Detection of Intracranial Hypertension using Deep Learning

Abstract

Intracranial Hypertension, a disorder characterized by elevated pressure in the brain, is typically monitored in neurointensive care and diagnosed only after elevation has occurred. This reaction-based method of treatment leaves patients at higher risk of additional complications in case of misdetection. The detection of intracranial hypertension has been the subject of many recent studies in an attempt to accurately characterize the causes of hypertension, specifically examining waveform morphology. We investigate the use of Deep Learning, a hierarchical form of machine learning, to model the relationship between hypertension and waveform morphology, giving us the ability to accurately detect presence hypertension. Data from 60 patients, showing intracranial pressure levels over a half hour time span, was used to evaluate the model. We divided each patient’s recording into average normalized beats over 30 sec segments, assigning each beat a label of high (i.e. greater than 15 mmHg) or low intracranial pressure. The model was tested to predict the presence of elevated intracranial pressure. The algorithm was found to be 92.05± 2.25% accurate in detecting intracranial hypertension on our dataset.

Intracranial Pressure Monitoring Signals After Traumatic Brain Injury: A Narrative Overview and Conceptual Data Science Framework

“…A three-layer Convolutional Neural Network (CNN) was trained on MOCAIP data and tested with a 3-fold cross-validation; its accuracy reached 87.19% ( 82 ). An autoencoder was used to reconstruct the features for pre-training enhancement and increased the accuracy further to 92.05% ( 81 ).…”

“…Control episodes are selected either from segments at least 1 h prior to ICP elevation from patients with at least one episode of IH, or segments from patients without a single episode of IH (80). If only IH and control episodes are compared, the problem would be IH detection (81,87), whereas IH prediction compares the pre-IH episodes and IH episodes.…”

Künstliche Intelligenz in der Neurointensivmedizin

“…Auch höhere Frequenzen können mit neuronalen Netzen und Autoencodern klassifiziert werden (400-Hz-Abtastrate, n = 60; [ 40 ]). Bei diesen wellenförmigen Daten besteht das Problem, dass aufgrund der Menge nicht alle Daten gleichzeitig genutzt werden können und es sich meistens um kleine Patientenkohorten handelt.…”

Forecasting intracranial hypertension using multi-scale waveform metrics

“…To obviate the need for explicit feature engineering on historical time series, deep learning architectures have been proposed, which detect intracranial hypertension from the raw pulse waveform [49]. Simpler dimensionality reduction approaches, such as principal component analysis, have also been used to find non-correlated features [50] that describe ICH.…”