Data scientists and neurotrauma surgeons from the University of Pittsburgh have successfully used automated brain scans and machine learning to inform outcomes in patients with severe traumatic brain injuries (TBI).
The team’s advanced machine-learning prognostic model analyses patient clinical data and brain scans to predict survival and recovery six months post-injury. Developed for use in patients at risk of having life-saving care removed when on life support, the team hope that the tool could identify patients likely to make a ‘meaningful recovery’.
Co-senior author David Okonkwo, Professor of Neurological Surgery, said:
“The majority of people who survive a critical period in an acute care setting make a meaningful recovery—which further underscores the need to identify patients who are more likely to recover.”
Researchers developed a custom artificial intelligence model to processed multiple brain scans from each patient. When combined with an estimate of coma severity and information about the patient’s vital signs, blood tests and heart function, the model accurately predicted patients’ risk of death and poor outcomes at six months following injury. It was tested on the records of over 700 patients across the US.
Shandong Wu, Associate Professor of radiology, bioengineering and biomedical informatics said:
“There is a great need for better quantitative tools to help intensive care neurologists and neurosurgeons make more informed decisions for patients in critical condition.
This collaboration gave us an opportunity to use our expertise in machine learning and medical imaging to develop models that use both brain imaging and other clinically available data to address an unmet need.”