Artificial Intelligence Can Improve Patient Care
Modeling shows that computers can improve medical decisions and outcomes at lower costs
TUESDAY, Feb. 19 (HealthDay News) -- Use of an artificial intelligence (AI) framework can improve patient outcomes at one-third of the costs of the current standard of care, according to a study published online Jan. 2 in Artificial Intelligence in Medicine.
Casey C. Bennett, from the Centerstone Research Institute in Nashville, Tenn., and Kris Hauser, Ph.D., from Indiana University in Bloomington, have developed a general purpose (non-disease-specific) computational/AI framework that combines Markov decision processes and dynamic decision networks to learn from clinical data and develop complex plans via simulation of alternative sequential decision paths while capturing the interactions of the various components of the health care system. The framework was evaluated using real electronic health record data.
The researchers found that the approach was feasible, with the AI framework outperforming the current treatment-as-usual (TAU) case-rate/fee-for-service models of health care. For AI versus TAU, the cost per unit of outcome change was $189 versus $497, where lower was considered optimal. Additionally, the AI approach could obtain a 30 to 35 percent increase in patient outcomes. This advantage could be further enhanced with tweaks to the AI model parameters, making obtaining approximately 50 percent more improvement (outcome change) for roughly half the costs feasible.
"Given careful design and problem formulation, an AI simulation framework can approximate optimal decisions even in complex and uncertain environments," the authors write.