AI ventures: Creating an AI instruction platform for health care

A startup active in the healthtech field approached us to design a software platform to revolutionize how healthcare instructions are being given to patients. The founding team had spotted a large problem that needed to be solved. Many patients get instructions from their health care practioners they have to execute at home. These instructions range from the intake of medicines to more complex instructions such as creating a stoma. Currently patients first practise these instructions together with a health care practioner for the most complex instructions but many times patients are left with instructions folders full of text missing clear visual explanations. On top of this many instructions are not accesible to people with disabilities. When you imagine that in The Netherlands approximately 2,5 million people suffer from low literacy problems, it becomes clear that folders with a lot a text and unclear instructions lacking visual content are not the best way to educate patients about the home instructions they have to execute themselves. 

The founding team of this startup wanted to create a digital instructions platform which would create on one hand a very straighforward digital instructions platform making heavily use of visual content, and on the other hand wanted to use artificial intelligence algorithms to make the instructions accesible to disabled people.

After having discussed the requirements for the platform we became very excited about the vision and goal of the founding team. We decided to participate in the platform by exposing our technical knowledge to them and creating the artificial intelligence engine of the platform. This artificial intelligence engine would convert instructions automatically to speech for disabled people with low literacy or reading problems. On top of that the artificial intelligence engine would automatically classify patients struggling with the execution of their instructions. By doing this health care practioners could be proactively warned when patients are struggling with certain instructions.

At this moment the artificial intelligence engine we developed is integrated into the main software architecture of the platform. When this engineering step is finished the platform will go into beta mode, to be tested by a group of health care practioners.

This is a very exciting project and we are happy to contribute to the great vision of the founding team. We really hope they will succeed in their mission by providing patients with much better healthcare instructions than the way it is currently done. In future blog posts we will share updates about the development of the platform and exciting milestones reached by this startup.