, since existing tongue datasets were captured with standardized devices or are no longer available.

Load-bearing premise

The load-bearing premise is that the three practitioners' labels on single static smartphone images are accurate and consistent enough across all eight attributes, particularly Pale and Ecchymosis, for the reported accuracy and F1 gaps to be meaningful.

Editorial extensions

If this is right

  • Remote tongue screening becomes feasible with ordinary smartphone or laptop cameras, not just standardized tongue-imaging devices.
  • The TongueDx dataset, if released as promised, gives the community a common benchmark for multi-label tongue attribute recognition under telehealth conditions.
  • The ATFE standardization (detect-segment-orient) transfers a benefit of about a 2% average F1 gain to a generic backbone, suggesting preprocessing is as important as the classifier.
  • SignNet's color, fur, body, and edge decomposition provides a template for embedding clinical sign knowledge into multi-label medical image classifiers.
  • Performance approaching registered practitioners suggests automated pre-screening could support, not replace, remote tongue assessment.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Because no inter-annotator agreement is reported, the reported margins over practitioners could be as much a measure of label noise as of model skill; an independent re-annotation study would settle this.
  • The color-branch simplification into four color categories could be seen as a learned soft color-calibration, and a testable extension is to replace it with explicit color constancy preprocessing on paired controlled and uncontrolled images.
  • The upright-orientation algorithm explicitly assumes the tongue appears in a normal, non-upside-down orientation, so extending to arbitrary pose would require a more general keypoint or rotation regressor.
  • If the public release includes demographic labels, the dataset could support age- and gender-stratified analysis of tongue signs, which the current experiments do not touch.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper introduces TongueDx, a telehealth-oriented tongue image dataset with 5109 images from 4650 subjects and eight binary attribute labels, and proposes a Sign-Oriented multi-label attribute detection framework. The framework's Adaptive Tongue Feature Extraction (ATFE) module detects the tongue with YOLOv5-MobileNetV3 (trained on GroundingDINO boxes), segments it with MobileSAM, orients it upright, and separates it into body and edge regions. The SignNet classifier then uses whole-tongue, body, and edge branches plus explicit color and fur branches with self-attention fusion to predict the eight attributes. In experiments, SignNet is reported to improve average accuracy by about 1% and average F1 by about 4% over ResNet50, with ablations attributing gains to the ATFE module and the color/fur branches; the authors also compare the model against registered practitioners and laypersons.

Significance. If the label foundation holds, TongueDx would be a genuinely useful public resource for remote tongue diagnosis, and the paper's explicit modeling of attribute-sign-region relationships is a sensible architectural idea that could transfer to other structured medical image tasks. The ablation study is well designed at the component level: it separately evaluates detection, segmentation, upright orientation, and the color and fur branches, and the reported numbers are internally consistent with the proposed mechanism. The manuscript also includes a candid supplementary discussion of stability and lighting limitations. However, the headline comparisons against ResNet50 and against practitioners rest on unvalidated ground-truth labels and on an incompletely specified evaluation protocol, so the strength of the contribution is not yet established.

