AI Enabled Healthcare

For anyone who knows me, the thought of me writing about anything regarding technology is comical. I can barely work my phone! My interest, however, in Artificial Intelligence (AI) in medicine was piqued after listening to a fantastic lecture on AI at ACOI 2024. The lecture was provided by Charlene LePane, DO, MBA, MSPH, FACOI, FACG, who serves as Chief Medical Information Officer at AdventHealth-Central Florida Division and is an ACOI board member extraordinaire. I am amazed at what already exists and what is on the horizon. I would highly recommend Dr. LePane's AI and Medicine lecture. You have until December 31 to review this and all the recorded sessions from ACOI 2024.  

After the Annual Convention, I wanted to learn more. My next step was to recruit Virginia Irwin-Scott, DO, MBA, CPE, FASN, FACOI, National Director of Kidney Care at ChenMed, to answer a few of my questions. I hope you find this helpful in your AI journey!

Why do you think it is important for physicians to learn, embrace, and utilize AI?

Dr. Irwin-Scott: Medicine and our relationships with patients stand to be profoundly enhanced by AI. The recent collapse of Forward Health—a $650 million venture by former Google and Uber executives aiming to "disrupt healthcare"—serves as a cautionary tale. Their ambitious plan for a self-serve, AI-driven CarePod kiosk, envisioned as an “autonomous doctor’s office,” failed because it overlooked a vital element: the human experience. On the September 30, 2024 episode of the Acquired podcast, hosts David Rosenthal and Ben Gilbert interviewed Spotify CEO Daniel Ek and Meta CEO Mark Zuckerberg, both of whom credited their success to balancing technological innovation with a deep focus on human connection. They emphasized that the human experience is just as critical as technology—a lesson equally applicable to the future of medicine.  

In the book ChatGPT, MD, Dr. Robert Pearl examines how generative AI can transform healthcare by improving diagnostics, reducing errors, and empowering both patients and doctors with accessible medical knowledge. He also addresses challenges like bias, privacy, and job displacement, presenting a balanced vision for AI's integration into medicine.  

ChatGPT wants physicians to understand AI will transform healthcare by improving diagnostic accuracy, enabling personalized medicine, and streamlining administrative tasks, which enhances efficiency and reduces burnout. Familiarity with AI helps doctors critically evaluate its recommendations, address ethical concerns like bias and data privacy, and maintain their leadership roles in patient care. By leveraging AI knowledge, physicians can foster trust with patients and actively shape their future to meet real-world clinical needs.

For those who do not yet use AI, how do you recommend they get started?

Dr. Irwin-Scott:

1. Educate Yourself on the Basics of AI

• Foundational Understanding: Learn the basics of AI, machine learning, natural language processing, and deep learning. Free online resources like Coursera, Khan Academy, or AI-specific modules (e.g., Stanford’s free AI for Medicine course) are excellent starting points. Certificate courses with universities start with foundational understanding and build over 6-10 weeks (MIT, Cornell, Harvard).
• Focus on Applications in Medicine: Seek examples of AI applications in your specialty (e.g., radiology, pathology, cardiology, nephrology). Publications, webinars, and conferences often cover emerging AI tools.

2. Stay Informed About AI in Healthcare

• Read Journals and News: Follow reputable medical journals (e.g., JAMA, NEJM) for articles about AI applications. LinkedIn is a great source of AI medical learning.
• Join AI-Focused Groups: Many professional societies (e.g., AMA, RSNA, ACC) have sections dedicated to AI in healthcare. Joining these groups offers access to curated resources and discussions.

3. Experiment with Entry-Level AI Tools

• Clinical Support Tools: Explore tools like symptom checkers, drug interaction platforms, or clinical decision support systems (e.g., IBM Watson, UpToDate’s AI features). Start using an LLM, ChatGPT or other LLM’s (Gemini, Copilot, Claude, Perplexity Grok) -practice asking the question for precision answers. Ask more than one LLM, ask for citations and review the citations. Look for similarities and differences. You are part of the training for these models with your queries.  
• Administrative AI: Try scheduling or documentation assistants.
• Image Analysis AI (for imaging-based fields): Test software that supports diagnostic interpretation (e.g., Aidoc, Zebra Medical Vision).

4. Attend Workshops and Conferences

• Many medical conferences now feature AI tracks. Attending these sessions can provide hands-on experience and allow interaction with experts.
• AI-Focused Conferences: RISE, ABAIM, College of AI in Medicine.

5. Collaborate with Tech-Savvy Colleagues

• Partner with colleagues already using AI tools. Ask for demonstrations or guidance in integrating AI into your practice.
• Leverage Hospital/Institutional Resources: Many larger institutions have AI innovation labs or partnerships with tech companies.

6. Advocate for Ethical and Practical Use

• Learn about the limitations of AI and the importance of data privacy, security, and reducing bias in AI systems.
• Understand local and international regulations to ensure compliant AI use.

