English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Hundreds of millions of surgical procedures take place annually across theworld, which generate a prevalent type of electronic health record (EHR) datacomprising time series physiological signals. Here, we present a transferableembedding method (i.e., a method to transform time series signals into inputfeatures for predictive machine learning models) named PHASE (PHysiologicAlSignal Embeddings) that enables us to more accurately forecast adverse surgicaloutcomes based on physiological signals. We evaluate PHASE on minute-by-minuteEHR data of more than 50,000 surgeries from two operating room (OR) datasetsand patient stays in an intensive care unit (ICU) dataset. PHASE outperformsother state-of-the-art approaches, such as long-short term memory networkstrained on raw data and gradient boosted trees trained on handcrafted features,in predicting five distinct outcomes: hypoxemia, hypocapnia, hypotension,hypertension, and phenylephrine administration. In a transfer learning settingwhere we train embedding models in one dataset then embed signals and predictadverse events in unseen data, PHASE achieves significantly higher predictionaccuracy at lower computational cost compared to conventional approaches.Finally, given the importance of understanding models in clinical applicationswe demonstrate that PHASE is explainable and validate our predictive modelsusing local feature attribution methods.

Forecasting adverse surgical events using self-supervised transfer learning for physiological signals