In the past decade or so, researchers have piloted more and more studies exploring the applications of AI in clinical settings. Here are five key studies that have been published in September so far:
- “Assessment of a deep learning model to predict hepatocellular carcinoma in patients with hepatitis C cirrhosis“: Researchers found that recurrent neural network models that use raw, longitudinal EHR data outperform traditional regression models in predicting the risk of hepatocellular carcinoma for patients with hepatitis C-related cirrhosis.
- “Artificial intelligence predictive analytics in the management of outpatient MRI appointment no-shows“: The study found that machine learning predictive analytics performed “moderately well” in identifying patients who are most at risk for missing MRI appointments.
- “Prescriptive analytics for reducing 30-day hospital readmissions after general surgery“: The research team developed predictive machine learning models to forecast 30-day patient readmissions and prescriptive machine learning models to provide more specific recommendations on actionable steps to reduce readmission rates. The models’ prediction accuracy exceeded 87 percent.
- “Machine learning to predict serious bacterial infections in young febrile infants“: The study revealed artificial intelligence can effectively help identify well-appearing infants with a fever who have a low risk for severe bacterial infection.
- “Illuminating the dark spaces of healthcare with ambient intelligence“: Researchers examined the ways hospitals could benefit from gathering information via ambient intelligence, which refers to physical space that is sensitive and responsive to human presence.
More articles on artificial intelligence:
How AI sensors in smart hospitals could reduce fatal medical errors: Stanford study
US Military Health System taps Google Cloud to prototype AI predictive cancer diagnostic tool
The future of AI in healthcare, according to 6 hospital innovation execs
