– Scientists at Massachusetts General Hospital (MGH) are utilizing machine learning and natural language processing tools in their EHRs to diagnose the chance that a patient will eventually receive a dementia or Alzheimer’s diagnosis, according to a new report.
Members of MGH’s Center for Quantitative Health, the Harvard T.H. Chan School of Publish Health, and the Harvard Brain Tissue Resource Center developed an algorithm based on machine learning to first build a list of key clinical terms associated with cognitive symptoms identified by clinical experts.
After creating the algorithm, the scientists used national language processing (NLP) to find those clinical terms within the EHR. Once that task was completed, researchers took the data and estimated the chances that the patient would develop dementia.
Researchers found that 2.4 percent of patients developed dementia over the next eight years of a follow up appointment. The team studied data on 267,855 patients admitted to one of two hospital systems.
“The most exciting thing is that we are able to predict risk of new dementia diagnosis up to eight years in advance,” said Thomas McCoy, Jr., MD, the first author of the report that was published in Alzheimer’s & Dementia.
Currently, Alzheimer’s affects more than 5.5 million US citizens and this number is believed to increase in the future. Finding an early diagnosis of dementia is key to finding effective treatments and improving care for patients.
Using EHRs rather than the current early detection tools saves the hospital money and extra steps to collect data. The new tool that is embedded into the EHR will hopefully accelerate research and makes better use of auto-generated data during care.
“This method was originally developed as a general ‘cognitive symptom’ assessment tool. But we were able to apply it to answer particular questions about dementia,” explained McCoy. “This study contributes to a growing body of work on the usefulness of calculating broad symptom burden scores across neuropsychiatric conditions.”
McCoy and his team of researchers utilized the general cognitive symptom tool in past studies to predict suicide and accidental death. The team also believes that the tool can be used to answer more questions revolving around other brain diseases.
“We need to detect dementia as early as possible to have the best opportunity to bend the curve,” said Roy Perlis, MD, senior author of the study and director of the MGH Center for Quantitative Health. “With this approach we are using clinical data that is already in the health record, that doesn’t require anything but a willingness to make use of the data.”
Overall, the researchers believe that this tool can be utilized to improve and accelerate dementia research.
“This approach could be duplicated around the world, giving us more data and more evidence for trials looking at potential treatments,” said Rudolph Tanzi, PhD, vice-chair of Neurology and co-director of the MGH McCance Center for Brain Health at the MGH Institute for Neurodegenerative Diseases.
In a 2019 US News & World Report, MGH was named the No. 2 hospital on its list of “America’s Best Hospitals.” The hospital is also home to the MGH Research Institute, which is the largest hospital-based research program in the US, with over 8,500 researchers working with over a $1 billion research budget.
In an effort to increase interoperability and health information exchange, MGH implemented Epic Systems EHR back in 2016.
The implementation of Epic allowed MGH to connect to other local healthcare systems such as Boston Medical Center, Lahey Health, Southcoast Health System, and UMass Memorial Health Care.
Credit: Google News