Mantis Biotech Creates Digital Twins to Solve Data Gaps in Medicine

Mantis Biotech, based in New York, is developing a solution to the data availability problem in medicine by creating so-called "digital twins" of the human body. These physics-based models could significantly accelerate genomics research, improve diagnostics, and support clinical decision-making. By leveraging vast datasets, Mantis Biotech aims to overcome the limitations faced by traditional models, especially in edge cases like rare diseases where reliable data is often scarce.

The company's platform integrates diverse data sources, including textbooks, motion capture cameras, and medical imaging, to create synthetic datasets. These datasets can be used to build predictive models that assist in studying new medical procedures and training surgical robots. For instance, a sports team could predict the likelihood of an NFL player sustaining an Achilles injury based on their recent performance and health status.

Mantis employs an LLM-based system to route, validate, and synthesize data, which is then processed through a physics engine to create high-fidelity models. This enables the company to transform disparate data sources into predictive models useful across various medical fields.

Founder Georgia Witchel emphasized the potential of their technology in the biomedical industry, where access to information about procedures and patients is often limited. She highlighted the importance of generating synthetic data to address ethical and regulatory challenges associated with using real patient data.

So far, Mantis has achieved success in professional sports by creating digital representations of athletes and analyzing their performance metrics. Recently, the startup raised $7.4 million in seed funding, which will be used to enhance their technology and market presence. The next step for Mantis is to continue developing the platform and eventually launch it for the general public, focusing on preventive healthcare.