Step 01
Define the types
Define MS types using MRI and blood tests to better reflect underlying abnormalities.
MS-PINPOINT · MS classification with AI
Defining the types of MS that MRI and blood tests can see.
We are developing an AI tool that detects different types of multiple sclerosis (MS) using magnetic resonance imaging (MRI) and blood tests.
Chapter 01
What we are building
Our goal is to help doctors recommend the most appropriate treatment with the fewest side effects for each patient.
To do that, the tool has to learn from more than one clinic. MS-PINPOINT reaches back ten to twenty years into the archives of twelve hospitals, so the data behind it reflects the diversity of everyone living with MS — including patients who are under-represented in clinical trials.
Chapter 02
The aim
Our aim is to identify patterns in MRI and blood tests that characterize specific types of MS. This information will help us predict how symptoms may change, when MS may worsen, and which drugs are most likely to help.
Chapter 03
The methods
Three pieces of work, each one a prerequisite for the next.
Step 02
Measure what treatment does
Build AI tools that measure and predict treatment effectiveness at an individual level.
Step 03
Learn without moving data
Develop AI tools that can learn from each other without sharing patients’ data across centers, ensuring personalized predictions from diverse backgrounds.
Chapter 04
What changes for patients
We will measure how the new MS types complement the existing ones, calculate potential cost savings, and improve healthcare by recommending appropriate treatments earlier and delaying disability.
Chapter 05
Patient information and opt-out
Data for research is anonymised inside the hospital, and every person living with MS at a participating hospital can ask to be left out of the project.