Prediction of a Healing Trajectory of Chronic Wounds Using a Machine Learning Approach

Download PDF

Key Takeaways: 

• Early change in wound size is the dominant predictor of healing.  

• Machine learning substantially outperforms traditional regression models.  

• Predictions become highly accurate by weeks 4–5 of treatment (AUC 0.90–0.92).  

• The model could enable earlier intervention for wounds unlikely to heal, improving outcomes and resource utilization. 

Read the full publication: prediction of a healing trajectory of chronic wounds using a machine learning approach