In this Q&A, Professor Huanmei Wu of Temple’s Barnett College of Public Health discusses digital twin technology and how it has the potential to help doctors better predict disease and improve personalized care.

Imagine having a virtual version of yourself that allows doctors to predict how a devastating disease you may be developing could unfold before symptoms take a serious turn. For people living with amyotrophic lateral sclerosis (ALS)—a disease with no cure and limited life expectancy—one of medicine’s biggest challenges is predicting how the disease is likely to progress from person to person.  

To address that challenge, Temple Health’s ALS clinical team—with support from researchers at the Barnett College of Public Health—is using digital twin technology to create a dynamic virtual replica of a person’s ALS conditions, leveraging real-time and longitudinal data to simulate disease scenarios and predict outcomes. Through Temple’s DT4PM project, clinicians and researchers are building AI-enabled digital twins of people living with ALS—continuously updated models designed to anticipate disease progression allow physicians to forecast more precise treatment strategies before applying them in real life.  

Temple Now spoke with Huanmei Wu, professor and department chair of health services administration and policy at the Barnett College of Public Health and co-leader of the ALS research, to answer questions about how digital twin technology works, its role in Temple’s research, and how it can potentially help patients make more informed decisions related to both their healthcare in general and their future. 

Temple Now: What is a digital twin and how can it impact healthcare? 

Huanmei Wu: A human digital twin in healthcare is a dynamic, virtual representation of a real individual, built through multiscale modeling of multimodal data, including clinical records, imaging, genomics, wearable sensor data, lifestyle behaviors, treatment history, environmental exposures and social determinants of health. Unlike a static medical record, a digital twin continuously updates as new data becomes available, which enables clinicians and patients to simulate “what-if” scenarios, such as forecasting disease progression or estimating how a specific treatment may affect outcomes. In health, they support dynamic treatment planning, real-time monitoring, early risk detection and prevention strategies. They also enhance shared decision-making and can improve care efficiency while reducing costs. Ultimately, digital twins help deliver the right intervention to the right patient at the right time. 

TN: We have heard about your digital twin project with ALS patients and the impact it is having. What other conditions can digital twin research be used for? 

HW: ALS is a powerful example for digital twin research because it is a progressive and highly variable neurological disease, where personalized modeling can significantly improve progression monitoring, care planning and prognosis. Beyond ALS, digital twins for health are especially valuable for other complex and heterogeneous conditions, including cancer, diabetes, cardiovascular diseases like heart failure, neurodegenerative diseases such as Parkinson’s and Alzheimer’s, and other rare diseases. Where patients respond differently to treatment or experience diverse disease trajectories, digital twins can help tailor care to the individual. In addition, digital twins can be used to simulate hospital operations, optimize clinical workflows, evaluate medication effectiveness, enhance clinical trial design and support training for healthcare professionals. 

TN: Who is needed on a team when doing this work? Can you talk about the collaborative nature of it? 

HW: Digital twins in healthcare are effective only when technology, clinical expertise and patient values are fully integrated. No single field can alone build a meaningful digital twin, which makes this work inherently interdisciplinary. A strong team includes clinicians (to define relevant clinical questions and assess outcomes), data scientists and AI/ML experts (to develop predictive models), biomedical engineers (to acquire and translate biological signals), health informatics specialists (to integrate and manage health data), software engineers (to build the technical infrastructure), biostatisticians (to ensure scientific rigor), ethicists and privacy experts (to protect patient data), and patient representatives (to ground the work in real-world needs). 

TN: What sparked your interest in digital twins and how long have you been doing this work? 

HW: I have been working on digital twins for healthcare since late 2021, and on ALS digital twins since 2024.  

My interest in digital twins for health has evolved over the years, alongside my interdisciplinary research projects. It began with recognizing the gap between the vast amount of health data we have collected and the difficult decisions patients still face. Although we generate enormous volumes of data, only a fraction is meaningfully used in real-time clinical care. Working on interdisciplinary teams allowed me to see healthcare challenges from multiple perspectives, such as those of clinicians, nurses, informaticians, medical physicists, engineers, social workers and patients. I was particularly drawn to complex, progressive diseases where uncertainty heavily impacts patients and families. In addition, my collaboration with national and international partners further shaped my vision. Together, we have worked across the spectrum from building AI-ready data infrastructure to developing clinical decision support tools. Through this work, the idea of creating a living, dynamic model of a patient’s health with digital twins to predict outcomes rather than simply reacting to them felt transformative and ultimately led me to focus my recent efforts on digital twins for health. 

TN: How important can digital twin work be moving forward when it comes to helping patients make decisions related to both their care and future? 

HW: Digital twins have the potential to transform shared decision-making fundamentally. Currently, many medical decisions rely on population averages, such as “patients like you typically experience …” However, patients want to understand what is likely to happen to them as individuals. Digital twins can provide personalized insights based on individual data and personal preferences. They can forecast individualized disease progression, estimate patient-specific treatment responses, and help plan long-term care while considering social and economic circumstances. It is especially significant for chronic and life-altering diseases, where decisions extend beyond treatment to career planning, family life, finances and long-term goals. In addition, digital twins are designed as decision-support tools, not decision-makers. If developed responsibly and equitably, digital twins could become one of the most impactful advances in future precision medicine and personalized healthcare.