Shah Lab

In A ‘Green Button’ For Using Aggregate Patient Data At The Point Of Care we envision a “green button” function within EHRs for clinicians to use aggregate patient data for real time decision making at the point of care. Check out our recently launched Informatics Consult Service that puts this idea in action.

Profiling risk factors for chronic uveitis in juvenile idiopathic arthritis: We report a new association between allergic conditions and chronic uveitis. Covered in Stanford Medicine Mag – Great Medical Mines and in the Wall Street Journal

Androgen Deprivation Therapy and Future Alzheimer's Disease Risk: This study found an association between the use of ADT in the treatment of prostate cancer and an increased risk of Alzheimer's disease in a general population cohort.

Pharmacovigilance using clinical notes: Uses textual clinical notes for detecting single drug–adverse event associations (AUC of 80.4%) and for detecting drug–drug interactions (AUC of 81.5%). Press in Forbes, GigaOM. Our efforts were the focus of an editorial commentary titled Advancing the Science of Pharmacovigilance.

An unsupervised learning method to identify reference intervals from a clinical database: We show that it is possible to use laboratory results and coded diagnoses to learn laboratory test reference intervals from clinical data warehouses.

Postmarket Surveillance of Point-of-Care Glucose Meters through Analysis of Electronic Medical Records: We show that it is possible to assess device accuracy using coincident testing of point-of-care and central laboratory blood glucose measurements in a large cohort of critically ill patients. Editorial in Clinical Chemistry.

Learning Effective Treatment Pathways in Patients With Type 2 Diabetes Treated With Metformin: Where we examine the effectiveness of second-line treatment of type 2 diabetes after initial therapy with metformin via an open collaborative research network

Detecting Chemotherapeutic Skin Adverse Reactions in Social Health Networks Using Deep Learning: Where we identify a cutaneous adverse event of a chemotherapeutic agent by analyzing content in a health social network.

Predicting Diagnoses of Depression: We developed a model that uses electronic medical record (EMR) data for predicting the diagnosis of depression up to 12 months before the diagnosis of depression.

Rapid identification of slow healing wounds: We demonstrate that it is possible to build a model for identifying delayed healing wounds with an Area Under the Curve (AUC) of 0.842 that works across all wound types.

Predicting patient ‘cost blooms’: We develop models that identify new entrants to the upper decile of per capita healthcare expenditures in the next year.