Hospitals are using artificial intelligence in emergency rooms, inpatient wards and intensive care units to help identify and treat patients at highest risk for sepsis and an impending cardiac arrest or stroke, The Wall Street Journal reported April 10.
Here are five AI projects the health systems' have created and deployed to help patients, according to the Journal:
- AI and sepsis response: Durham, N.C.-based Duke University Hospital created Sepsis Watch, a deep learning model that provides an early warning system within Duke Health for patients who are at risk of sepsis. The model can predict sepsis quickly and accurately with data from the hospital's own patient records.
- Advanced Alert: Oakland, Calif.-based Kaiser Permanente developed a predictive model called Advance Alert Monitor, that scans patient data continuously, assigning scores that predict the risk of transfer to the ICU or death.
- Sepsis prediction: HCA Healthcare developed a predictive algorithm, Spot—for Sepsis Prediction and Optimization of Therapy, to continuously monitor vital signs, lab results, nursing reports and other data — firing an alert directly to nurses at the moment signals converge that indicate impending sepsis. The algorithm was able to detect Sepsis six hours earlier and more accurately than clinicians.
- Nate: HCA also developed the Next-Gen Analytics for Treatment and Efficiency, which uses machine learning to detect other critical or life-threatening conditions such as shock in trauma patients, complications after surgery and early signs of deterioration in all patients.
- Cancer screening: Danville, Pa.-based Geisinger Health System partnered with Medial EarlySign to identify patients overdue for a colorectal-cancer screening. They also used a machine-learning algorithm to flag those at higher risk and were able to schedule screenings for 68.1 percent of the patients flagged.