November 01, 2019
Most of what we do has no proven downstream outcome effects. IMHO the main reasons the radiology AI companies have not met investor expectations are:— Curt Langlotz (@curtlanglotz) November 2, 2019
1. high quality clinical training data is a lot harder to come by than radiology AI startups anticipated. https://t.co/FM6wyackDm
The points made by Dr. Curt Langlotz are incredibly important, and are matters that we, as a biomedical startup had to consider when starting out.
Biomedical data is indeed very hard to come by, and is often provided by only one hospital causing issues with generalization. That is why we have partnerships with hospitals all around the world (and constantly adding new ones!) which are using devices that we supply to supply us our own data set. We also perform constant quality checks to ensure the data coming in is useable and practical.
While many companies do aim to solve familiar problems, we are going for issues that no one else is attempting and are fairly difficult. Beyond that, our solution isn't simply a 'nice to have', it is a revolution in the diagnostic world.
As for workflow, our team includes world class radiologists that help us understand the clinical environment, and address any issues before they come up. As well, our technology is device agnostic, wrapping around whatever ultrasound tool is currently in use, making it more powerful than before. With those things combined, integration and adoption should present no issue.
If you have any questions about us please reach out to one of our team members!