Predicting recovery in patients suffering from traumatic brain injury by using admission variables and physiological data: a comparison between decision tree analysis and logistic regression

Andrews, Peter; Sleeman, Derek; Statham, Patrick; McQuatt, A.; Corruble, Vincent; Jones, Patricia A.; Howells, Timothy; Macmillan, Carol S. A.

doi:10.3171/jns.2002.97.2.0326

Cited by 202 publications

(127 citation statements)

References 29 publications

Supporting

Mentioning

118

Contrasting

Unclassified

Order By: Relevance

“…2). Even though there is a lack of consensus regarding the ideal range of CPP after TBI, and the error difference of more than 10 mm Hg may or may not be clinically relevant depending on the duration of such a discrepancy, 4,41 both the normal CPP range of 70-85 mm Hg 41 and suggestions for maintaining a CPP of at least 70 mm Hg after head injury, 35 or between 50 and 70 mm Hg, 7 or at a static autoregulation range (i.e., 60-70 mm Hg) 12 are within the range of the highest accuracy of eCPP. This ability of eCPP to estimate CPP most accurately at an expected or suggested level of CPP seems promising for clinical use of the method.…”

Section: Cerebral Perfusion Pressure As a Numbermentioning

confidence: 99%

“…6,15,29,31,39 Hence, monitoring and management of CPP will remain one of the cornerstones of TBI management, even though it is recognized that further work is required in terms of patient-specific thresholds. 4,41 For this reason, the development of noninvasive methods to monitor CPP is reasonable and could provide an additional option in resource-limited settings or for patients who have contraindications for invasive monitoring.…”

Section: The Role Of Icpmentioning

confidence: 99%

“…4,41 The invasive part of CPP calculation is ICP, the invasive nature of which can make the calculation of CPP and thus the CPP-oriented management unfeasible, such as in some clinical scenarios in which invasive measurement of ICP is either not available or unobtainable (i.e., for patients with contraindications for direct measurement). 3,8,27 In contrast, ABP can be measured and monitored noninvasively, 20,30,37 i.e., with a Finapres finger plethysmograph, the use of which has been shown to provide a good level of agreement between Finapres-derived indices of cerebral autoregulation and their corresponding estimates obtained from invasive measurements of aortic ABP.…”

mentioning

confidence: 99%

See 2 more Smart Citations

A noninvasive estimation of cerebral perfusion pressure using critical closing pressure

Varsos

Kolias

Smielewski

et al. 2015

JNS

View full text Add to dashboard Cite

T he circulation of cerebral blood flow (CBF) is driven by cerebral perfusion pressure (CPP), which is defined as the vascular pressure gradient across the cerebral bed and can be calculated as the difference between arterial blood pressure (ABP) and pressure in cortical or bridging veins. 15,24 Due to difficulties in measuring the pressure of bridging veins, invasive intracranial pressure (ICP) measurements are used instead as an approximation, defining CPP as ABP − ICP. 1,26,34,39 Cerebral perfusion pressure in clinical practice is considered an esabbreviatioNs ABP = arterial blood pressure; AUC = area under the curve; CBF = cerebral blood flow; CPP = cerebral perfusion pressure; CrCP = critical closing pressure; CVR = cerebrovascular resistance; eCPP = noninvasive estimator of CPP; FV = flow velocity; GCS = Glasgow Coma Scale; GOS = Glasgow Outcome Scale; ICP = intracranial pressure; IQR = interquartile range; MCA = middle cerebral artery; nICP = noninvasive estimator of ICP; ROC = receiver operating characteristic; TAU = time constant of the cerebrovascular arterial bed; TBI = traumatic brain injury; TCD = transcranial Doppler. obJect Cerebral blood flow is associated with cerebral perfusion pressure (CPP), which is clinically monitored through arterial blood pressure (ABP) and invasive measurements of intracranial pressure (ICP). Based on critical closing pressure (CrCP), the authors introduce a novel method for a noninvasive estimator of CPP (eCPP). methods Data from 280 head-injured patients with ABP, ICP, and transcranial Doppler ultrasonography measurements were retrospectively examined. CrCP was calculated with a noninvasive version of the cerebrovascular impedance method. The eCPP was refined with a predictive regression model of CrCP-based estimation of ICP from known ICP using data from 232 patients, and validated with data from the remaining 48 patients. results Cohort analysis showed eCPP to be correlated with measured CPP (R = 0.851, p < 0.001), with a mean ± SD difference of 4.02 ± 6.01 mm Hg, and 83.3% of the cases with an estimation error below 10 mm Hg. eCPP accurately predicted low CPP (< 70 mm Hg) with an area under the curve of 0.913 (95% CI 0.883-0.944). When each recording session of a patient was assessed individually, eCPP could predict CPP with a 95% CI of the SD for estimating CPP between multiple recording sessions of 1.89-5.01 mm Hg. coNclusioNs Overall, CrCP-based eCPP was strongly correlated with invasive CPP, with sensitivity and specificity for detection of low CPP that show promise for clinical use.

