GE-backed deep learning startup Arterys bumps up Series A to $12M+ as it nears cardio launch, goes into oncology

Tools
Image created by Arterys' system--Courtesy of Arterys

Medical imaging deep learning startup Arterys has a major partnership with GE ($GE) and a cardio product launch teed up as part of that for this fall. Now, it's lined up a more than $12 million Series A financing to help it transition into its next field: oncology. The San Francisco, CA-based startup had previously disclosed a smaller sum for that Series A of $7 million late last year.

GE was able to integrate Arterys' work into its products quickly due to an existing relationship between Stanford University and GE. They're also both investors through GE Ventures and Stanford-StartX Fund, which was co-founded by the StartX accelerator, Stanford University and Stanford Health Care. Emergent Medical Partners led the round, with participation from Norwich Ventures as well as existing investors Asset Management Ventures, AME Cloud Ventures and Morado Ventures.

The GE deal is a co-marketing agreement; the Arterys System will be available via the ViosWorks application for GE MRI scanners this fall. It will be downloadable software that's compatible with the thousands of new and installed GE Healthcare MRI scanners. It's already in use at research centers globally.

The diagnostic software connects to a standard MRI machine to enable noninvasive quantification of cardiac blood flow. The deep learning analysis is based on a simple, 10-minute MRI scan. After a few minutes of analysis with the Arterys system, it can assess patients with all sorts of cardiovascular disorders including structural heart disease, congenital heart disease, carotid and neurovascular as well as renal vascular disease. The idea is to help doctors more easily and usefully process massive amounts of data.

"It's not possible for physicians to process all this data themselves," co-founder and CEO of Arterys Fabien Beckers told FierceMedicalDevices in an interview. "We take advantage of neural networks to process all this data. We have reached the limit on how much a single computer and a single doctor can do with data."

Remarkably for such an early startup, Arterys is bringing in revenues from the GE deal and will continue to do so. Beckers said that the partners have learned a lot about each of their relative advantages.

"The important piece is about how important it is to work with the vendor from the startup perspective," he said. "We are benefiting from each other's strength. They can make it deployable very quickly and we can be very focused and tailored on what we do best."

Up next, Arterys is tackling the daunting world of oncology. It aims to apply deep learning--in which the system is constantly adapting based upon new inputs--to all sorts of oncology data on a single patient including imaging scans such as CTs or MRIs, electronic health records, genomic data, blood tests and even demographic information to help physicians to get to better patient treatment faster.

"There's a lot of data tailored for clinical purposes. They system can learn dynamically all the time. It's a whole different way of thinking about it--no one else has that. It's a two-way street between doctor and software. It's very important to have and really leverage deep learning as a tool, as a means to an end," summed up Beckers.

- here is the announcement

Special Report: Big data dominates as deep learning, AI seal the deal

Related Articles:
Dell, Zebra Medical offer machine learning algorithms for medical image analysis
GE, Arterys ready launch for deep learning diagnostic system for cardiac MRIs
Philips, Hitachi partner to develop universal, next-gen medical image management
Former Philips Imaging CEO heading machine learning startup to improve stroke diagnosis
Artificial intelligence startup nabs $12M+ to monitor medication adherence via smartphone
Stealth machine learning startup raises $10M to develop a heart disease detection algorithm
Startup gets $7M Series A, GE partnership to apply machine learning to medical imaging
Deep learning radiology startup Enlitic wins Aussie partner, $10M Series B