David Alexander Sontag

This paper introduces scSet, a diffusion-based autoencoder that learns patient-level representations from single-cell data, outperforming existing methods in clinical prediction with limited labeled data.

An evaluation of scBERT and scGPT on cell type annotation against L1-regularized logistic regression, with ablations of pre-training and tests of fine-tuning stability.

The paper evaluates the Angus ICD9-CM sepsis abstraction criteria, finding high specificity but moderate sensitivity in identifying suspected infections in emergency department patients.

This paper demonstrates a contextual autocomplete system that improves structured data capture and quality in emergency department documentation using machine learning.

The paper develops and validates a machine learning algorithm to link EMS electronic patient care reports to emergency department records, achieving high sensitivity and positive predictive value.