Predicting Cardiopulmonary Arrest In Critically Ill Children

Cardiopulmonary arrest (CPA) is a devastating complication in critically ill children associated with death and significant disability amongst survivors. The only effective treatment for CPA is cardiopulmonary resuscitation (CPR). However, preventing CPA from happening in the first place would help improve outcomes for this population of critically ill children.

Effective prevention requires an ability to reliably predict the likelihood a patient may experience CPA. SickKids researchers are applying machine learning to this problem using convolutional and recurrent neural networks and continuous physiological signals captured from the patient bedside, in an effort to identify potential risks and predict CPA before it happens. The ensemble risk model developed by the team is being deployed in a silent trial to evaluate the model’s performance as a clinical risk mitigation tool.