Skip to main content
. 2016 Dec 2;475(3):872–880. doi: 10.1007/s11999-016-5180-x

Fig. 2.

Fig. 2

The Bayesian network for the TA3 and TA4 models are shown. The arrow and bar edges are assigned to positive and negative correlations, respectively. The probability estimates for posttibialis anterior muscle strength necessitates direct predictors (posttibialis anterior muscle strength and duration). Indirect predictors (age, hernia, and pain) are not necessary for the prediction given the information of the two direct predictors (Table 5). Pre-TA = preoperative tibialis anterior muscle strength; post-TA = postoperative tibialis anterior muscle strength; duration = duration of drop foot before surgery.