Neural Networks to Predict Radiographic Brain Injury in Pediatric Patients Treated with Extracorporeal Membrane Oxygenation

Abstract

Brain injury is a significant source of morbidity and mortality for pediatric patients treated with Extracorporeal Membrane Oxygenation (ECMO). Our objective was to utilize neural networks to predict radiographic evidence of brain injury in pediatric ECMO-supported patients and identify specific variables that can be explored for future research. Data from 174 ECMO-supported patients were collected up to 24 h prior to, and for the duration of, the ECMO course. Thirty-five variables were collected, including physiological data, markers of end-organ perfusion, acid-base homeostasis, vasoactive infusions, markers of coagulation, and ECMO-machine factors. The primary outcome was the presence of radiologic evidence of moderate to severe brain injury as established by brain CT or MRI. This information was analyzed by a neural network, and results were compared to a logistic regression model as well as clinician judgement. The neural network model was able to predict brain injury with an Area Under the Curve (AUC) of 0.76, 73% sensitivity, and 80% specificity. Logistic regression had 62% sensitivity and 61% specificity. Clinician judgment had 39% sensitivity and 69% specificity. Sequential feature group masking demonstrated a relatively greater contribution of physiological data and minor contribution of coagulation factors to the model's performance. These findings lay the foundation for further areas of research directions.

Utilizing Machine Learning to Predict Neurological Injury in Venovenous Extracorporeal Membrane Oxygenation Patients: An Extracorporeal Life Support Organization Registry Analysis

“…This is supported by a single-center pediatric ECMO study (68 VV-ECMO and 106 VA-ECMO patients) demonstrating a high prevalence of ABI (51%) where standardized neuromonitoring with neuroimaging protocol was implemented. 16 This study also used artificial intelligence to predict ABI primarily in VA-ECMO patients and demonstrated a better AUC-ROC (0.76) 16 versus our AUC-ROC with the ELSO Registry (0.67 in VV-ECMO patients).…”

Predicting Acute Brain Injury in Venoarterial Extracorporeal Membrane Oxygenation Patients with Tree-Based Machine Learning: Analysis of the Extracorporeal Life Support Organization Registry

“…In fact, a single-center study of 106 VA-ECMO and 68 VV-ECMO pediatric patients using ML to predict CNS ischemia and ICH showed a superior AUC-ROC (0.76) than ours with the ELSO Registry (0.67). ( 36 ) This result may not be surprising given the institution’s rigorous advanced neuroimaging technique to determine ABI and adjudication system by multiple clinicians. Accordingly, their prevalence of ABI (51% in VA/VV-ECMO mixed population) was much higher than ours with the ELSO Registry (7.7% in VA-ECMO and 16.5% in ECPR).…”

The use of machine learning and artificial intelligence within pediatric critical care

“…40 Two common types of neural networks include recurrent neural networks-which can process large amounts of data and "learn" from missed predictions and convolutional neural networks which specialize in transforming imaging data. 40 While several applications have been published, [41][42][43] historic limitations include their "black box" nature and difficulty in determining clinical importance. Recent advances such as detector randomized input sampling or generative adversarial networks have substantially reduced the "black box" nature of neural networks, these techniques have allowed researchers to even determine which portions of an x-ray were important to an algorithm in predicting if an image belonged to a COVID-19 positive or negative patient.…”