D
DiabetesAI
Risk screening
ML model online

Diabetes risk,estimated with ML.

Enter the patient measurements below to receive a model-predicted diabetes risk estimate.

Patient information

Enter the values used by the trained model.

For glucose, blood pressure, skin thickness, insulin, and BMI, a value of 0 is treated by the model as a missing measurement and imputed during preprocessing.

This tool provides a machine-learning risk estimate and is not a medical diagnosis.

Prediction result

Waiting
+

Ready for analysis

Fill in the patient information and run the model to see the predicted probability.

MODEL OVERVIEW
Algorithm
Gradient Boosting
ROC-AUC
81.86%
Recall
74.07%
Threshold
34.3%
How it works
01Enter patient measurements.
02FastAPI sends the values through the saved ML pipeline.
03The model returns a probability and risk classification.