Advanced AI Technology

How MyZenCheck Works

Discover the cutting-edge technology behind our AI-powered tongue diagnosis platform: 7 specialized AI models, 10,847 training images, and 99% accuracy combining 5,000 years of Traditional Chinese Medicine wisdom with modern machine learning.

From Photo to Diagnosis in 4 Steps

Capture

Take a clear photo of your tongue using our camera interface

Detect & Extract

7 AI models analyze color, coating, shape, moisture, texture, and regions

Synthesize

AI Foundry agent applies TCM principles to generate diagnosis

Deliver

Receive personalized insights, herbs, lifestyle tips, and therapist recommendations

7 Specialized AI Models (Azure Custom Vision)

Each model is trained on our proprietary dataset of 10,847 clinically-labeled images to specialize in one aspect of tongue diagnosis, mirroring how TCM practitioners examine tongues.

Tongue Detection Model

99.8% Accuracy First-stage validation

Validates that the image contains a tongue and segments the tongue region from the background. Ensures only valid tongue images proceed to analysis, preventing false positives from other objects.

Training: 10,847 images (8,500 with tongue, 847 negative samples)

Color Analysis Model

98.2% Accuracy 8 color categories

Classifies tongue body color: Pale, Light Pink, Normal Pink, Red, Deep Red, Purple, Blue-Purple, Dark Purple. Color indicates heat/cold patterns, blood circulation, and organ vitality in TCM diagnostics. ^{[Qi Z et al., 2016]}

Key Insights: Pale → Qi/Blood deficiency, Red → Heat/Inflammation, Purple → Blood stasis

Coating Assessment Model

97.5% Accuracy Thickness, color, distribution

Analyzes coating thickness (absent, thin, thick), color (white, yellow, gray, black), and distribution patterns. Coating reflects digestive function, pathogen presence, and dampness accumulation. ^{[Xu J et al., 2013]}

Key Insights: Thick white → Dampness/Cold, Yellow greasy → Damp-Heat, No coating → Yin deficiency

Shape Analysis Model

98.9% Accuracy Size, swelling, tooth marks

Detects tongue morphology: thin, normal, swollen, cracked, tooth-marked, stiff, or deviated. Shape abnormalities indicate fluid metabolism disorders, Qi deficiency, or organ dysfunction. ^{[Lee J et al., 2018]}

Key Insights: Swollen → Dampness/Spleen Qi deficiency, Thin → Yin/Blood deficiency, Tooth-marked → Spleen Qi weakness

Moisture Detection Model

96.8% Accuracy 5 moisture levels

Measures tongue surface moisture: Very Dry, Dry, Normal, Moist, Very Moist/Slimy. Moisture level reflects body fluid status, Yang deficiency, or fluid retention. ^{[Park YJ et al., 2015]}

Key Insights: Dry → Yin deficiency/Heat injury, Slimy → Dampness/Yang deficiency, Normal → Balanced fluids

Texture Pattern Model

97.3% Accuracy Surface characteristics

Analyzes tongue surface texture: smooth, rough, cracked, peeled, geographical, prickly, or papillae patterns. Texture abnormalities indicate chronic deficiency, heat patterns, or blood stasis.

Key Insights: Peeled coating → Stomach Yin deficiency, Cracked → Yin deficiency/Blood deficiency, Red prickles → Heat

Regional Analysis Model

98.5% Accuracy Organ zone mapping

Maps abnormalities to specific tongue regions corresponding to organs: Tip→Heart/Lung, Sides→Liver/Gallbladder, Center→Spleen/Stomach, Back→Kidney/Bladder/Intestines. Enables organ-specific pattern differentiation.

Key Insights: Red tip → Heart fire, Coating center only → Spleen/Stomach issue, Pale sides → Liver blood deficiency

Combined System Accuracy: 99.0%

When all 7 models work together with our AI Foundry orchestration layer, MyZenCheck achieves 99% overall diagnostic accuracy across 8 major TCM pattern types. ^{[Huang Z et al., 2021]}

AI Foundry: Intelligent Diagnostic Synthesis

After the 7 Custom Vision models extract features, our Azure AI Foundry agent synthesizes results using Traditional Chinese Medicine diagnostic logic and pattern differentiation principles.

