Machine Learning

Probing Automated Treatment of Urinary Tract Infections for Bias: A Case-Study Where Machine Learning Perpetuates Structural Differences and Racial Disparities

Urinary tract infections (UTI) are a common indication for antibiotic treatment worldwide. Fluoroquinolone antibiotics are widely prescribed for these infections, despite being considered 'second-line' treatments. The use of these therapies …

Clinical Trial of an AI COVID-19 Model to Identify Patients Ready for Discharge

We previously developed a predictive model to help clinicians identify adults with COVID-19 who may be ready for discharge given their low risk of adverse events. This algorithm was integrated into our EHR to display regularly updated risk scores in …

Expanding the Reach of Structured EHR Data with Clinical Notes: Improving End-of-Life Prediction

Appropriate treatment decisions and end-of-life planning for patients with serious, life-limiting disease rely on accurate prognosis estimates. Many existing methods rely on structured electronic health record data which may limit generalizability …

Supporting Acute Advance Care Planning with Precise, Timely Mortality Risk Predictions

Patients with serious, life-limiting disease benefit from end-of-life conversations, goal setting, and palliative care. Hospitalized patients at high risk of near-term death are likely to benefit from such interventions. As NYU Langone Health …

A Validated, Real-Time Prediction Model for Favorable Outcomes in Hospitalized COVID-19 Patients

The COVID-19 pandemic has challenged front-line clinical decision-making, leading to numerous published prognostic tools. However, few models have been prospectively validated and none report implementation in practice. Here, we use 3345 …

Development, Implementation, and Prospective Validation of a Model to Predict 60-Day End-of-Life in Hospitalized Adults Upon Admission at Three Sites

**Background**: Automated systems that use machine learning to estimate a patient’s risk of death are being developed to influence care. There remains sparse transparent reporting of model generalizability in different subpopulations especially for …

Estimating Real World Performance of a Predictive Model: A Case-Study in Predicting End-of-Life

**Objective**: One primary consideration when developing predictive models is downstream effects on future model performance. We conduct experiments to quantify the effects of experimental design choices, namely cohort selection and internal …