prev next front |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 |16 |17 |18 |19 |20 |21 |22 |23 |24 |review
Low rates of agreement between algorithms and GI were found (average 43%). Total agreement was not found for any algorithm. None of the adverse events was imputed equally by all the algorithms. Algorithms presented high rates of sensitivity (average of 93%) and low rates of specificity (average of 7%).