Skip to main content

Table 3 Correlation coefficients of true vs. predicted intracranial pressure (ICP), and cerebral perfusion pressures (CPP), from a non-linear multivariate analysis method (artificial neural networks)a

From: Analyses of cerebral microdialysis in patients with traumatic brain injury: relations to intracranial pressure, cerebral perfusion pressure and catheter placement

 

Pericontusional

Nonpericontusional

ICP

0.165 ± 0.013

0.372 ± 0.019

   Permuted

0.129 ± 0.016

0.265 ± 0.021

CPP

0.134 ± 0.018

-0.143 ± 0.020

   Permuted

0.059 ± 0.021

-0.111 ± 0.030

  1. aICP, intracranial pressure; CPP, cerebral perfusion pressure. Microdialysis markers (glucose, lactate, pyruvate and glycerol) and ratios (LP ratio and LG ratio) were used as explanatory variables predicting mean ICP/CPP (0 to 60 minutes prior to microdialysis (MD) sample time). In addition, correlations were analyzed after randomly permuting (scrambling) MD data per subject in relation to ICP/CPP to access what remained of correlations if no longer serially related. Negative correlation (modeling positive prediction) is interpreted as "even worse." Data are presented as means ± standard deviation (SD). SD values represent the variation of artificial neural network solutions from 200 separate runs with random initializations.