Karavela is on a mission to collect one of the largest datasets of brain scans in the world. We believe that this data can help us better understand the human brain and develop new treatments for neurological and psychiatric disorders. Because this data is both very valuable and sensitive, we are committed to sharing it responsibly all the while protecting the privacy and rights of individuals who participate in our research.

• For science - We believe science is safer when shared, double-checked and openly discussed, which is why we will share part of our data, models and results with the scientific community.
• For patients - We are committed to put our data and models to good use, which is why we prioritise utility in clinical research centered on psychiatry and neurology related disorders.
• For global impact - All brains are different, and we intend to collect data that can benefit the largest number of people across the globe, regardless of age, gender, or background.

Open science is faster and safer

Research is most valuable when it is shared and discussed: we pledge to openly share our model training methods and results for the community to inspect our work and to foster reproducibility and allow this field to move at the fastest pace.

Moreover, data acquisition and model training are very expensive and time-consuming, which is why we will allow other teams to use some of our hardly-obtained ressources to advance our common understanding of the brain.

Clinical research is at the center of our mission

We are a passionate team of researchers, engineers and clinicians who are fascinated by the human brain and its mysteries. Nevertheless, we want to make sure that our work converts into activable scientific insights that can bring tangible benefits to patients and the broader community.

Our data collection and modelling endeavours are not the end game, but necessary steps to building tools that are useful for patients and medical professionals.

Every brain is different, and our data reflects that

Brains are widely different from one person to another. Building models of the brain which are useful to the largest proportion of the population requires a diverse dataset that captures this variability. We pledge to acquire data from participants with very diverse backgrounds, including age, gender, ethnicity, and neurological conditions, to ensure our models are representative and effective for everyone.

Moreover, our data acquisition programs are designed to include many international partners, ensuring a wide range of perspectives and expertise in our research efforts all the while making sure that the protocols we develop can be applied globally.