Researchers from Stanford University have created a data set containing information about how different terms and concepts co-occur in 20 million clinical notes and patient narratives spanning 19 years. Using this data set, scientists can estimate the probability that a patient with a certain condition will take a certain drug or use a certain device. The data set’s creators hope it will help researchers in a wide range of applications, including outcome prediction for medical treatments and analyzing patterns of comorbidity, when patients have two or more conditions.
Data Set Maps Relationship Between Drugs, Diseases, Devices, and Procedures