major comments (4)
  1. [The TongueDx Dataset; Quantitative Results] The manuscript does not report any inter-annotator agreement for the eight binary attribute labels, and it states that final annotations were completed by three professional practitioners through direct observation in a clinical setting, while test images were captured at home under uncontrolled lighting. This creates an unquantified train/test label mismatch: a label may encode what was observed live even when the image alone does not support it. The concern is load-bearing because the average-F1 advantage of SignNet over ResNet50 in Table 3 is concentrated in Pale (32.68 to 46.00) and Ecchymosis (24.72 to 38.02), the two attributes the authors themselves identify in the Supplementary Material as most sensitive to lighting. Please report per-attribute inter-annotator agreement (for example Cohen's or Fleiss' kappa) on a representative sample, describe the adjudication process, and clarify whether the three practitioners used in the human evaluation are the same individuals who created the training labels. Without this evidence, the model-model and model-human comparisons rest on unverified ground truth.
  2. [Experiments; Supplementary: Five-fold Cross-validation Setting] Table 3 is captioned 'five-fold cross-validation,' but the Supplementary describes a five-fold cross-validation combined with a separate hold-out test set of 895 subjects, with the hold-out test set reserved for human comparison. These are different protocols, and the paper does not state clearly whether the numbers in Tables 3 and 4 come from the five training folds, from the hold-out test set, or from a combination of both. Please specify which split produced every reported metric, confirm whether the hold-out test set is excluded from the five folds, and describe exactly how model selection was performed (the Supplementary mentions selecting on validation F1). This is essential for interpreting the headline 1%/4% improvements and for reproducing the human-evaluation results.
  3. [Tongue Detection and Segmentation] The localization evaluation is circular: GroundingDINO's outputs are used as the ground-truth bounding boxes for training YOLOv5-MobileNetV3, and the reported 'over 99% mAP' measures agreement with that same teacher model, not with true tongue location. This does not validate the ATFE claim of 'high-quality annotations' or robustness. Please evaluate the detection network on a human-verified subset of bounding boxes, or at least report the agreement between GroundingDINO and human-annotated boxes, so that the localization step is independently validated.
  4. [Quantitative Results] The claim that SignNet 'approaches the performance of practitioners' is based on single averaged accuracy/F1 rows for practitioners, with no variance and no statistical test, and no significance tests are reported for any model-model difference either. Given the large standard deviations in Pale and Ecchymosis in Table 3, please provide paired or matched comparisons (for example per-fold paired tests or bootstrap confidence intervals) for SignNet versus ResNet50 and for SignNet versus practitioner performance on the same test images. Without such tests, the strength of the 4% F1 claim and the practitioner-level claim cannot be assessed.
minor comments (5)
  1. [Loss Function] Equation (8) is malformed as printed: it reads alpha_j = median / F_j, but the surrounding text says m is the median frequency; it should read alpha_j = m / F_j, with m defined in the equation rather than only in prose.
  2. [Supplementary Table 5; main text] Please standardize attribute names across tables, figures, and text: for example, 'Furrthick' in Supplementary Table 5 and 'Tipsidered' in Table 5 should be consistent with 'FurThick' and 'TipSideRed' used elsewhere.
  3. [Figure 5] Figure 5 is hard to read because several labels are garbled or truncated (for example the TIUO/TEBRS inset and the mathematical symbols in the branch labels); please provide a higher-resolution version with clean and complete notation.
  4. [Algorithm 1] The contour filtering condition in Algorithm 1 is underspecified: step 3 says 'filter points where angle between points > alpha' without defining how that angle is computed or how the threshold interacts with the later smoothing steps; please provide a precise geometric definition.
  5. [Supplementary: ROC Curve Analysis] The ROC discussion states that SignNet and ResNet50+ATFE are 'fairly comparable to the baseline ResNet,' which is in tension with the claimed 4% average F1 improvement in Table 3; please clarify whether the ROC/AUC comparison uses a different protocol or subset, or reconcile the two statements.

Circularity Check

1 steps flagged · score 4.0 of 10

The central attribute-recognition claim is independent, but the reported >99% detection mAP is teacher-student agreement by construction: YOLOv5 is trained on GroundingDINO outputs that are also called the ground truth for the mAP.

  1. fitted input called prediction [The TongueDx Dataset, second paragraph; Methodology, 'Adaptive Tongue Feature Extraction Module', 'Tongue Detection and Segmentation']
    "we utilized GroundingDINO to generate precise ground truth bounding boxes for each tongue image in our dataset ... we used GroundingDINO’s outputs as high-quality training data for a more lightweight network ... Given the simplicity of this single-object localization task, we can easily achieve over 99% mAP performance."

