AI system could boost survival chances for sepsis patients

A newly-developed artificial intelligence (AI) system may be the answer for treating and preventing fatal outcomes for sepsis patients, according to a recent study.

An AI Clinician system matched or made better treatment decisions than human physicians 98 percent of the time, according to the study. Patient mortality was lowest when human doctors treatments matched the AI system’s suggestion. When physician’s decisions were different from the system, patients chances of survival were reduced.

“The AI Clinician was able to ‘learn’ from far more patients than any doctor could see in a lifetime. It has learnt from 100,000 patients and ‘remembered’ them all equally, whereas doctors are always susceptible to recall bias, where they particularly remember recent cases or unusual cases,” Anthony Gordon, MD, FRCA, FFICM, senior author of the study and Imperial College London professor, said in a statement.

The system’s abilities were detailed in a study conducted by researchers with the Imperial College London and published in Nature Medicine.

The AI Clinician system analyzed the medical records of 96,000 American patients with sepsis, a potentially fatal complication from a blood infection. By using reinforcement learning, the system was able to assess each patient’s case and decide the best strategies for keeping that patient alive.  

With the system proven to be a reliable treatment option for sepsis patients, researchers now hope to test its abilities in UK hospitals.

“The only way for any technology to help a patient is to turn it into a product that doctors and hospitals can prescribe, therefore we are seeking to [commercialize],” study author A. Aldo Faisal, PhD, said in a statement.