Researchers from Osaka University’s Graduate School of Medicine’s Department of Traumatology and Acute Critical Medicine created an AI algorithm that predicted the risk of mortality for patients who had suffered a major injury. Using the records of over 70,000 patients with blunt-force trauma from the Japan Trauma Data Bank from 2013 to 2017, the researchers were able to identify significant factors that could more precisely guide treatment strategies.
In the emergency room, trauma specialists frequently have to make life-or-death decisions. The difficulties include a lack of complete understanding of the factors that indicate the likelihood of poor clinical outcomes, as well as the fact that the body’s own inflammatory and blood clotting changes in response to significant injuries occasionally cause more harm than good. Trauma care clearly necessitates a more thorough and stringent approach.
Researchers at Osaka University Graduate School of Medicine used machine learning algorithms to analyse a database of all trauma cases ever reported in Japan. The patient’s age and type of injury were listed alongside other details. Serum from trauma patients was also subjected to mass spectrometry and proteome analysis at the Osaka hospital. This provided more detailed information on blood markers that could indicate an increase or decrease in a specific protein’s level.
The study’s principal author, Jotaro Tachino, a graduate medical student at Osaka University, said, “Our study has important clinical implications. It can help identify the patients at highest risk who may benefit most from early intervention.”
READ MORE: A Mathematical Model to Counter Malaria
The researchers used a hierarchical clustering analysis to identify the 11 factors most strongly associated with a higher mortality rate. Among these factors were the type and severity of the injury. They also discovered that the most vulnerable patients had an acute inflammatory response, if not excessive inflammation. They also discovered protein markers that strongly suggested that coagulation was downregulated and was linked to poor outcomes.
Hiroshi Ogura, senior author, says, “The method that we used for this project can also be extended to the development of new treatment strategies and therapeutic agents for other medical conditions for which large datasets are available.” This work could significantly improve the distribution of limited ER healthcare resources to save more lives. The research team also hopes that it will provide insight into how to control the inflammatory pathways that can become out of control after traumatic injuries.