    YOLOv5 is trained on bounding boxes whose 'ground truth' is defined as the highest-confidence GroundingDINO output for the prompt 'tongue'; no independent human annotation of boxes is reported. The mAP is then reported against these same GroundingDINO-derived boxes. The metric therefore measures student-teacher agreement: how well YOLO reproduces the teacher's labels on the same dataset that supplied its training targets. This is forced by construction, since any sufficiently expressive detector trained on those labels will score highly against them. The paper presents the >99% mAP as evidence that the ATFE detection step is accurate, but it is a self-consistency measure, not an externally validated detection accuracy.

full rationale

The paper's central contribution — SignNet's multi-label attribute recognition with about 4% higher average F1 than ResNet50 — is not circular: the eight attribute labels are provided by three practitioners, the model is trained and evaluated on held-out subjects in five-fold cross-validation, and the comparison against ResNet50, ViT, TransFG, laypersons, and practitioners is an external empirical benchmark. No load-bearing self-citation appears: the same-author references (Yuan et al. 2023 as a reproduced baseline; Ho et al. 2024 in future work) are not used to justify a derivation. The one self-referential element is the tongue-detection evaluation: GroundingDINO's outputs are used both to train YOLOv5 and as the 'ground truth' boxes against which the >99% mAP is reported, so that number measures teacher-student agreement rather than absolute localization accuracy. This is a supporting component rather than the central claim, and the supplementary materials themselves deprioritize it ('almost any type of network can easily achieve 99% mAP'), so I score it 4 rather than higher. Missing inter-annotator agreement and the clinical-setting versus static-image label mismatch are real validity threats, but they are not circularity and do not enter the score.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central performance claim rests on the quality of practitioner labels in a single dataset, several hand-set hyperparameters, and the expert-defined attribute-sign table. No new physical entities, forces, or particles are introduced.

free parameters (4)
  • Loss weights w_color and w_fur = w_color=1, w_fur=0.6
    Empirically set in the supplementary Implementation Details; these weights balance the color, fur, and attribute losses in Equation 6 and affect the reported performance.
  • Tongue edge width ratio r = 0.191
    Default value in Algorithm 2 for separating body and edge masks; chosen by hand and no sensitivity analysis is shown.
  • Upright orientation angle threshold alpha = 60 degrees
    Chosen from 50 to 70 degrees in Algorithm 1; the authors report similar results across that range.
  • Class frequency weights alpha_j = Computed from training fold frequencies via Eq. 8
    Used to reweight the binary attribute losses; this is data-derived rebalancing rather than a hand-tuned constant, but it is a modeling choice that affects the final F1 trade-off.
assumptions (4)
  • domain assumption The eight selected tongue attributes are visually identifiable from static 2D smartphone images and are meaningful diagnostic signs
    Underlies both dataset construction and all experiments; no external clinical validation is provided to show that these attributes can be reliably read from static telehealth photos.
  • domain assumption Practitioner annotations are accurate ground truth for all eight attributes
    Labels were created by three professional practitioners, but no inter-annotator agreement is reported, and the same practitioner pool may have been used for the human evaluation.
  • domain assumption The upright orientation algorithm assumes the tongue contours appear as upper and lower curves, not horizontal or upside-down
    Acknowledged in Algorithm 1; failure on extreme tongue poses is an explicit limitation of the preprocessing pipeline.
  • ad hoc to paper The attribute-sign relationships in Table 2 correctly describe where each attribute appears and its relationship to color, shape, region, and fur
    This expert-derived table fixes the branch structure of SignNet. If the relationships are incorrect, the architecture's inductive bias is incorrect.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Dr. Tongue: Sign-Oriented Multi-label Detection for Remote Tongue Diagnosis." pith.science (2026). https://pith.science/paper/NRZEHYHG

@misc{pith2026250103053,
  author       = {Pith},
  title        = {Pith review of: Dr. Tongue: Sign-Oriented Multi-label Detection for Remote Tongue Diagnosis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NRZEHYHG}},
  note         = {Machine review of arXiv:2501.03053}
}
read the original abstract

Tongue diagnosis is a vital tool in Western and Traditional Chinese Medicine, providing key insights into a patient's health by analyzing tongue attributes. The COVID-19 pandemic has heightened the need for accurate remote medical assessments, emphasizing the importance of precise tongue attribute recognition via telehealth. To address this, we propose a Sign-Oriented multi-label Attributes Detection framework. Our approach begins with an adaptive tongue feature extraction module that standardizes tongue images and mitigates environmental factors. This is followed by a Sign-oriented Network (SignNet) that identifies specific tongue attributes, emulating the diagnostic process of experienced practitioners and enabling comprehensive health evaluations. To validate our methodology, we developed an extensive tongue image dataset specifically designed for telemedicine. Unlike existing datasets, ours is tailored for remote diagnosis, with a comprehensive set of attribute labels. This dataset will be openly available, providing a valuable resource for research. Initial tests have shown improved accuracy in detecting various tongue attributes, highlighting our framework's potential as an essential tool for remote medical assessments.

