Search Results

You are looking at 1 - 3 of 3 items for

  • Author or Editor: Jonathan R. Jagid x
  • Refine by Access: all x
Clear All Modify Search
Free access

Machine learning to predict passenger mortality and hospital length of stay following motor vehicle collision

John Paul G. Kolcun, Brian Covello, Joanna E. Gernsback, Iahn Cajigas, and Jonathan R. Jagid

OBJECTIVE

Motor vehicle collisions (MVCs) account for 1.35 million deaths and cost $518 billion US dollars each year worldwide, disproportionately affecting young patients and low-income nations. The ability to successfully anticipate clinical outcomes will help physicians form effective management strategies and counsel families with greater accuracy. The authors aimed to train several classifiers, including a neural network model, to accurately predict MVC outcomes.

METHODS

A prospectively maintained database at a single institution’s level I trauma center was queried to identify all patients involved in MVCs over a 20-year period, generating a final study sample of 16,287 patients from 1998 to 2017. Patients were categorized by in-hospital mortality (during admission) and length of stay (LOS), if admitted. All models included age (years), Glasgow Coma Scale (GCS) score, and Injury Severity Score (ISS). The in-hospital mortality and hospital LOS models further included time to admission.

RESULTS

After comparing a variety of machine learning classifiers, a neural network most effectively predicted the target features. In isolated testing phases, the neural network models returned reliable, highly accurate predictions: the in-hospital mortality model performed with 92% sensitivity, 90% specificity, and a 0.98 area under the receiver operating characteristic curve (AUROC), and the LOS model performed with 2.23 days mean absolute error after optimization.

CONCLUSIONS

The neural network models in this study predicted mortality and hospital LOS with high accuracy from the relatively few clinical variables available in real time. Multicenter prospective validation is ultimately required to assess the generalizability of these findings. These next steps are currently in preparation.

Free access

Simple wound closure compared with surgery for civilian cranial gunshot wounds

Evan M. Krueger, Ronald J. Benveniste, Victor M. Lu, Ruby R. Taylor, Rahul Kumar, Joacir G. Cordeiro, and Jonathan R. Jagid

OBJECTIVE

A carefully selected subset of civilian cranial gunshot wound (CGSW) patients may be treated with simple wound closure (SWC) as a proactive therapy, but the appropriate clinical scenario for using this strategy is unknown. The aim of this study was to compare SWC and surgery patients in terms of their neurological outcomes and complications, including infections, seizures, and reoperations.

METHODS

This was a single-center, retrospective review of the prospectively maintained institutional traumatic brain injury and trauma registries. Included were adults who sustained an acute CGSW defined as suspected or confirmed dural penetration. Excluded were nonfirearm penetrating injuries, patients with an initial Glasgow Coma Scale (GCS) score of 3, patients with an initial GCS score of 4 and nonreactive pupils, and patients who died within 48 hours of presentation.

RESULTS

A total of 67 patients were included; 17 (25.4%) were treated with SWC and 50 (74.6%) were treated with surgery. The SWC group had a lower incidence of radiographic mass effect (3/17 [17.6%] SWC vs 31/50 [62%] surgery; absolute difference 44.4, 95% CI −71.9 to 16.8; p = 0.002) and lower incidence of involvement of the frontal sinus (0/17 [0%] SWC vs 14/50 [28%] surgery; absolute difference 28, 95% CI −50.4 to 5.6; p = 0.01). There were no differences in the frequency of Glasgow Outcome Scale–Extended scores ≥ 5 between the SWC and surgery groups at 30 days (4/11 [36.4%] SWC vs 12/35 [34.3%] surgery; OR 1.1, 95% CI 0.3–4.5; p > 0.99), 60 days (2/7 [28.6%] SWC vs 8/26 [30.8%] surgery; OR 0.9, 95% CI 0.3–3.4; p > 0.99), and 90 days (3/8 [37.5%] SWC vs 12/26 [46.2%] surgery; OR 0.7, 95% CI 0.1–3.6; p > 0.99). There were no differences in the incidence of infections (1/17 [5.9%] SWC vs 6/50 [12%] surgery; OR 0.5, 95% CI 0.1–4.1; p = 0.67), CSF fistulas (2/11 [11.6%] SWC vs 3/50 [6%] surgery; OR 2.1, 95% CI 0.3–13.7; p = 0.60), seizures (3/17 [17.6%] SWC vs 9/50 [18%] surgery; OR 1, 95% CI 0.2–4.1; p > 0.99), and reoperations (3/17 [17.6%] SWC vs 4/50 [8%] surgery; OR 2.5, 95% CI 0.5–12.4; p = 0.36) between the SWC and surgery groups.

CONCLUSIONS

There were important clinically relevant differences between the SWC and surgery groups. SWC can be considered a safe and efficacious proactive therapy in a carefully selected subset of civilian CGSW patients.

Restricted access

2017 AANS Annual Scientific Meeting Los Angeles, CA • April 22–26, 2017