Add to Favourites
To login click here
This study examined the impact of implementing predictive models on the subsequent performance of those and other models in two major healthcare institutions. The researchers found that using the models to adjust how care is delivered can alter the baseline assumptions that the models were “trained” on, often for worse. They simulated critical care scenarios and investigated three key scenarios, including retraining models to address performance degradation over time and creating a new model after one has a significant impact on patient care. The study concluded that machine learning models in healthcare can be victims of their own success.
