Artificial intelligence could help screen trauma patients who misuse alcohol
By Jose Florez
February 21, 2019
Researchers have found an artificial intelligence method that may soon be used to identify trauma patients who have misused alcohol.
Many trauma cases are associated with alcohol misuse; screening, brief intervention, and referral to treatment (SBIRT) may help lessen alcohol consumption, reduce injury recurrence by 50 percent and decrease DUI incidents.
“The brief intervention typically includes providing information on the link between drinking and injury, encouraging patients to think about how drinking may have contributed to their injuries and giving professional advice about the need to reduce risk by cutting down or quitting drinking,” a news release reads.
Today, screening methods commonly used, like the 10-item Alcohol Use Disorders Identification Test (AUDIT), have drawbacks. Answers pertaining to alcohol use may not be honest or staff may be unavailable to administer the test. And so, the use of artificial intelligence to test for alcohol misuse may overpass these issues.
Researchers at Loyola Medicine and Loyola University Chicago looked through a wealth of electronic health records utilizing natural language processing and machine learning. The AI technique uses computational methods, which helps computers understand human language.
In the study, researchers probed 1,422 adult patients admitted to Loyola’s Level 1 trauma center over a period of three years. Researchers also collected 91,045 clinician notes in electronic health records, containing 16,091 medical concepts used to predict alcohol misuse. Signs of intoxication, alcoholism, liver imaging, and thiamine deficiency are among the concepts.
Utilizing the clinical Text Analysis and Knowledge Extraction System, researchers took linguistic processing of clinical notes, with the binary classification of alcohol misuse as the primary analysis.
Researchers concluded that natural language processing has adequate predictive validity for identifying alcohol misuse in the trauma setting. In most cases, about 78 percent, the AI technique successfully differentiated between patients who misused alcohol and non-drinkers.
The findings, according to researchers, “provides an automated approach to potential overcome staffing and patient barriers for SBIRT programs at trauma centers.”
The study was published in the Journal of the American Medical Informatics Association.