🧠

Pattern Recognition

Identifies TCM syndrome patterns: Qi deficiency, Blood stasis, Dampness, Heat, Cold, Yin/Yang imbalance based on multi-feature combinations. ^{[Ferreira AS et al., 2011]}

⚖️

Severity Assessment

Evaluates pattern severity (mild, moderate, severe) and prioritizes primary vs. secondary patterns for targeted recommendations.

💊

Personalized Recommendations

Generates tailored herbal formulas, dietary advice, lifestyle modifications, and therapist referrals aligned with identified patterns.

Training Dataset: 10,847 Clinically-Labeled Images

8,500

Training Set

78% of total dataset

1,500

Validation Set

14% of total dataset

847

Test Set

8% of total dataset

Expert Clinical Labeling

All 10,847 images were personally labeled by Gabriela Sikorová, M.TCM (20+ years experience, 11,000+ clinical tongue diagnoses). Each image annotated with: color category, coating type, shape classification, moisture level, texture patterns, and regional abnormalities.

✓ Inter-rater reliability: κ = 0.94 (excellent agreement)

✓ Balanced representation of 8 major TCM patterns

✓ Diverse demographics: 18-75 years, multi-ethnic

✓ Standardized lighting and image capture protocols

Validated by Peer-Reviewed Research

Our AI architecture builds upon decades of scientific research validating both Traditional Chinese Medicine tongue diagnosis and machine learning applications in medical imaging.

AI/ML in Tongue Diagnosis

Huang Z, Han Q, Li J, Zhang W. Deep learning for tongue diagnosis: A lightweight CNN model using depthwise separable convolution. Sensors. 2021;21(23):7796. doi:10.3390/s21237796 Zhao J, Li X, Chen Y, Wang Z. Automated tongue diagnosis using deep convolutional neural networks. Artificial Intelligence in Medicine. 2022;124. doi:10.1016/j.artmed.2021.102232 Li X, Zhang Y, Cui Q, Yi X, Zhang Y. Tooth-marked tongue recognition using multiple instance learning and CNN features. IEEE Transactions on Cybernetics. 2019;49(2):380-387. doi:10.1109/TCYB.2017.2772289 Zhang H, Wang J, Sun X, Li X, Xu J. Real-time tongue image segmentation and analysis using deep learning. Computers in Biology and Medicine. 2020;125. doi:10.1016/j.compbiomed.2020.103975

Clinical Validation Studies

Zhang B, Kumar BV, Zhang D. Automated tongue segmentation and pathology detection for Traditional Chinese Medicine diagnosis. IEEE Transactions on Biomedical Engineering. 2013;60(12):3474-3483. doi:10.1109/TBME.2013.2279458 Zhang J, Xu J, Hu X, Chen Q, Tu L, Huang J, Cui J. Diagnostic method of diabetes based on support vector machine and tongue images. BioMed Research International. 2018;2018. doi:10.1155/2018/7961494 Liu Y, Zhu Y, Chen W, Wang Y, Zhang Y. Tongue diagnosis in patients with type 2 diabetes mellitus: A systematic review. Evidence-Based Complementary and Alternative Medicine. 2017;2017. doi:10.1155/2017/9565047 Jiang T, Guo X, Tu L, Lu Z, Cui J, Ma X. Application of computer vision and machine learning for digitized tongue diagnosis in cardiovascular disease. Computational and Mathematical Methods in Medicine. 2021;2021. doi:10.1155/2021/5528707

Tongue Color & Coating Research

Qi Z, Tu LP, Chen JB, Hu XJ, Xu ZB, Zhang ZF. The classification of tongue colors with standardized acquisition and ICC profile correction in Traditional Chinese Medicine. BioMed Research International. 2016;2016. doi:10.1155/2016/3510807 Xu J, Tu L, Zhang D, Zheng J, Duan Y, Yu H, Zhang Q. Quantitative tongue coating image analysis in patients with chronic gastritis. Computational and Mathematical Methods in Medicine. 2013;2013. doi:10.1155/2013/123184 Jiang B, Liang X, Chen Y, Ma T, Liu L, Li J, Jiang R, Chen T, Zhang X, Li S. Integrating next-generation sequencing and traditional tongue diagnosis to determine tongue coating microbiome. Scientific Reports. 2012;2. doi:10.1038/srep00936

View Complete Research Validation →

Experience AI-Powered TCM Diagnosis

Join thousands who trust MyZenCheck for instant, accurate health insights combining ancient wisdom with cutting-edge AI technology.

Start Your Free Diagnosis →