Artificial intelligence alerts doctors seven minutes before danger occurs in operating rooms
Artificial intelligence has quickly become an essential tool in modern operating rooms. In Vienna, researchers from MedUni Vienna and the Vienna University of Technology have developed a model that can detect dangerous drops in blood pressure during surgery up to seven minutes before they occur. By analyzing routine data such as blood pressure, pulse, oxygen saturation, and ventilation, AI gives doctors crucial extra time to intervene and prevent complications.
The new model, known as the Temporal Fusion Transformer, was trained on data from over 73,000 patients at Vienna General Hospital. Its predictions were remarkably accurate, with an average deviation of only 4 mmHg when predicting blood pressure trends. It was able to identify critical drops with an accuracy of over 90 percent. Chief physician Johann Nootzer confirms that these developments demonstrate how artificial intelligence can significantly enhance the safety of operating rooms.
Artificial intelligence is already being used to adjust medications in real-time during surgery. Automated systems control drugs like propofol based on EEG data, circulation, and respiratory parameters, allowing for the maintenance of anesthesia stability while reducing the amount of medication used by up to 30 percent. Notzer points out that this leads to increased patient safety and efficiency of surgical teams.
Artificial intelligence also alleviates the administrative burden on doctors. Tools like Smart Anesthesia Manager take over data documentation and recording almost automatically, allowing doctors to focus more on patients. Future systems are expected to further support medical teams through voice recognition and real-time recommendations.
Training and research programs also benefit. AI-supported simulations analyze how medical teams work and provide personalized feedback, improving learning outcomes by up to 35 percent. Virtual and augmented reality tools provide realistic training environments. In scientific research, AI can analyze complex medical records to identify patterns and potential risks, helping to prevent complications such as low blood pressure during surgery or breathing problems after the procedure.
Professor Judith Martini from the University Hospital of Innsbruck emphasizes the need for responsible use of artificial intelligence. She warned that technology must be based on high-quality data and remain transparent, as there is no place for a 'black box' in science. Ethical standards and data protection must be adhered to in order to maintain patient privacy and community trust.
Despite the rapid progress, experts emphasize that human expertise cannot be replaced. As Martini and Nootzer agree, artificial intelligence will be a strong partner in the medical field—but it will not replace medical responsibility.