CHICAGO — Tempus AI, Inc. has announced the launch of a new initiative, dubbed Fuses, aimed at revolutionizing diagnostic and therapeutic research through the application of artificial intelligence and one of the most expansive healthcare datasets in the world.
Fuses will leverage Tempus’ proprietary multimodal dataset, which includes over 40 million research records. This dataset features more than 1.5 million records linking clinical and genomic data, 2 million imaging records, and roughly 300,000 records combining genomic and transcriptomic information. By applying its advanced machine learning capabilities to this massive trove of data, Tempus aims to create powerful diagnostic and predictive tools that could transform personalized medicine.
The program is designed to accelerate the development of Tempus’ AI-driven diagnostic portfolio, providing physicians with an expanded suite of algorithmic tests. These diagnostics aim to inform treatment decisions more accurately by identifying patients unlikely to benefit from specific therapies or those at elevated risk for severe side effects. One such innovation already underway is the company’s Immune Profile Score (IPS), a multimodal biomarker used to assess prognosis in patients with metastatic pan-solid tumors eligible for immune checkpoint inhibitor therapies.
Tempus CEO and founder Eric Lefkofsky said the launch of Fuses represents the culmination of a decade of work collecting and curating an industry-leading data library. He emphasized that the new program positions the company to develop what could become the world’s largest diagnostic AI model, capable of answering previously unsolvable questions in precision medicine.
“Our goal is to make diagnostics smarter and more personalized, and we are on the verge of transforming how therapies are matched to patients,” Lefkofsky said. “We are now equipped to deliver tangible, transformative change in clinical practice.”
At the heart of Fuses is a foundation model designed to learn generalizable principles related to prognosis and therapeutic efficacy in real-world settings. This model will support drug development by offering new insights into why certain clinical trials fail, proposing new uses for investigational drugs, and guiding more effective trial designs. Additionally, it may reveal mechanisms behind drug response and resistance, fueling the development of next-generation companion diagnostics and targeted treatments.
Through Fuses, Tempus seeks not only to refine its clinical offerings but also to support the broader medical community in advancing therapeutic innovation and improving patient outcomes at scale.
