Week 22, How Can AI Help?

Continuing on previous week’s topic on artificial intelligence, this week was mainly filled up by discussions on what AI can, and is allowed, to do for us in the future.

Sensotrend is one of two SME’s in the WARIFA project. The EU project has 12 partners in total, all approaching AI and healthcare from different angles.

The name WARIFA is an acronym of Watching the Risk Factors. The aim of the project is to research how artificial intelligence can be used to produce personalized recommendations based on rich sets of data from citizens and from their environment.

In the application phase the project got full points in EU’s evaluation. It’s the first time for all participants to be involved in a project that has received such a great evaluation.

The project focuses on cancer, cardiovascular diseases, diabetes and chronic respiratory diseases. We bring to the project our expertise in getting data from all medical devices used in treatment of diabetes, as well as our know-how in integrating data from wellness trackers.

Another specific interest for Sensotrend is to learn whether AI could help merge perhaps conflicting advice from different branches of healthcare into a justified personal recommendation for a patient. Let’s say a patient has both diabetes and a high risk for melanoma, and perhaps some depression too. For treatment of diabetes, some exercise would be beneficial. For depression, this could be combined with the advice to go out and enjoy the sun and nature. However, for melanoma risk it would be good to avoid excessive sunlight. How should these different recommendations be merged? Should the risk factors be weighed against each other? Should we involve the patient and their personal goals and values? Could the algorithm come up with a recommendation to go out and do some exercise, but with enough protective clothing?

All of this is very much an oversimplification, of course. It’s just an example illustrating what kind of questions the project is considering.

This week was a project meeting week. We should have been discussing with each other face to face in Porto, but the meeting was changed to online only mode at the last minute. Meeting only online certainly affects the team dynamics in this kind of projects. Through Teams, we mostly discuss things on the agenda. If we would have been face to face in Porto for three days, there would have been more time for more informal discussions and for free flow of ideas, over breakfasts and dinners. Hopefully the next meeting can be organized in a physical location.

I should note that academic projects operate on a completely different frequency than startups. One and a half years have passed of the four-year project, and in my view very little tangible results have been produced. Surely we have created documents and published some research papers, which are of course meaningful. But there’s no app yet, and the infrastructure for sharing data is only now being set up. To a startup mainly involved in software development, this feels slow.

However, we’re also not in a big hurry to jump on the AI train. It will take some time for us to learn where AI can best help healthcare, and where it will be allowed to help.

This post is part of a series looking at what we at Sensotrend do besides our everyday product development activities. See the full list of articles.