SAN FRANCISCO—Intraoperative hypotension (IOH) can be successfully predicted using deep neural networks, which analyze continuous waveforms of invasive arterial blood pressure, new research has concluded.
A large intraoperative data set of more than 17,000 waveforms was used. The study found that the novel deep neural network model achieved significantly better performance than a benchmark model, which predicted IOH with only a single value of invasive arterial blood pressure.
“Recent