Figures

Figures reproduced from arXiv: 2501.03053 by the authors.

Figure 1
Figure 1. Challenges in tongue diagnosis imaging for tele [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Cropped images with eight representative tongue [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Original samples of tongue images from our [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Two examples of the Tongue Image Upright Orien [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: Overview of our Sign-Oriented Attributes Detection Framework. The framework consists of three main stages: (1) [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: Receiver Operating Characteristic (ROC) curves [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]
Figure 7
Figure 7. Figure 7: Examples of tongue images with bounding box [PITH_FULL_IMAGE:figures/full_fig_p013_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

53 extracted references · 43 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Bansal, A.; Sikka, K.; Sharma, G.; Chellappa, R.; and Divakaran, A. 2018. Zero-shot object detection. In Proceedings of the European conference on computer vision (ECCV), 384--400

    work page 2018

  4. [4]

    Bucher, M.; Vu, T.-H.; Cord, M.; and P \'e rez, P. 2019. Zero-shot semantic segmentation. Advances in Neural Information Processing Systems, 32

    work page 2019

  5. [5]

    Fu, H.; Cohen-Or, D.; Dror, G.; and Sheffer, A. 2008. Upright orientation of man-made objects. In ACM SIGGRAPH 2008 papers, 1--7

    work page 2008

  6. [6]

    He, J.; Chen, J.-N.; Liu, S.; Kortylewski, A.; Yang, C.; Bai, Y.; and Wang, C. 2022. Transfg: A transformer architecture for fine-grained recognition. In Proceedings of the AAAI conference on artificial intelligence, volume 36, 852--860

    work page 2022

  7. [7]

    H.; Lee, H

    Heo, J.; Lim, J. H.; Lee, H. R.; Jang, J. Y.; Shin, Y. S.; Kim, D.; Lim, J. Y.; Park, Y. M.; Koh, Y. W.; Ahn, S.-H.; et al. 2022. Deep learning model for tongue cancer diagnosis using endoscopic images. Scientific reports, 12(1): 6281

    work page 2022

  8. [8]

    Howard, A.; Sandler, M.; Chu, G.; Chen, L.-C.; Chen, B.; Tan, M.; Wang, W.; Zhu, Y.; Pang, R.; Vasudevan, V.; et al. 2019. Searching for mobilenetv3. In Proceedings of the IEEE/CVF international conference on computer vision, 1314--1324

    work page 2019

Show all 53 references
  1. [9]

    Hu, J.; Han, S.; Chen, Y.; Ji, Z.; et al. 2015. Variations of tongue coating microbiota in patients with gastric cancer. BioMed research international, 2015

    2015

  2. [10]

    Hu, Y.; Wen, G.; Liao, H.; Wang, C.; Dai, D.; and Yu, Z. 2019. Automatic construction of chinese herbal prescriptions from tongue images using CNNs and auxiliary latent therapy topics. IEEE transactions on cybernetics, 51(2): 708--721

    2019

  3. [11]

    Huang, B.; Wu, J.; Zhang, D.; and Li, N. 2010. Tongue shape classification by geometric features. Information Sciences, 180(2): 312--324

    2010

  4. [12]

    Jia, C.; Yang, Y.; Xia, Y.; Chen, Y.-T.; Parekh, Z.; Pham, H.; Le, Q.; Sung, Y.-H.; Li, Z.; and Duerig, T. 2021. Scaling up visual and vision-language representation learning with noisy text supervision. In International conference on machine learning, 4904--4916. PMLR

    2021

  5. [13]

