There are 5 important steps when getting started with AI in the medical field. Here is number 1: Defining the correct problem statement!
We know we aren’t the first company to shout out loud that AI brings a lot of opportunities in the medical field. If everybody is dreaming about endless possibilities, what are the barriers to start make them happen. Here is a short list of some of the most recurring questions and how to tackle them. Enjoy the first entry of the medical AI series on defining the correct problem statement.
Let’s explore AI… But, what does it offer?
Like any other product or service, AI is a tool that can be used to solve problems. Not any problem, but problems with a sort of rhythm, a routine. AI will create new insights from data and can be used to double check a diagnosis or find new insights that are difficult to interpret or see with a human eye. Moreover, AI can be used to find signatures and anomalies in massive amounts of structured and unstructured datasets and it will make medical workflow more efficient.
Every time people ask if we can explore the possibilities of AI, my first questions are: “What data is available or can be made available? Are there repetitive tasks in your work flow that you have to do over and over again to gain crucial information? Do you have problems that in your opinion can be solved by exploring data, but that is manually impossible?”
Defining the correct problem statement
So if you consider exploring the possibilities of AI. Let’s start with searching and defining the correct problem statement. This challenge can’t be solved by data engineers only, other disciplines are required too. The availability of human factor engineering, design thinking and service or business engineers is crucial.
The first discipline will help analyze the workflow and its users, this is required to build an in-depth understanding of the needs and pains of your users. By interviewing and observing the different stakeholders in a well-structured methodological way, you can capture hidden unmet needs or assess current dissatisfaction and inefficiencies. Design thinking allows to visualize problems and new user interactions, as well as rethink the workflow that of course will need to be enabled by the AI tool. Design thinking even allows to create a new ‘what if’ future and solutions to problems we will only encounter in the future. In addition, business and service designers will help to define and size opportunities, and detect the high willingness to pay for the new values offered with your AI solution.
Any questions on this post? Submit them below and we’ll get back to you soon!