AI relies heavily on data, yet much of the health data remains underutilized due to concerns over patient privacy, regulations, and intellectual property protection.
“This is the core challenge” in developing AI solutions for life sciences and related sectors like pharmaceuticals, explained German entrepreneur Robin Röhm. Additionally, collaboration on sensitive data often presents difficulties. His startup, Apheris, seeks to address these issues through federated computing. This method enables secure access to data for AI model training without transferring it, taking a decentralized approach.
APHERIS AI:
Apheris‘ clientele includes Roche and several hospitals, Röhm added.
Federated computing works by executing computations locally where the data resides, with only the results (such as model parameters) being aggregated centrally, said Marcin Hejka, co-founder and managing partner at OTB Ventures. Hejka co-led an $8.25 million Series A round for Apheris with deep tech investor eCAPITAL.
He believes Apheris could play a crucial role in the emerging federated data networks. “We see a maturing ecosystem of third-party software tools (open-source federation engines, data quality tools, and security products),” Hejka told TechCrunch. “Apheris also facilitates seamless integration with complementary privacy-enhancing technologies (like homomorphic encryption, differential privacy, and synthetic data).”
The company’s latest funding follows a pivot. Röhm and co-founder Michael Höh initially launched the company in 2019 to create a federated learning framework to rival open-source alternatives. After raising a significant seed round in 2022, they shifted focus in 2023 to address data owners’ needs, particularly in the pharma and life sciences sectors.
This pivot paid off, with the company finding product-market fit and increasing its revenue fourfold since launching its new product in late 2023. Backed by existing investors, including Octopus Ventures and Heal Capital, the new funding round brings Apheris’ total funding to $20.8 million. The money will be used to hire senior talent with expertise in life sciences and business development.
AISB:
Apheris’ software agent, the Apheris Compute Gateway, acts as a bridge between local data and AI models. It’s already being used by the AI Structural Biology (AISB) Consortium, a collaboration between companies like AbbVie, Boehringer Ingelheim, Johnson & Johnson, and Sanofi to advance AI-driven drug discovery.
The company will focus further on protein complex prediction with this new funding. While its solutions are agnostic to specific use cases, Apheris recognizes its potential to unlock valuable and diverse data in life sciences, especially when limited public data is available, and companies are reluctant to share their data without proper assurances.
“Without addressing the concerns of data owners regarding AI data sharing, we believe the full potential of AI cannot be realized. That’s the core mission behind what we’re building,” Röhm said.
