Educators in Medicine,
In this newsletter, we continue our journey through the fundamentals of AI, its applications in medicine, and its transformative role in faculty development and education. Let’s dive into learning.
AI 101 - A Primer on Artificial Intelligence
On Clinical Decision Support (CDS)
Glass Health recently announced the introduction of Glass Odyssey, their groundbreaking AI Clinical Decision Support (CDS) system, which now seamlessly integrates with hospital EHRs. This innovation allows Glass Health users to access clinical decision support through every step of the patient encounter. Their announcement video showcases how AI can expand on history taking and suggest orders for the physician user, illustrating the potential to enhance clinical workflows and improve patient care.
The potential of tools like Glass Odyssey is vast. By integrating CDS directly into EHRs, clinicians can receive real-time, evidence-based recommendations that support decision-making processes. This can lead to more accurate diagnoses, optimized treatment plans, and improved patient outcomes. Additionally, the system's ability to continuously learn and adapt from new data means that its recommendations will become increasingly precise over time. However, I think there are potential flaws to consider. Always, as an educator, I have thoughts about the reliance on AI for decision-making, which may lead to overdependence and a reduction in clinicians' critical thinking skills.
I went to their website to try it out, and found they have an interesting tool that is free in a limited version.
You can “consult” Glass to ask it a clinical question, or get a differential. They also have a tool for trainees (and attendings) to share clinical notes, pearls, and thought processes. Seems like a great resource if they can gain the traction and buy-in from users, in my limited experience though, seems like they’re trying to do too many things? At least in my experience, products should aim to do one thing well before scaling horizontally.
AI in Education
The Role of Interviews in Medical School and Residency Applications
Interviewing is a noisy prediction problem, as Erik Bern eloquently discusses in his blog post on the subject. The insights from Sarah Gebauer's MLforMDs blog led me to explore this further. The traditional interview process often fails to predict future performance accurately due to its subjective nature and inherent biases. This raises the question: as a residency program director why don't we make our interviews more critical thinking-oriented?
Incorporating assessments into interviews could provide a more accurate gauge of a candidate's potential. In my mind, the idea of using AI to 'listen' to interviews is also intriguing. Medicine is looking to leverage voice as a diagnostic/prognostic factor - why not do the same for job interviews? AI could analyze responses for indicators of critical thinking, problem-solving skills, and emotional intelligence, potentially improving the selection process.
We are working on designing our interview processes for this fall at our program. What are your best practices? How can this space be disrupted to make it more impactful? Comment below!
What Can I Do Now?
AI Diagnosing Acute Otitis Media in Children
The article "Development and Validation of an Automated Classifier to Diagnose Acute Otitis Media in Children" presents a fascinating development in AI's role in diagnosing conditions. The AI classifier for acute otitis media could significantly reduce the guesswork involved in diagnosing this common condition, leading to better antibiotic stewardship and potentially altering the role of physicians.
I am curious, would this technology lead to more accurate diagnoses and reduce unnecessary antibiotic prescriptions? Are we watching a shift in the physician's role towards more complex decision-making, leaving routine diagnoses to AI? On the flip side, would concerned parents lose trust if their children don't receive prescriptions as frequently? This is an area ripe for discussion, and I'm curious to hear readers' thoughts on how this technology could impact pediatric care.
Chief Health AI Officer — An Emerging Role for an Emerging Technology
I was excited to read the article "Chief Health AI Officer — An Emerging Role for an Emerging Technology" in NEJM AI. Just as health systems were forced to move towards the CMIO role, the Chief Health AI Officer is a role that will follow closely behind. Informatics is too broad a field to be focused on one space, so I imagine this position will work shoulder to shoulder with the CMIO. Doctors and health systems need an ambassador and translator for this language. Across the country, physicians want to use AI in their practice but few know enough about implementation. I see this role should be flourishing in the next 12-24 months, as health systems don't want to fall behind. The CHAIO will be crucial in ensuring that AI technologies are implemented effectively and ethically, bridging the gap between clinical practice and advanced technology.
As always - get in touch and let me know your thoughts!
Thank you for joining us on this adventure. Stay tuned for more AI insights, best practices, and more future editions of AI+MedEd.
For education and innovation,
Karim
Share this with someone - have them sign up here.