Sophia: Data driven medicine for genome diseases

Sophia is a new data driven medicine based on collective artificial intelligence. Sophia can reads and aggregate the genetic code of DNA to predict and diagnose genetic diseases such as cancer. Sophia relies on DNA data sent to the company by 170 hospitals in Sophia DDM, a clinical genomics platform for molecular diagnostics, to improve the ability to detect and diagnose genomic conditions.

The technology will soon be available to hospitals, clinicians and members of the Sophia DDM community. The secure and private pooling of more patients’ genomic profiles on Sophia DDM will allow for similar advances from Sophia for 40 other genome diseases; oncology, hereditary cancers, cardiology, metabolic disorders and paediatrics, and the advent of a true Data-Driven Medicine for patients.

Sophia passes the test
The results revealed from Sophia Genetics’ study on Sophia were obtained from 30’000 patients’ genomic tests building on the information pooled by hospitals in Sophia DDM. An initial match of 85% from 10.000 analysed patients was obtained which improved to a 96% match with 20.000 tests, and a 98% with expert clinicians’ variant pathogenicity predictions for BRCA genes mutations, which bear a potential risk of susceptibility to breast cancer. Sophia continuously analyses the collected pool of DNA data sent to the company by 170 hospitals globally and in doing so, improves its own ability to detect and diagnose genomic conditions.

Jurgi Camblong, CEO and co-founder of Sophia Genetics says: “Sophia will constantly examine the genomic data of thousands of patients in our network and gradually teach itself to not only identify genomic diseases but to become more accurate in its diagnosis. This is the first time we have introduced machine learning into our operation. As it constantly learns, it will vastly improve the speed and precision with which we can identify potentially life-threatening diseases, and allow us to diagnose many times more patients than we were able to previously.”