    Jiang, B.; Liang, X.; Chen, Y.; Ma, T.; Liu, L.; Li, J.; Jiang, R.; Chen, T.; Zhang, X.; and Li, S. 2012. Integrating next-generation sequencing and traditional tongue diagnosis to determine tongue coating microbiome. Scientific reports, 2(1): 936

    2012

  6. [14]

    Jiang, T.; Guo, X.-j.; Tu, L.-p.; Lu, Z.; Cui, J.; Ma, X.-x.; Hu, X.-j.; Yao, X.-h.; Cui, L.-t.; Li, Y.-z.; et al. 2021 a . Application of computer tongue image analysis technology in the diagnosis of NAFLD. Computers in Biology and Medicine, 135: 104622

    2021

  7. [15]

    Jiang, T.; Hu, X.-j.; Yao, X.-h.; Tu, L.-p.; Huang, J.-b.; Ma, X.-x.; Cui, J.; Wu, Q.-f.; and Xu, J.-t. 2021 b . Tongue image quality assessment based on a deep convolutional neural network. BMC Medical Informatics and Decision Making, 21(1): 1--14

    2021

  8. [16]

    Jiang, T.; Lu, Z.; Hu, X.; Zeng, L.; Ma, X.; Huang, J.; Cui, J.; Tu, L.; Zhou, C.; Yao, X.; et al. 2022. Deep Learning Multi-label Tongue Image Analysis and Its Application in a Population Undergoing Routine Medical Checkup. Evidence-Based Complementary and Alternative Medicine, 2022

    2022

  9. [17]

    Jiang, X.; Zheng, C.; Xu, X.; Liu, B.; Zheng, W.; Zhang, H.; and He, S. 2024. VrdONE: One-stage Video Visual Relation Detection. In Proceedings of the 32nd ACM International Conference on Multimedia, 1437--1446

    2024

  10. [18]

    Jocher, G.; Chaurasia, A.; Stoken, A.; Borovec, J.; Kwon, Y.; Michael, K.; Fang, J.; Wong, C.; Yifu, Z.; Montes, D.; et al. 2022. ultralytics/yolov5: v6. 2-yolov5 classification models, apple m1, reproducibility, clearml and deci. ai integrations. Zenodo

    2022

  11. [19]

    Ke, L.; Ye, M.; Danelljan, M.; Tai, Y.-W.; Tang, C.-K.; Yu, F.; et al. 2024. Segment anything in high quality. Advances in Neural Information Processing Systems, 36

    2024

  12. [20]

    C.; Kanugula, A

    Kichloo, A.; Albosta, M.; Dettloff, K.; Wani, F.; El-Amir, Z.; Singh, J.; Aljadah, M.; Chakinala, R. C.; Kanugula, A. K.; Solanki, S.; et al. 2020. Telemedicine, the current COVID-19 pandemic and the future: a narrative review and perspectives moving forward in the USA. Family...

    2020

  13. [21]

    C.; Lo, W.-Y.; et al

    Kirillov, A.; Mintun, E.; Ravi, N.; Mao, H.; Rolland, C.; Gustafson, L.; Xiao, T.; Whitehead, S.; Berg, A. C.; Lo, W.-Y.; et al. 2023. Segment anything. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 4015--4026

    2023

  14. [22]

    Lee, T.-C.; Lo, L.-C.; Wu, F.-C.; et al. 2016. Traditional Chinese medicine for metabolic syndrome via TCM pattern differentiation: Tongue diagnosis for predictor. Evidence-Based Complementary and Alternative Medicine, 2016

    2016

  15. [23]

    Li, J.; Li, D.; Xiong, C.; and Hoi, S. 2022. Blip: Bootstrapping language-image pre-training for unified vision-language understanding and generation. In International conference on machine learning, 12888--12900. PMLR

    2022

  16. [24]

    Li, X.; Zhang, Y.; Cui, Q.; Yi, X.; and Zhang, Y. 2018. Tooth-marked tongue recognition using multiple instance learning and CNN features. IEEE transactions on cybernetics, 49(2): 380--387

    2018

  17. [25]

    Liu, S.; Zeng, Z.; Ren, T.; Li, F.; Zhang, H.; Yang, J.; Li, C.; Yang, J.; Su, H.; Zhu, J.; et al. 2023. Grounding dino: Marrying dino with grounded pre-training for open-set object detection. arXiv preprint arXiv:2303.05499

