Freenome has announced an expanded collaboration with NVIDIA to accelerate its artificial intelligence (AI) and deep learning initiatives aimed at advancing personalized multi-cancer detection through its blood-based screening tests, leveraging NVIDIA’s accelerated computing technologies to enhance analytical speed and accuracy in identifying cancer-specific signals in blood at the earliest stages of disease. The partnership focuses on scaling training of Freenome’s proprietary cell-free DNA (cfDNA) fragment-level deep learning (FLDL) models, addressing large dataset challenges with high-performance computing, and developing an open-source cfDNA methylation foundational model to benefit broader genomic research. The companies will work on optimizing the FLDL model to handle millions of cfDNA fragments and billions of base pairs per sample, overcoming training and inference bottlenecks as commercial testing scales, and employ the NVIDIA BioNeMo framework to build a foundation model contextualizing methylation patterns for multiple applications. “NVIDIA’s expertise in accelerated computing hardware and software has helped Freenome solve data loading and other training bottlenecks we experienced with our FLDL model,” said C. Jimmy Lin, M.D., Ph.D., MHS, chief scientific officer at Freenome, noting that NVIDIA’s frameworks like BioNeMo and Parabricks could transform deep learning R&D, commercial testing processes, and foundational model development for the genomics community.
Also Read: Chai Discovery Secures $130M Series B to Accelerate AI-Driven Molecular Discovery
In addition to model advancements, the collaboration aims to empower healthcare organizations (HCOs) by exploring access to de-identified clinical and molecular real-world data (RWD), enabling accelerated AI research for future diagnostic and prognostic applications. Freenome’s multiomics platform, which profiles DNA methylation, RNA, protein, and other analytes to generate billions of data points per individual, relies on AI and machine learning to detect subtle cancer signals that traditional methods may miss, with its first test, SimpleScreen™ CRC for colorectal cancer, currently under FDA review with approval expected in the second half of 2026, and additional cancer detection tests planned for launch that year. This expanded initiative underscores the strategic integration of AI, deep learning, and high-performance computing to improve early cancer detection and broaden the impact of Freenome’s screening technologies.