Dr. Christopher Kellner · NYC Neurosurgeon
Verified account This account is verified. Learn more@chriskellnerMD
Cerebrovascular & Stroke Neurosurgeon @MountSinaiNeuro · Assoc. Professor & Researcher @MountSinai · Chair @cvsection · Co-Director, Precision Recovery
1.4K Following 3.7K Followers 415 Posts
Latest posts 5
Some thoughts on taking call.
0:24
Collaboration when problem-solving is a key lesson I've learned from my mentors, @ConnollyES, J Mocco, Alejandro Berenstein, @joshbedersonmd. What are lessons you'll never forget from your training?
0:24
0:31
AI in healthcare will ultimately be judged not by what it can demonstrate in a lab, but by whether we can deploy it safely and effectively in the real world.
A new study in @NEJM Catalyst describes how leaders at @MountSinaiNeuro evaluated, vetted, integrated, and deployed an AI-powered digital triage platform across our health system.
The results are impressive:
• ~22,000 completed AI-guided triage encounters
• 80% completion among patients who initiated triage
• 71% of use occurred outside normal business hours
• 88–96% concordance with physician triage
• System Usability Score of 85.5 (“excellent”)
• Zero reported patient-safety incidents
Perhaps most interestingly, among patients who already thought they knew where they should seek care, 62% had an initial plan that differed from the AI triage recommendation, and in 79% of those cases, the system recommended more urgent care.
This is where AI can become transformative: not simply answering questions, but creating a 24/7 digital front door that helps patients get the right care, in the right place, at the right time while making better use of limited healthcare resources.
Congratulations to Bilal Naved, Nicholas Gavin, @girish_nadkarni, Robert Freeman, [and the entire multidisciplinary @MountSinaiNYC team for demonstrating what thoughtful, clinically grounded AI implementation can look like at scale.
Read more ⬇️
catalyst.nejm.org
#ArtificialIntelligence #HealthcareAI
A new study in @NEJM Catalyst describes how leaders at @MountSinaiNeuro evaluated, vetted, integrated, and deployed an AI-powered digital triage platform across our health system.
The results are impressive:
• ~22,000 completed AI-guided triage encounters
• 80% completion among patients who initiated triage
• 71% of use occurred outside normal business hours
• 88–96% concordance with physician triage
• System Usability Score of 85.5 (“excellent”)
• Zero reported patient-safety incidents
Perhaps most interestingly, among patients who already thought they knew where they should seek care, 62% had an initial plan that differed from the AI triage recommendation, and in 79% of those cases, the system recommended more urgent care.
This is where AI can become transformative: not simply answering questions, but creating a 24/7 digital front door that helps patients get the right care, in the right place, at the right time while making better use of limited healthcare resources.
Congratulations to Bilal Naved, Nicholas Gavin, @girish_nadkarni, Robert Freeman, [and the entire multidisciplinary @MountSinaiNYC team for demonstrating what thoughtful, clinically grounded AI implementation can look like at scale.
Read more ⬇️
catalyst.nejm.org
#ArtificialIntelligence #HealthcareAI