    2023 arXiv

  18. [26]

    Y.; et al

    Lo, L.-c.; Chen, Y.-F.; Chen, W.-J.; Cheng, T.-L.; Chiang, J. Y.; et al. 2012. The study on the agreement between automatic tongue diagnosis system and traditional Chinese medicine practitioners. Evidence-Based Complementary and Alternative Medicine, 2012

    2012

  19. [27]

    A.; Odone, A.; et al

    Ohannessian, R.; Duong, T. A.; Odone, A.; et al. 2020. Global telemedicine implementation and integration within health systems to fight the COVID-19 pandemic: a call to action. JMIR public health and surveillance, 6(2): e18810

    2020

  20. [28]

    Pang, B.; Zhang, D.; Li, N.; and Wang, K. 2004. Computerized tongue diagnosis based on Bayesian networks. IEEE Transactions on biomedical engineering, 51(10): 1803--1810

    2004

  21. [29]

    Portnoy, J.; Waller, M.; and Elliott, T. 2020. Telemedicine in the era of COVID-19. The Journal of Allergy and Clinical Immunology: In Practice, 8(5): 1489--1491

    2020

  22. [30]

    W.; Hallacy, C.; Ramesh, A.; Goh, G.; Agarwal, S.; Sastry, G.; Askell, A.; Mishkin, P.; Clark, J.; et al

    Radford, A.; Kim, J. W.; Hallacy, C.; Ramesh, A.; Goh, G.; Agarwal, S.; Sastry, G.; Askell, A.; Mishkin, P.; Clark, J.; et al. 2021. Learning transferable visual models from natural language supervision. In International conference on machine learning, 8748--8763. PMLR

    2021

  23. [31]

    Shi, Y.; Hu, X.; Cui, J.; Cui, L.; Huang, J.; Ma, X.; Jiang, T.; Yao, X.; Lan, F.; Li, J.; et al. 2021. Clinical data mining on network of symptom and index and correlation of tongue-pulse data in fatigue population. BMC Medical Informatics and Decision Making, 21(1): 1--14

    2021

  24. [32]

    Wang, H.; Zhang, X.; and Cai, Y. 2014. Research on teeth marks recognition in tongue image. In 2014 International Conference on Medical Biometrics, 80--84. IEEE

    2014

  25. [33]

    Wang, J.; Lan, C.; Liu, C.; Ouyang, Y.; Qin, T.; Lu, W.; Chen, Y.; Zeng, W.; and Philip, S. Y. 2022. Generalizing to unseen domains: A survey on domain generalization. IEEE transactions on knowledge and data engineering, 35(8): 8052--8072

    2022

  26. [34]

    Wang, X.; Zhang, B.; Yang, Z.; Wang, H.; and Zhang, D. 2013. Statistical analysis of tongue images for feature extraction and diagnostics. IEEE Transactions on Image Processing, 22(12): 5336--5347

    2013

  27. [35]

    W.; Dai, Z.; Tsvetkov, Y.; and Cao, Y

    Wang, Z.; Yu, J.; Yu, A. W.; Dai, Z.; Tsvetkov, Y.; and Cao, Y. 2021. Simvlm: Simple visual language model pretraining with weak supervision. arXiv preprint arXiv:2108.10904

    2021 arXiv

  28. [36]

    H.; Schiele, B.; and Akata, Z

    Xian, Y.; Lampert, C. H.; Schiele, B.; and Akata, Z. 2018. Zero-shot learning—a comprehensive evaluation of the good, the bad and the ugly. IEEE transactions on pattern analysis and machine intelligence, 41(9): 2251--2265

    2018

  29. [37]

    Xu, C.; Chen, Z.; Mai, J.; Xu, X.; and He, S. 2023. Pose-and attribute-consistent person image synthesis. ACM Transactions on Multimedia Computing, Communications and Applications, 19(2s): 1--21

    2023

  30. [38]

