Artificial Intelligence and Health Care: Predicting Patient Deterioration
The patient took a turn for the worse. No one wants to hear that particular news about their hospitalized loved one from a health care practitioner, especially when they aren’t expecting it. However, studies, such as one conducted by the National Patient Safety Agency in the UK, estimate that 11 percent of all in-hospital deaths occur because patients’ conditions deteriorated before medical staff recognized and could respond to the signs in a timely manner. Predicting and identifying when a patient’s condition worsens and applying interventions early is vital when someone is in the hospital.
Artificial intelligence has shown great promise for shaping the future of health care. Can machine learning accurately identify the risk factors for patient deterioration and correctly predict its onset?
As part of a medical research partnership with the US Department of Veterans Affairs (VA), the team of scientists and engineers at DeepMind, the artificial intelligence group at Alphabet (Google’s parent company), will work on the global issue of patient deterioration during hospital care.
Patterns from approximately 700,000 historical, de-personalized medical records will be analyzed to develop machine learning algorithms to identify risk factors for a patient’s health worsening and predict its beginning. Initial research will focus on Acute Kidney Injury (AKI), one of the most common conditions associated with patient deterioration and one that is often asymptomatic, yet sudden onset. For example, AKI may occur following routine procedures and operations, such as a hip replacement. AKI can lead to the need for dialysis or even death, but hopefully early detection could lead to better outcomes for patients.
“This project has great potential intelligently to detect and prevent deterioration before patients show serious signs of illness. Speed is vital when a patient is deteriorating: The sooner the right information reaches the right clinician, the sooner the patient can be given the right care,” according to Mustafa Suleyman, co-founder of DeepMind, in the VA announcement.