According to a story from Biospace, in Palo Alto, AI company Emedgene plans to dive into helping the undiagnosed with their new rare disease initiative. The company will be paired with the world’s top research institutions to conduct analysis using algorithms that try to solve unanswered patient cases using new information and research as it get released. The research institute will be conducting a genomics analysis using ‘Pathorolo’, the first machine-learning algorithm that assesses how likely it is that a disease case can be solved and continuously reruns the cases as new evidence becomes available.

Emedgene’s Method

Emedgene is a genomic interpretation platform that uses AI to conduct their genomic analysis in order to speed up the process and make more discoveries. The platform is utilized by many of the leading genetic labs such as Baylor Genetics, TGEN, Medical College of Wisconsin, and many others. The platform runs and reruns their algorithms, searching through databases and literature until it finds information to confirm that a case could be solved. At that point, it alerts the patient who is awaiting diagnosis. While half of the cases are solved on their first analysis, their process vitally continues the search. Emedgene’s CEO Einat Metzer explains:

“Genomics knowledge is growing at a fast pace, and at re-run, an estimated 10% of cases can be solved with new information. In this era of extremely rapid discovery, ironically, computational AI approaches can demonstrate human compassion.”

Their algorithm holds up: 93% of the 553 patients cases they deemed solvable were eventually solved. Once the case is deemed solvable, they recommend patients see a geneticist, armed with the algorithms findings to seek out answers.