    Xu, C.; Qu, W.; Xu, X.; and Liu, X. 2022. Multi-scale flow-based occluding effect and content separation for cartoon animations. IEEE Transactions on Visualization and Computer Graphics, 29(9): 4001--4014

    2022

  31. [39]

    Yu, Y.; Liu, B.; Zheng, C.; Xu, X.; Zhang, H.; and He, S. 2024. Beyond textual constraints: Learning novel diffusion conditions with fewer examples. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 7109--7118

    2024

  32. [40]

    Yuan, L.; Yang, L.; Zhang, S.; Xu, Z.; Qin, J.; Shi, Y.; Yu, P.; Wang, Y.; Bao, Z.; Xia, Y.; et al. 2023. Development of a tongue image-based machine learning tool for the diagnosis of gastric cancer: a prospective multicentre clinical cohort study. EClinicalMedicine, 57

    2023

  33. [41]

    U.; Bae, S.-H.; Lee, S.; and Hong, C

    Zhang, C.; Han, D.; Qiao, Y.; Kim, J. U.; Bae, S.-H.; Lee, S.; and Hong, C. S. 2023. Faster segment anything: Towards lightweight sam for mobile applications. arXiv preprint arXiv:2306.14289

    2023 arXiv

  34. [42]

    Zhao, Y.; Gou, X.-j.; Dai, J.-y.; Peng, J.-h.; Feng, Q.; Sun, S.-j.; Cao, H.-j.; Zheng, N.-n.; Fang, J.-w.; Jiang, J.; et al. 2013. Differences in metabolites of different tongue coatings in patients with chronic hepatitis B. Evidence-Based Complementary and Alternative Medicine, 2013

    2013

  35. [43]

    Zheng, Y.; Wu, J.; Qin, Y.; Zhang, F.; and Cui, L. 2021. Zero-shot instance segmentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2593--2602

    2021

  36. [44]

    Zhi, L.; Zhang, D.; Yan, J.-q.; Li, Q.-L.; and Tang, Q.-l. 2007. Classification of hyperspectral medical tongue images for tongue diagnosis. Computerized Medical Imaging and Graphics, 31(8): 672--678

    2007

  37. [45]

    Zhu, P.; Wang, H.; and Saligrama, V. 2019. Zero shot detection. IEEE Transactions on Circuits and Systems for Video Technology, 30(4): 998--1010

    2019

  38. [46]

    Zuo, W.; Wang, K.; Zhang, D.; and Zhang, H. 2004. Combination of polar edge detection and active contour model for automated tongue segmentation. In Third International Conference on Image and Graphics (ICIG'04), 270--273. IEEE

    2004

  39. [47]

    Afifi, M.; and Brown, M. S. 2020. Deep White-Balance Editing. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    2020

  40. [48]

    G.; Ommer, B.; and Brown, M

    Afifi, M.; Derpanis, K. G.; Ommer, B.; and Brown, M. S. 2021. Learning Multi-Scale Photo Exposure Correction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 9157--9167

    2021

  41. [49]

    Afifi, M.; Price, B.; Cohen, S.; and Brown, M. S. 2019. When Color Constancy Goes Wrong: Correcting Improperly White-Balanced Images. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    2019

  42. [50]

    Bianco, S.; and Cusano, C. 2019. Quasi-unsupervised color constancy. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 12212--12221

    2019

  43. [51]

    Chang, W.-H.; Chen, C.-C.; Wu, H.-K.; Hsu, P.-C.; Lo, L.-C.; Chu, H.-T.; and Chang, H.-H. 2024. Tongue feature dataset construction and real-time detection. Plos one, 19(3): e0296070

    2024

  44. [52]

    C.; Chen, Y.; Xie, Y

    Ho, S. C.; Chen, Y.; Xie, Y. J.; Yeung, W. F.; CHEN, S. C.; and Qin, J. 2024. Visceral Condition Assessment through Digital Tongue Image Analysis. Frontiers in Artificial Intelligence, 7: 1501184

    2024

  45. [53]

    Wang, X.; and Zhang, D. 2010. An Optimized Tongue Image Color Correction Scheme. IEEE Transactions on Information Technology in Biomedicine, 14(6): 1355--1364

    2010

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.