Artificial intelligence is making critical health care decisions. The sheriff is MIA - POLITICO
2/18/24
This is a challenging read when all "AI" is generally bucketed across all healthcare segments. When we begin a conversation with a healthcare executive, we start with addressing an AI solution's function and operation. We've seen the segments broken down a few different ways and typically talk about AI that runs back-office and moves closer to the bedside.
Most recently we heard Aidoc describe the segments as:
Financial: Billing & Claims, Insurance Eligibility, Revenue Cycle Mgmt.
Operational: Registration Forms, Scheduling, Customer Service.
Documentation & Staff Augmentation: Supply Chain, Facilites, Process Optimization
Clinical: Diagnostic, Monitoring and Care Delivery
Does the FDA need more funding to be able to do more? We believe so.
Should there be a Private-Public partnership to support the testing? We believe so.
But, auditing one time at a regional AI Assurance Lab and through the FDA isn't enough. Without support, local community and critical care hospitals won't be able to adopt new AI tools, because they won't have the resources, bandwidth and specialized knowledge to govern locally.
All healthcare systems, regardless of size or location are experiencing medical professional shortages, staff burn-out and an aging population. AI is one of a few solutions that will allow healthcare providers to do more with less. But it is critical that each hospital has the ability for local assessment and monitoring. Each AI solutions is affected by the data it is run on, how it's deployed and integrated, and more importantly how it is utilized by the medical professional.
“You know as a practicing physician that different environments are different,” Mark Sendak, population health and data science lead at Duke University’s Institute for Health Innovation, told senators at a Finance Committee hearing on artificial intelligence in health care this month. “Every health care organization needs to be able to locally govern AI.”