How to get started using AI to solve sick care problems

Dear Dr Meyers,

"It seems all everyone is talking about these days is artificial intelligence in medicine. I'm a grunt in the trenches working for the man and I'm doing all I can do to keep my head above water let alone integrate AI into my practice. Please help me."

You are not alone. Everyone else is in the same boat, so take it a step at a time.

Here are some ways to start becoming AI competent:

1. Acquire education, and access to resources, networks, experience, mentors, peer to peer support and career guidance.
2. Start by starting your life-long AI education, training, and develop by taking inexpensive or free basic courses e.g, on LinkedIn Learning, Coursera or The American Board of Artificial Intelligence in Medicine
3. Don't let school or getting more degrees get in the way of your education. Most of your learning will come from project-based learning working with other collaborators and colleagues. Volunteer to work on an AI project team where you work or with an AI startup , accelerator, or venture firm.
4. Attend AI conferences, webinars, and workshops including those that have nothing to do with using AI in medicine. Sick care cannot be fixed from inside so learn how other industries are solving problems that plague sick care, like search, simulations, rapid prototyping, workflow and process mapping, etc. Here is what you just missed .
5. Talk to first year medical students and take advantage of reverse mentoring
6. Start AI rounds with data scientists, similar to what bioengineering students do during their first hear of undergraduate bioengineering programs.
7. Create knowledge transfer programs.
8. Volunteer to teach AI. The best way to learn something is to teach it.
9. Learn to speak "data" without an accent and pass it forward by teaching your data colleagues how to speak "medicine"
10. Start playing with free, readily available large language models to do things, like writing the draft of a LinkedIn article like this one.
11. Admit that you don't know what you don't know
12. Pass it forward to other members of your community of practice who are on social media and Slack channels, Substack, LInkedIn groups and local meetups
13. Network, network, network
14. Work on your personal and professional development and career plan
15. Join an angel network to get the investor perspective. Many don't make you pay to play.

Try these three applications of the 70-20-10 rule.

Application 1: How people learn

The 70-20-10 Model for Learning and Development is a commonly used formula within the training profession to describe the optimal sources of learning by successful managers. It holds that individuals obtain 70% of their knowledge from job-related experiences, 20% from interactions with others, and 10% from formal educational events. Are you sure you want to get an MBA?

Application 2: The amount of time you should spend on getting stuff done, strategic thinking and strategic planning i.e. making your present business model obsolete before someone else does

The rule states that you should spend 70% on the now, 20% on the next and 10% on the new.

Google says it follows the rule. But, it seems, the reality might be different.

"We encourage our employees, in addition to their regular projects, to spend 20% of their time working on what they think will most benefit Google," they wrote .?"This empowers them to be more creative and innovative. Many?of our significant advances have happened in this manner."

Huge 20% products include the development Google News, Gmail, and even AdSense.

However, whether or not 20% time actually exists anymore has been a matter of debate. In 2013, Chris Mims wrote for Quartz that 20% time was "as good as dead" because it became too difficult for employees to take time off from their normal jobs.

Quiet quitters, by definition, have killed any discretionary effort beyond doing what they get paid to do.

The 70-20-10 ratio was always an average though. Some companies, like technology companies which need to produce new offerings more quickly, might have a ratio that is more like 45-40-15. At the other end of the spectrum might be large consumer goods companies with established core offerings, where the ratio might be 80-18-2.

Application 3: Who does most of the work in teams

This a take on the Pereto Principle. The Pareto Principle states that around 80% of outputs from a system are caused by about 20% of inputs. It can also mean that around 80% of effects are brought about by 20% of causes, or that 20% of effort creates about 80% of the results — the Pareto Principle just describes any system where there’s an 80/20 split in productivity.? In other words, if you have 10 people on a team, 2 will do 80% of the work that the other 8 should be sharing. There is a lesson here for teachers trying to get all members of student project teams engaged.

Application 4: Impact of your innovation portfolio allocation

Companies that allocated about 70% of their innovation activity to core initiatives, 20% to adjacent ones, and 10% to transformational ones outperformed their peers, typically realizing a P/E premium of 10% to 20% (see the exhibit “Is There a Golden Ratio?”). Google knows this well: Cofounder Larry Page told Fortune magazine that the company strives for a 70-20-10 balance, and he credited the 10% of resources that are dedicated to transformational efforts with all the company’s truly new offerings.

Pick a problem to solve and go get Sh#T done.

You passed organic chemistry for God's sake. You can do this.

Arlen Meyers, MD, MBA is the President and CEO of the Society of Physician Entrepreneurs on Substack