Digital Twins in Medicine: Science Fiction or Modern Data Miracle?
The term “digital twin” may evoke images straight from the pages of science fiction, such as Nobel Prize winner Kazuo Ishiguro’s dystopian novel, Never Let Me Go, in which biological clones are raised solely for their organs. In reality, the concept of a digital twin in medicine—while still emerging—is no longer confined to fiction. The development of digital twins in silico (as computer models) presents ethical and unprecedented opportunities to advance medical science.
Digital twins have long been used in manufacturing and other industries as cost-effective tools for simulating design choices. NASA famously utilized digital twin concepts during the Apollo 13 mission, simulating spacecraft conditions to help bring astronauts safely back to Earth.
Today, advances in computer science and artificial intelligence (AI) are enabling medicine to explore digital twins for healthcare research and treatment.
What Are Digital Twins?
A digital twin is a highly detailed virtual replica of a physical entity. Unlike a static model, a digital twin is designed to mirror its real-world counterpart in real time, analyze its behavior, and provide predictive insights using advanced simulation, machine learning, and reasoning. This analytical and predictive capability distinguishes digital twins from simple replicas. Digital twins generally consist of three components: a physical entity, a virtual replica, and automated bi-directional data communications for real-time analysis.
In healthcare, digital twins can represent everything from entire populations to specific organs, such as the human heart. By creating virtual copies, healthcare professionals can conduct research or assist with diagnoses without risk to patients.
Digital Twin Successes in Medical Research
A recent study in Nature Cardiovascular Research demonstrates how digital twins can advance heart disease research. In the study, researchers created over 3,800 anatomically accurate digital hearts to investigate how age, sex, and lifestyle factors influence cardiac function. These digital twins were built using real patient data and ECG readings, with machine learning and AI automating much of the process.
The study found that differences in ECG readings between men and women are primarily due to heart size, not electrical conduction. Insights like these allow clinicians to refine treatments, such as tailoring heart device settings or identifying new drug targets for specific populations. As one researcher noted:
“The digital heart models we’ve built lay the foundation for the next step in our research – linking heart function to genes. This could help us understand how genetic variations influence heart function in a way that’s never been done before. This could lead to more precise and personalized care for patients in the future.”
Digital Twins for Surgical Planning
Digital twins are proving valuable in complex surgeries. Virtual models of anatomical structures allow surgeons to simulate procedures before operating. When combined with AI, entire surgeries can be simulated, enabling surgeons to plan by assessing different entry points, angles, and depths. For example, heart surgeons have used computer-simulated models of transcatheter aortic valve replacement to determine the best approach for individual patients.
At Johns Hopkins University, researchers are studying the use of digital twins to improve outcomes and reduce complications for ablation, a medical procedure for an irregular heart rhythm known as atrial fibrillation. By creating computer simulations of a patient’s heart based on preoperative images, surgeons can practice and plan the procedure, identifying target areas for ablation and implementing customized plans in the operating room.
Similarly, in orthopedics, digital twins help surgeons select optimal stabilization methods and postoperative treatments tailored to individual patients.
“When you have this model, you can personalize with certain features, certain anatomy, then you can try things. In heart surgery, you can’t try 20 different things; you only have one shot.”
Personalized Medicine
Digital twins are poised to revolutionize personalized medicine, allowing physicians to design individualized care plans. At Stanford University, researchers are studying how medical images and electronic health records can be combined to enhance cardiovascular risk predictions. Building a digital twin follows the same premise: “From a holistic perspective, can we simulate potential interventions, such as weight loss or starting a new treatment, to understand what will be best for a patient at a given time?”
In clinical oncology, digital twins are being explored to select the best treatment modalities based on simulated therapy outcomes.
Drug Development and Clinical Trials
Drug development is costly, and animal testing does not always yield clear data for humans. Pharmaceutical companies are investigating digital twins to reduce the number of participants needed for clinical trials, potentially saving billions and reducing patient exposure to unnecessary tests.
The FDA is cautiously open to this approach. In a May 2023 discussion paper, the FDA cited a sponsor’s proposal to use digital twins to generate “patient prognostic scores” for predicting placebo outcomes, enabling reduced placebo sample sizes in phase 2 and 3 trials. The FDA agreed to exploratory use in phase 2 and possibly primary analysis in phase 3.
The Path Forward
While digital twin technology is emerging as a powerful tool for medical research and personalized medicine, challenges remain. Building dynamic models using vast amounts of real-time data is costly. Data accuracy and quality are critical—incorrect or incomplete data can lead to unreliable insights. Data must be sourced from health records, imaging, wearables, and genetic databases, raising privacy and security concerns. Compliance with HIPAA, GDPR, and other regulations adds complexity.
Ethical considerations are also paramount. Health data may be skewed toward certain demographics, potentially exacerbating bias. Issues include informed consent, data ownership, patient autonomy, and healthcare equity.
Despite these challenges, the development and deployment of digital twins in medicine offer tremendous opportunities for research and customized healthcare. As researchers, clinicians, and technologists continue to collaborate and innovate, digital twins stand poised to redefine the boundaries of medical science—transforming challenges into opportunities and paving the way for a future where healthcare is truly personalized, predictive, and precise.
Authored by Maria-Cristina Smith, Berkley Life Sciences, VP, Products & Professional Liability Specialist