Error-correction learning for artificial neural networks using the Bayesian paradigm. Application to automated medical diagnosis

- A novel Bayesian-based strategy for training MLPs is proposed.
- Six medical datasets (breast and lung cancer, heart attack and diabetes) were used for assessment.
- Statistical benchmark has revealed the effectiveness of the model.
- The corresponding algorithm is easy to understand and implement.
- The model is prone to easily adapt to different medical decision-making issues.