Search Results

You are looking at 1 - 3 of 3 items for

  • Author or Editor: Oscar D. Guillamondegui x
Clear All Modify Search
Restricted access

Andrew T. Hale, David P. Stonko, Jaims Lim, Oscar D. Guillamondegui, Chevis N. Shannon and Mayur B. Patel

OBJECTIVE

Pediatric traumatic brain injury (TBI) is common, but not all injuries require hospitalization. A computational tool for ruling in patients who will have a clinically relevant TBI (CRTBI) would be valuable, providing an evidence-based way to safely discharge children who are at low risk for a CRTBI. The authors hypothesized that an artificial neural network (ANN) trained on clinical and radiologist-interpreted imaging metrics could provide a tool for identifying patients likely to suffer from a CRTBI.

METHODS

The authors used the prospectively collected, publicly available, multicenter Pediatric Emergency Care Applied Research Network (PECARN) TBI data set. All patients under the age of 18 years with TBI and admission head CT imaging data were included. The authors constructed an ANN using clinical and radiologist-interpreted imaging metrics in order to predict a CRTBI, as previously defined by PECARN: 1) neurosurgical procedure, 2) intubation > 24 hours as direct result of the head trauma, 3) hospitalization ≥ 48 hours and evidence of TBI on a CT scan, or 4) death due to TBI.

RESULTS

Among 12,902 patients included in this study, 480 were diagnosed with CRTBI. The authors’ ANN had a sensitivity of 99.73% with precision of 98.19%, accuracy of 97.98%, negative predictive value of 91.23%, false-negative rate of 0.0027%, and specificity for CRTBI of 60.47%. The area under the receiver operating characteristic curve was 0.9907.

CONCLUSIONS

The authors are the first to utilize artificial intelligence to predict a CRTBI in a clinically meaningful manner, using radiologist-interpreted CT information, in order to identify pediatric patients likely to suffer from a CRTBI. This proof-of-concept study lays the groundwork for future studies incorporating iterations of this algorithm directly into the electronic medical record for real-time, data-driven predictive assistance to physicians.

Restricted access

Michael F. Stiefel, Alejandro Spiotta, Vincent H. Gracias, Alicia M. Garuffe, Oscar Guillamondegui, Eileen Maloney-Wilensky, Stephanie Bloom, M. Sean Grady and Peter D. LeRoux

Object. An intracranial pressure (ICP) monitor, from which cerebral perfusion pressure (CPP) is estimated, is recommended in the care of severe traumatic brain injury (TBI). Nevertheless, optimal ICP and CPP management may not always prevent cerebral ischemia, which adversely influences patient outcome. The authors therefore determined whether the addition of a brain tissue oxygen tension (PO2) monitor in the treatment of TBI was associated with an improved patient outcome.

Methods. Patients with severe TBI (Glasgow Coma Scale [GCS] score < 8) who had been admitted to a Level I trauma center were evaluated as part of a prospective observational database. Patients treated with ICP and brain tissue PO2 monitoring were compared with historical controls matched for age, pathological features, admission GCS score, and Injury Severity Score who had undergone ICP monitoring alone. Therapy in both patient groups was aimed at maintaining an ICP less than 20 mm Hg and a CPP greater than 60 mm Hg. Among patients whose brain tissue PO2 was monitored, oxygenation was maintained at levels greater than 25 mm Hg. Twenty-five patients with a mean age of 44 ± 14 years were treated using an ICP monitor alone. Twenty-eight patients with a mean age of 38 ± 18 years underwent brain tissue PO2-directed care. The mean daily ICP and CPP levels were similar in each group. The mortality rate in patients treated using conventional ICP and CPP management was 44%. Patients who also underwent brain tissue PO2 monitoring had a significantly reduced mortality rate of 25% (p < 0.05).

Conclusions. The use of both ICP and brain tissue PO2 monitors and therapy directed at brain tissue PO2 is associated with reduced patient death following severe TBI.

Restricted access

Andrew T. Hale, David P. Stonko, Jaims Lim, Oscar D. Guillamondegui, Chevis N. Shannon and Mayur B. Patel

OBJECTIVE

Pediatric traumatic brain injury (TBI) is common, but not all injuries require hospitalization. A computational tool for ruling in patients who will have a clinically relevant TBI (CRTBI) would be valuable, providing an evidence-based way to safely discharge children who are at low risk for a CRTBI. The authors hypothesized that an artificial neural network (ANN) trained on clinical and radiologist-interpreted imaging metrics could provide a tool for identifying patients likely to suffer from a CRTBI.

METHODS

The authors used the prospectively collected, publicly available, multicenter Pediatric Emergency Care Applied Research Network (PECARN) TBI data set. All patients under the age of 18 years with TBI and admission head CT imaging data were included. The authors constructed an ANN using clinical and radiologist-interpreted imaging metrics in order to predict a CRTBI, as previously defined by PECARN: 1) neurosurgical procedure, 2) intubation > 24 hours as direct result of the head trauma, 3) hospitalization ≥ 48 hours and evidence of TBI on a CT scan, or 4) death due to TBI.

RESULTS

Among 12,902 patients included in this study, 480 were diagnosed with CRTBI. The authors’ ANN had a sensitivity of 99.73% with precision of 98.19%, accuracy of 97.98%, negative predictive value of 91.23%, false-negative rate of 0.0027%, and specificity for CRTBI of 60.47%. The area under the receiver operating characteristic curve was 0.9907.

CONCLUSIONS

The authors are the first to utilize artificial intelligence to predict a CRTBI in a clinically meaningful manner, using radiologist-interpreted CT information, in order to identify pediatric patients likely to suffer from a CRTBI. This proof-of-concept study lays the groundwork for future studies incorporating iterations of this algorithm directly into the electronic medical record for real-time, data-driven predictive assistance to physicians.