It is one thing to develop a medical algorithm, but another to prove that it works. It's hard to find medical data to do that. One startup is ready to provide that in spades, along with the tools to make validation studies easier.
500 Global led a $2 million seed round for Gesund. The company has viable platforms, 30 clients in their sales pipeline and revenue expected this quarter, according to CEO and founder Enes Hosgor.
Gesund is a Contract Research Organization that is used for the development of medical algorithms and models. The same way a CRO might design a clinical trial for a drug or medical device company, Gesund's platform creates 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299 888-353-1299
I like to think of us as a machine learning ops company.
There is evidence that getting diverse and usable data sets can be a challenge, and that a medical algorithm is only as good as the data it is trained on. A study that was published in the Journal of the American Medical Association in 2020 analyzed 74 scientific papers describing deep learning and found that 70% of the data was from New York, California and Massachusetts.
34 U.S. states did not contribute any data to the training of these algorithms, which raises questions about how generalizable they might be to a wider population.
There is an issue across different types of healthcare providers. You could use the data collected at the hospital to train an algorithm. If you want to deploy that in a small community hospital, it's not certain it will work.
According to a meta-review of 152 studies published in the BMJ, the data sets used to train algorithms are smaller than they should be. There are some success stories, but this is an industry-wide problem.
You can provide data that isn't there in the first place if you use technology alone. Genetic studies for people of non-European ancestry are sorely lacking. Making existing data easier to access and creating partnerships that open up new avenues for data sharing are the issues that Gesund is focused on.
Gesund's validation platform is shown in the picture.
Existing data sharing agreements in place with clinical sites are what gives Gesund its data pipeline. The University of Chicago Medical Center, Massachusetts General Hospital, and Berlin are the places where Gesund is focused right now. The company plans to expand in the future.
Aggregating and delivering data for use in machine learning applications is also being done by others, like the Nightingale Open Science Project, which will freely provide clinical data sets to researchers. The technology stack that Hosgor sees as the company's secret weapon is the data itself.
The average healthcare provider doesn't have a cloud, so all that goes out the window. The bread and butter of machine learning is not reliant on any third-party managed services.
There is a low code interface on the platform. Doctors and providers can drag and drop the data they need and test their own data against it.
We built the first product that allows model owners to run their algorithms against data in high compliance environments. He explained that it was their secret sauce.
Nightingale is providing some of its services for free at the moment. The Community Edition of the company allows academics to test their existing algorithms for free, but they will have to submit their own data sets.
The company's premium version will be paid for by the artificial intelligence companies. The paying customers will have access to proprietary data sets. They will pay for the data they need. Gesund claims to have 30 potential clients, and expects to make money this quarter.
We were at theRSNA in Chicago in November and every single company we talked to said I needed evidence yesterday.
The $2 million pre-seed round is all of the funding for the company. The company will focus on R&D in the near future.
All Rights Reserved © 2024
