Mantis Biotech is building synthetic human datasets that power physics-based digital twins โ predictive models of anatomy, physiology, and behavior. The technology is advancing fast. The health systems that could benefit most are not keeping pace.
While the promise of digital twins in healthcare draws growing attention from investors and researchers, a quieter problem is hardening beneath the surface: most providers do not yet have the data infrastructure, governance frameworks, or IT maturity to deploy these tools at scale.
How Digital Twins in Healthcare Are Being Built โ and Why Data Is the Bottleneck
According to TechCrunch, Mantis Biotech’s platform integrates disparate data sources to produce synthetic datasets. It then uses an LLM-based system to route, validate, and synthesize those streams before running them through a physics engine to generate high-fidelity human models.
Mantis founder and CEO Georgia Witchel described one use case to TechCrunch: a sports team predicting the likelihood of a specific NFL player developing an Achilles heel injury, drawing on recent performance, training load, diet, and career length. The model is specific, dynamic, and data-hungry.
That hunger is exactly what most clinical environments cannot currently satisfy. Patient records remain fragmented across systems, consent frameworks vary by jurisdiction, and real-world health datasets are riddled with gaps โ the very problem Mantis says its synthetic data approach is designed to solve.
Concrete Capabilities, Real Constraints
The benefits of the approach are not hypothetical. Business Insider reports that NASA’s Perseverance rover relies on AI-enhanced digital twins to navigate Mars terrain, while engineers use the same technique to monitor the James Webb Space Telescope. With AI layered on, digital twins can now make predictions, diagnose issues, and recommend actions in real time โ capabilities that took decades to develop.
NASA mission systems engineer Julie Van Campen noted that the Webb telescope’s digital twins were first developed in the early 2000s, long before modern AI existed. The expertise built over those years is only now being applied to next-generation missions. Healthcare is arguably earlier in that same curve.
Outside medicine, Dark Reading reports that JPMorgan Chase uses digital fingerprints and digital twins for cybersecurity threat hunting โ analyzing flagged behavioral anomalies and projecting how patterns might evolve over time. The bank’s system helps analysts sort through vast employee and AI-agent log data while reducing false alerts. Financial services, it turns out, built the data discipline healthcare still lacks.
The Infrastructure Problem Nobody Is Rushing to Fix
Healthcare IT leaders are increasingly aware of the readiness gap. According to MobiHealthNews, HIMSS Changemaker awardee Sepi Browning advises that organizations new to healthcare IT must first evaluate their readiness, governance, and infrastructure before selecting technology โ not after. The sequence matters, and it is frequently reversed in practice.
Patrick Bizeau, CIO of Swiss Medical Network and a HIMSS Senior Executive Changemaker Awardee, frames the problem differently: healthcare IT work begins at the bedside. Digital tools must reduce friction for clinicians and patients, not add to it. A digital twin that requires clean, continuous, multi-source data feeds is of little use in a ward where basic EHR interoperability is still unsolved.
Helsinki-based Digital Workforce is approaching the problem from the workflow side rather than the modeling side. CEO Jussi Vasama told MobiHealthNews the company uses robotic process automation and agentic AI to streamline healthcare and social care pathways โ supporting clinicians on productivity, long-term condition follow-up, and patient safety. It is a more incremental path, but one that engages the data-generating layer where it actually exists.
What to Watch: Adoption Gaps and Accountability Questions
The central open question is not whether digital twins in healthcare work โ the technical case is increasingly solid. The question is whether health systems will invest in the unglamorous prerequisite work: data governance, interoperability standards, and IT workforce development.
Synthetic data generation, as Mantis Biotech is pursuing, may offer a partial workaround by reducing dependence on complete real-world datasets. But synthetic data introduces its own validation challenges, particularly for regulatory approval of clinical decision tools. No synthetic dataset has yet cleared a major regulatory pathway for direct patient care use.
Meanwhile, the aerospace and financial sectors continue pulling ahead. Both industries built their digital twin capabilities on decades of structured, well-governed data collection โ a discipline healthcare has historically underinvested in. As long as that investment gap persists, the most sophisticated predictive models in medicine will remain tools for well-resourced research environments, not frontline care.
FAQ – Frequently Asked Questions
What are the key data governance frameworks required for deploying digital twins in healthcare?
Effective data governance for digital twins involves establishing clear policies on data ownership, consent management, and data quality. This includes implementing standards for data anonymization, pseudonymization, and secure data sharing. Healthcare organizations can refer to frameworks like the Data Governance Institute or the HIMSS Data Governance Toolkit for guidance.
How can healthcare providers assess their current IT maturity for digital twin adoption?
Healthcare providers can assess their IT maturity by evaluating their existing infrastructure’s ability to support real-time data integration, analytics, and AI-driven insights. They should also consider their organization’s ability to scale and adapt to new technologies. Utilizing IT maturity assessment tools, such as the HIMSS EMRAM (Electronic Medical Record Adoption Model) or the Gartner IT Maturity Model, can provide a comprehensive evaluation.
What are the potential return on investment (ROI) metrics for digital twin implementations in healthcare?
ROI metrics for digital twin implementations can include reduced hospitalization rates, improved patient outcomes, and enhanced operational efficiency. Healthcare organizations can also track metrics such as reduced length of stay, improved patient satisfaction, and decreased costs associated with preventable complications. Establishing clear ROI metrics will help organizations evaluate the effectiveness of their digital twin investments.
Last Updated on March 30, 2026 8:01 pm by Laszlo Szabo / NowadAIs | Published on March 30, 2026 by Laszlo Szabo / NowadAIs
