Fig 1.
Flow chart of implementation process.
Table 1.
Comparison of second AFB smear-positive status among different TB patients.
Fig 2.
The result of logistic regression.
Table 2.
Assignment table of dependent variable and independent variables.
Table 3.
Second AFB smear-test results of TB patients with different vitamin D levels.
Fig 3.
Bayesian network model for second AFB smear-positive in TB patients.
A: Initial network structure. B: Bayesian network structure after adding constraints according to prior knowledge. C: Trimmed Bayesian network structure.
Table 4.
The definition and assignment table of Bayesian network nodes.
Fig 4.
Conditional probability table of second AFB smear-positive.
Table 5.
Edges and edge type probabilities of Bayesian network node.