At yesterday’s WIRED Health (9 March), founder of Sophia Genetics, Jurgi Camblong, explained how his company applies machine learning to genomic knowledge shared across 45 countries to improve treatment at a constant rate. MedTech Engine sits down with the Swiss biologist to dig deeper into data-driven medicine.
Jurgi Camblong: To us, AI is “re-humanising medicine” in the sense that it is releasing clinicians from some time consuming processes of the clinical workflow, to re-focus on the relationship and the direct contact with the patient, which are so important in diagnostic. Regarding your second point, as such, the main characteristic of an AI is its faculty to learn automatically and continuously as it is used. Think of self-driving cars for instance. They are not programmed to know how to drive any road on the planet, but instead, they are developed in a way that any new road taken is an opportunity to learn how to deal with it, and then share this knowledge with the community of users. The same applies to SOPHiA: the more data SOPHiA crunches, the more it learns, as information is continuously shared on the platform to help clinicians advance their own knowledge and take better decisions. As to how this knowledge should be shared, I can only refer you to former US President Barack Obama 2016 statement for the launch of the Precision Medicine Initiative in the US, which outlines the necessary steps to democratize data-driven medicine:
As I am sure you’ve understood, those steps perfectly summarise the path actually followed by Sophia Genetics form start, and how its community share and scale knowledge across the globe already today. With more than 240 healthcare centres in 45 countries united around SOPHiA, our artificial intelligence, we’ve created the world’s largest clinical genomics community. The collective knowledge of this community allows SOPHiA to continuously refine diagnostic and treatment recommendations as more cases are analysed. Knowledge sharing is for us the cornerstone of the democratisation of data-driven medicine.
In some regards, part of our democratisation effort is to debunk the silos that might exist, and advocate for a “platform thinking”. Our track record to date, and especially the number of strategic partnerships we have concluded with the main players form our field – Devyser, Horizon Discovery, Qiagen, ArcherDx, and DNA sequencing giant Illumina to name a few – show that the value proposition of Sophia Genetics have been clearly articulated and that our stakeholders have understood the value of joining our clinical genomics community. Everybody has a role to play in data-driven medicine and we are proud to have federated a community that is together advancing genomic knowledge to the ultimate benefit of patients.
To clarify one thing: Sophia Genetics is first and foremost active in the diagnostics field, and not a purely R&D effort. Our technology is already used by clinicians from 250 institutions in 45 countries to diagnose patients and recommend tailored care. To date, 100.000 patients have benefited from our technology. That being said, as far as Brexit is concerned, this should not prevent us from continuing our R&D efforts in Switzerland, to ensure we continue to develop the most advanced tools in the field.
We believe blockchain technology shows an important potential in different fields. As far as Sophia Genetics and data privacy are concerned, our approach guarantees the privacy of patients’ data by stripping the package entered into the platform of any personal information such as the name of the patient. With the sensitive data available only to clinicians, we ensure the highest level of data privacy. As for the infrastructure, Sophia Genetics works with a network of private and secure servers, based at the country level to comply with the most stringent local regulations on this type of data. Finally, we continuously work to advance our tools in this area and we’ve recently published in Genome Research about a new technology called SECRAM that the company has developed to protect such kind of infrastructure:
We’re in the midst of a revolution, driven by the democratisation of data-driven medicine. Moving forward, we should enter a new era of “real time epidemiology” where access to a wide range of clinical data should allow us to go one step further in personalised care.
You're the expert! Write for The Engine or share your articles, papers and researchAdd your content
Add your content
Sign up for Ignition, our regular, ideas-packed newsletter