What is Panops.ai and how did the project start?
Panops.ai aims at identifying patients and segment kidney stones from the CT images, and predict the best treatment type based on the stone care characteristics. Urologists use an existing framework to perform treatments and make diagnosis. However, we noticed that there are quite some outliers in this process.
The project started at ACE AI Lab. This programme goes to big corporations and figures out a project that they could do with AI, but don't have the capacity or other means to achieve. ACE AI Lab gives this challenge to a student team. We collaborate with OLVG. The urology unit leader there, Ernst van Haarst, started researching into how to treat kidney stones better. And that's where we came into the picture.
So how does Panops.ai help medics in their work?
Well, it makes all the measurements that they do, but automatically. It makes the measurements faster and, more importantly, more precise. And since we can have a kidney stone segmented in a CT scan, we can reconstruct this in 3D and put a heat map on it, so doctors can actually see the stone and the actual density map of it. And they can also virtually cut it in half, so they can see inside.
Additionally, we can calculate the Skewness and the kurtosis of the distribution, which has a lot of predictive power, since it indicates whether there are big differences between the densities inside or something else.
What part does the ACE Incubation Program play in this project?
Their help is very valuable. They host a weekly sessions for us to learn how to perfect our skills regarding either sales, how to do your business model, how to talk to customers, how to find them, how to deal with the legal part, intellectual property, and so on. We also got really valuable contacts from them. So there's a lot of topics that they introduce us to, and we can ask questions and interact with the lecturers. We already incorporated quite a lot of information from the programme.
What’s in store for Panops.ai in the near future?
We are mainly focused on the on the technical side. We are developing our tool to actually predict outcomes. And if we have that, then we want to extend our partnership to other hospitals, and do a study together to actually evaluate our prototype and see how it performs in real life. In the end, with this gained knowledge, we want to actually commercialize it and bring it to markets.
In the end, Panops.ai has the goal to empower medical professionals to make better and more accurate, faster diagnosis. So we don't want to restrict ourselves to only deal with the kidney stones or even kidneys. We see this project as a great starting point for us to develop as a team and building meaningful partnerships with hospitals and doctors.