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

We consider the problem of predicting how the likelihood of an outcome ofinterest for a patient changes over time as we observe more of the patientdata. To solve this problem, we propose a supervised contrastive learningframework that learns an embedding representation for each time step of apatient time series. Our framework learns the embedding space to have thefollowing properties: (1) nearby points in the embedding space have similarpredicted class probabilities, (2) adjacent time steps of the same time seriesmap to nearby points in the embedding space, and (3) time steps with verydifferent raw feature vectors map to far apart regions of the embedding space.To achieve property (3), we employ a nearest neighbor pairing mechanism in theraw feature space. This mechanism also serves as an alternative to dataaugmentation, a key ingredient of contrastive learning, which lacks a standardprocedure that is adequately realistic for clinical tabular data, to ourknowledge. We demonstrate that our approach outperforms state-of-the-artbaselines in predicting mortality of septic patients (MIMIC-III dataset) andtracking progression of cognitive impairment (ADNI dataset). Our method alsoconsistently recovers the correct synthetic dataset embedding structure acrossexperiments, a feat not achieved by baselines. Our ablation experiments showthe pivotal role of our nearest neighbor pairing.

Temporal Supervised Contrastive Learning for Modeling Patient Risk Progression