Predicting the diabetic foot in the population of type 2 diabetes mellitus from tongue images and clinical information using multi-modal deep learning
2024

Predicting Diabetic Foot in Type 2 Diabetes from Tongue Images and Clinical Data

Sample size: 391 publication 10 minutes Evidence: high

Author Information

Author(s): Tian Zhikui, Wang Dongjun, Sun Xuan, Cui Chuan, Wang Hongwu

Primary Institution: Tianjin University of Traditional Chinese Medicine

Hypothesis

Can tongue images and clinical information be used to predict diabetic foot in patients with type 2 diabetes mellitus?

Conclusion

A diabetic foot prediction model was established that effectively combines tongue images and clinical data, demonstrating the value of tongue diagnosis in predicting diabetic foot risk.

Supporting Evidence

  • The model achieved an accuracy of 0.95 and sensitivity of 0.9286.
  • Diabetes duration was significantly higher in diabetic foot patients compared to those with type 2 diabetes.
  • Plantar hardness was greater in diabetic foot patients than in the type 2 diabetes group.

Takeaway

Doctors can use pictures of your tongue and some health information to tell if you might get foot problems from diabetes.

Methodology

The study used a ResNet-50 deep neural network to analyze tongue images and clinical data to predict diabetic foot.

Potential Biases

Potential biases may arise from the subjective interpretation of tongue images.

Limitations

The study may be limited by the sample size and the specific population from which participants were drawn.

Participant Demographics

The study included 391 participants, with a majority being male and an average age of 67 years in the diabetic foot group.

Statistical Information

P-Value

<0.001

Confidence Interval

0.8694–0.9993

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.3389/fphys.2024.1473659

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication