Publications by authors named "Theo Saarinen"

Article Synopsis
  • Guidelines suggest using risk scores to identify very-low-risk patients with gastrointestinal bleeding (GIB) for possible discharge from emergency departments.
  • A new machine learning model was developed and tested against existing scores (Glasgow-Blatchford and Oakland) using data from nearly 3,500 patients across different hospitals.
  • Results showed the machine learning model performed better in predicting very-low-risk patients, successfully identifying a higher percentage compared to the existing risk scores.
View Article and Find Full Text PDF