por
Lauren Dubinsky, Senior Reporter | May 16, 2017
Made possible with the latest
advances in artificial intelligence
After three years of research, a team of scientists from IBM Research and Sutter Health have developed methods that can predict heart failure using data from patients' EHRs.
The team used the most recent advancements in artificial intelligence, including natural language processing, machine learning, and big data analytics to train models to predict patients at risk.
Physicians typically use
CT angiography and stress tests with SPECT myocardial imaging to predict which patients are likely to suffer a heart attack or other adverse cardiovascular event. They also document signs and symptoms of heart failure in the patients' record.
Ad Statistics
Times Displayed: 41009
Times Visited: 663 Reveal Mobi Pro integrates the Reveal 35C detector with SpectralDR technology into a modern mobile X-ray solution. Mobi Pro allows for simultaneous acquisition of conventional & dual-energy images with a single exposure. Contact us for a demo at no cost.
However, patients are usually diagnosed with heart failure after an acute event, according to the research team. By that time, the patient is already hospitalized with advanced-stage disease and potentially irreversible and progressive organ damage.
While researching, the team found crucial insight into the practical trade-offs and types of data that are required to train models. They also developed potential new application methods that could attract more physicians to adopt future models.
For example, they found that only six of the 28 original risk factors contained within the Framingham Heart Failure Signs and Symptoms (FHFSS) were consistent predictors of a future heart failure diagnosis.
They also found that combining other data routinely collected in EHRs, including disease diagnoses, medication prescriptions and lab tests with FHFSS, could be helpful predictors of a patient's risk of heart failure.
The IBM Research and Sutter Health scientists will continue to work together to improve the accuracy and clinical relevance of the models and to test them for clinical use.
They noted that the availability of big data and advancements in cognitive computing could result in "dramatic advances in earlier disease detection." Based on a
MarketsandMarkets report from earlier this month, the use of big data is set to grow in coming years.
The report found that the global artificial intelligence in health care market will grow from $667 million in 2016 to $8 billion by 2022, with big data being one of the main drivers.