7. Join Online Communities or Courses

• ABAIM: CME and ability for board certification, office hours monthly.
• College of AI in Medicine: Founded in 2024 and growing.  

8. Reflect and Adapt

• After experimenting with AI tools, assess their impact on your workflow and patient care. Adjust your approach as needed and continue to explore and grow.  

What would be a way to incorporate AI into daily office practice now?

Dr. Irwin-Scott: The first step to incorporate AI in an office practice is identifying a specific problem or area where AI can provide the most immediate benefit to streamline operations, improve patient care, or reduce physician workload. Integrating AI-augmented transcription tools into Electronic Health Records (EHR) can significantly enhance the efficiency and accuracy of clinical documentation. When selecting an AI transcription tool, consider factors such as compatibility with your existing EHR system, data security measures, ease of use, and the ability to accurately capture medical terminology. Implementing these tools can lead to more efficient documentation processes, allowing healthcare providers to focus more on patient care.

Tomorrow’s office will utilize an AI agent, a software program that uses machine learning and automation to perform tasks, make decisions, and assist in workflows, often mimicking human intelligence. AI agents will revolutionize office-based physician practices by automating documentation, optimizing workflows, supporting diagnostics, enhancing patient communication, and providing personalized treatment recommendations. These agents will streamline administrative and clinical tasks, allowing physicians to focus more on patient care and less on routine processes.

What would be a way to incorporate AI into daily hospital practice now?

Dr. Irwin-Scott: AI is being utilized in nearly all health systems today to varying degrees, with some organizations leading the way and others just beginning to explore its potential. While much of AI's impact is integrated into process management, many layers of AI work quietly behind the scenes to transform care delivery. In the ER, AI is streamlining triage, predicting critical deterioration, and optimizing resource use during busy times. In the ICU, it monitors patient vitals in real-time, detects early signs of complications like sepsis, and provides predictive insights to guide care. Radiology leverages AI for faster and more accurate detection of abnormalities such as tumors or fractures, while in pathology, it enhances diagnostic precision and automates time-consuming tasks like tissue analysis.

Looking to the future, what are you most excited about regarding how AI may assist a physician’s practice and ultimately our patients?

Dr. Irwin-Scott: As we look to the future, there’s so much to be excited about when it comes to how AI is transforming healthcare. For starters, AI is making diagnostics faster and more accurate, helping us catch conditions earlier and provide better outcomes for our patients. It’s also opening doors to truly personalized medicine, where treatment plans can be tailored to everyone based on their unique data. This is the set up for precision medicine that incorporates the human experience. AI is taking over time-consuming administrative tasks, allowing us to spend more time where it matters most caring for patients. The possibilities are endless, and as these tools continue to evolve, I believe AI will become an invaluable partner in improving how we deliver care whether it is in the office or for an entire population.    

Do you have any reservations about utilizing AI in healthcare?

Dr. Irwin-Scott: While AI holds incredible promise for healthcare, it's important to recognize and address potential concerns. Issues like data privacy, algorithmic bias, and the risk of over-reliance on AI are critical to consider. Additionally, challenges with integration into workflows, regulatory hurdles, and ensuring transparency in how AI systems make decisions must be tackled. By approaching AI thoughtfully and prioritizing both ethical use and patient trust, we can harness its benefits while safeguarding the human aspects of medicine.  

In summary, AI is already becoming a part of our daily routines in healthcare, with many systems using it for decision support, imaging, and streamlining administrative tasks. As these technologies improve and integrate more seamlessly, their role will expand into areas like personalized medicine and predictive analytics. I believe that within the next 3-5 years, AI will be an essential tool we rely on every day, helping us deliver more precise, efficient, and patient-focused care. The future is exciting, and AI will be a key partner in shaping it. Here are a few resources Dr. Irwin-Scott shared, which you might find helpful in learning more about AI:

1. Philips Healthcare AI - https://www.philips.com/a-w/about/news/archive/features/2022/20221124-10-real-world-examples-of-ai-in-healthcare.html
2. Coursera - AI in Healthcare: https://www.coursera.org/articles/ai-in-health-care
3. McKinsey - Transforming Healthcare with AI: https://www.mckinsey.com/industries/healthcare/our-insights/transforming-healthcare-with-ai
4. Cleveland Clinic - AI in Healthcare: https://health.clevelandclinic.org/ai-in-healthcare
5. Le Monde - AI and Mental Health: https://www.lemonde.fr/en/pixels/article/2024/08/18/how-ai-is-shaking-up-the-mental-health-community-rather-than-pay-for-another-session-i-d-go-on-chatgpt_6717874_13.html  
6. Wikipedia - MIT Jameel Clinic -  https://en.wikipedia.org/wiki/MIT_Jameel_Clinic
7. American Medical Association (AMA) on AI - https://www.ama-assn.org/practice-management/digital/ama-future-health-emerging-landscape-augmented-intelligence-health-care

Thank you to Dr. Irwin-Scott for sharing her knowledge about AI.  

I hope you have a happy and healthy holiday season! 

Susan