Skip to main content
. 2019 Mar 7;14(3):e0213257. doi: 10.1371/journal.pone.0213257

Fig 4.

Fig 4

Results considering the more restrictive analysis using leave-one-out cross-validation in the best FOT parameter and FOT associated with machine learning methods (obtained in the first part of this research [28]). These results are compared with those obtained in the present study using FrOr modeling and FrOr modeling combined with machine learning methods in patients with normal (A) and abnormal (B) spirometry.