Mantis Biotech creates digital twins of people to address the shortage of medical data

What Mantis Does

Mantis Biotech's approach is built around the idea of gathering different types of information about a person in a single digital environment: anatomy, physiology, and behavior. From these fragments, the company forms synthetic datasets, which can then be used to create digital twins of the human body. Such a twin is not equivalent to a living patient and is not an exact copy in the literal sense.

Rather, it's about a computational model that helps describe how the organism is structured and how it potentially responds to different influences. The goal is to transform poorly compatible medical data into a more cohesive foundation for research. In medicine, needed information is often stored in different databases, collected under different protocols, and doesn't transfer well from one research context to another.

If Mantis can standardize this layer and make it suitable for modeling, drug developers will gain a new working tool even before expensive clinical testing stages.

Why Digital Twins Are Needed

For pharma, a data deficit is not an abstract problem but a direct speed limitation. Researchers may have a strong hypothesis about a disease mechanism or how a molecule works, but run into a shortage of comparable data arrays. Digital twins in this logic are needed not for a flashy AI showcase, but to quickly test scenarios, compare models of how the organism reacts, and find weak points in existing observation sets.

• combine anatomical, physiological, and behavioral data in a single model
• supplement real medical samples with synthetic data
• test hypotheses before more expensive development stages
• faster identify data gaps for specific diseases
• reduce dependence on rare or slowly replenished datasets

If such an approach works with sufficient accuracy, companies will be able to use synthetic datasets as an intermediate layer between raw observations and applied conclusions. This is especially important where real data is difficult to collect due to cost, privacy, or a limited number of suitable patients. In this scenario, a digital twin becomes not a replacement for clinical reality, but a way to more efficiently extract signal from it before the next round of expensive research.

The Main Question

Such systems have an obvious limitation: the quality of the result always depends on the quality of the source material. A synthetic dataset is useful only insofar as it accurately reflects real biological processes. If the original sources contain biases, gaps, or poor representativeness, the model can reproduce the same errors, only in a more convincing and technologically sophisticated wrapper.

This is why the conversation about synthetic data in medicine quickly comes down to validation, quality control, and confidence in conclusions. This is why for Mantis the main test will not be data generation itself, but trust in it. Pharmaceutical companies and research teams will look at reproducibility, transparency, and practical applicability of such models.

The market will ultimately assess not the loudness of the digital twin term, but whether it actually helps reduce research time and cost without losing scientific reliability and without unnecessary risk in subsequent development stages.

What It Means

Mantis Biotech is betting on one of the most pragmatic directions of AI in medicine: data infrastructure, not another interface on top of a model. If the company can reliably build digital twins from disparate medical sources, it could accelerate drug development where the process is currently slowed by a lack of quality data. For the market, this is an important signal: the next wave of AI in healthtech may be built not around chatbots, but around a better research foundation.