Researchers at the Department of Energy’s Pacific Northwest National Laboratory (PNNL), https://www.pnnl.gov Stanford University, Virginia Tech and John Snow Labs, have developed TransMED, a first of its kind Artificial Intelligence (AI) prediction tool to address emerging or rare diseases.
The multi-institutional team developed TransMED to analyze data from existing diseases to predict outcomes for emerging diseases. Early results indicate that TransMED outperforms current patient outcome prediction models, particularly for rarer outcomes. The researchers partly attribute this to TransMED’s ability to scrutinize a wide variety of medical information, including other respiratory diseases.
TransMED studies nearly all types of EHR data such as medical conditions, drugs, procedures, laboratory measurements, and information from clinical notes. By using this holistic view of the patients, TransMED is able to make predictions much the same way a clinician would.
TransMED is also able to transfer learning. Transfer learning works by having a machine learning model work on solving a problem where a lot of data exists. The model than transfers this knowledge to solve similar problems.
In the case of TransMED, researchers trained the model on known severe respiratory disease patient outcomes and applied that knowledge to predicting COVID-19 outcomes. TransMED when given a patient’s recent medical history can predict a patient’s need for ventilators, or other rare outcomes 5 to 7 days out into the future.
The scientists explained that the application of AI in real world healthcare settings is in its infancy but this work is a promising first step towards building a useful model for predicting patient outcomes. Although TransMED has yet to be tested in a clinical setting, the technology offers an encouraging glimpse into the future of healthcare.