Two artificial intelligence-based algorithms predicted the probability of a patient dying in the ICU within 30 days of traumatic brain injury with accuracies up to 81% and 84%, in this study from Finland.
The first algorithm is based on objective monitor data, and the second one includes data on the level of consciousness.
Because traumatic brain injury patients are unconscious, it is challenging to accurately monitor the patient's condition in the ICU, where many variables are continuously monitored (eg, intracranial pressure, mean arterial pressure, and cerebral perfusion pressure).
Just one variable, such as intracranial pressure, can yield hundreds of thousands of data points per day, making it impossible for the human brain to comprehend the resulting millions of daily collected data points.
This is why researchers at Helsinki University Hospital developed the algorithms. A model like this has not been presented before and reflects how and into what direction patient care in the ICU is evolving, the authors say.