show abstract

Section: Cerebral Perfusion Pressure As a Numbermentioning

confidence: 99%

Section: The Role Of Icpmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

A noninvasive estimation of cerebral perfusion pressure using critical closing pressure

Varsos

Kolias

Smielewski

et al. 2015

JNS

View full text Add to dashboard Cite

show abstract

“…Andrews et al, conducted a comparative study between decision trees and logistic regression to predict improvement in patients with ABI [13]. In Yi et al [14], compared different learning techniques (decision tree, adaboost, support vector machine and artificial neural networks) to be used as a decision support tool based on rules for the treatment of patients with TBI.…”

Section: Predictive Data Mining In Abi Rehabilitationmentioning

confidence: 99%

“…The best-known methods are decision trees (DT) and ANNs, both of which are widely used in the ABI literature [13,14].…”

Section: Predictive Data Mining In Abi Rehabilitationmentioning

confidence: 99%

Artificial metaplasticity prediction model for cognitive rehabilitation outcome in acquired brain injury patients

Marcano-Cedeño

Chausa

García

et al. 2013

Artificial Intelligence in Medicine

View full text Add to dashboard Cite

Objective: The main purpose of this research is the novel use of artificial metaplasticity on multilayer perceptron (AMMLP) as a data mining tool for prediction the outcome of patients with acquired brain injury (ABI) after cognitive rehabilitation. The final goal aims at increasing knowledge in the field of rehabilitation theory based on cognitive affectation. Methods and materials:The data set used in this study contains records belonging to 123 ABI patients with moderate to severe cognitive affectation (according to Glasgow Coma Scale) that underwent rehabilitation at Institut Guttmann Neurorehabilitation Hospital (IG) using the tele-rehabilitation platform PREVIRNEC®. The variables included in the analysis comprise the neuropsychological initial evaluation of the patient (cognitive affectation profile), the results of the rehabilitation tasks performed by the patient in PREVIRNEC® and the outcome of the patient after a 3-5 months treatment. To achieve the treatment outcome prediction, we apply and compare three different data mining techniques: the AMMLP model, a backpropagation neural network (BPNN) and a C4.5 decision tree. Results: The prediction performance of the models was measured by ten-fold cross validation and several architectures were tested. The results obtained by the AMMLP model are clearly superior, with an average predictive performance of 91.56%. BPNN and C4.5 models have a prediction average accuracy of 80.18% and 89.91% respectively. The best single AMMLP model provided a specificity of 92.38%, a sensitivity of 91.76% and a prediction accuracy of 92.07%. Conclusions: The proposed prediction model presented in this study allows to increase the knowledge about the contributing factors of an ABI patient recovery and to estimate treatment efficacy in individual patients. The ability to predict treatment outcomes may provide new insights toward improving effectiveness and creating personalized therapeutic interventions based on clinical evidence.

show abstract

Trauma in the Elderly Patient

Victorino

Chong²,

Pal³

2003

Arch Surg

107

View full text Add to dashboard Cite

dvanced age is a well-known risk factor for poor outcomes in trauma patients. Older patients can benefit from the intensive monitoring and aggressive management associated with trauma team involvement. Several common topics were chosen for discussion in which the treatment options may differ slightly because of the advanced age of the patient. We discuss the following selected topics: resuscitation of the elderly trauma patient, solid organ injuries, hip fractures, rib fractures, and head injuries.

show abstract

Predicting recovery in patients suffering from traumatic brain injury by using admission variables and physiological data: a comparison between decision tree analysis and logistic regression

Abstract: Decision tree analysis confirmed some of the results of logistic regression and challenged others. This investigation shows that there is knowledge to be gained from analyzing observational data with the aid of decision tree analysis.

Cited by 202 publications

References 29 publications

A noninvasive estimation of cerebral perfusion pressure using critical closing pressure

A noninvasive estimation of cerebral perfusion pressure using critical closing pressure

Artificial metaplasticity prediction model for cognitive rehabilitation outcome in acquired brain injury patients

Trauma in the Elderly Patient

Contact Info

Product

Resources

About