Digital Twin

What is a digital patient twin?

In healthcare, the digital patient twin is being researched and developed for specific organs like the heart and liver. But what, in fact, is a digital patient twin? Let's look into what this could be, and which applications contribute to its concept.

10 min

Peter Aulbach

Published on March 21, 2023

There are several concepts of digital twins currently being used in healthcare. One central approach is digital patient twins: With their help, doctors are increasingly able to better predict how a person ages, when illnesses appear, what course they take, and what the most effective treatment is.

Could a digital patient twin help sick people get back to normal life as soon as possible? Or could it even prevent people from getting sick in the first place? This sounds like a big dream from today’s perspective. See how it could actually work:

These simulations of a patient’s state of health are based on biophysiological data models created with algorithms. In principle, the digital patient twin might process recordable data received from a wide variety of sources without having to store the data in a single place – and ideally for the patient’s lifetime.

The digital patient twin would collect an individual’s existing health information in real-time and continually compares this data with the results from population studies, data on specific clinical pathologies, the course of specific diseases, and information on medications, diagnostics, and therapies used for other affected individuals.

Taking into account collected evidence, paired with clinical guidelines and economic considerations, the twin could enable doctors creating a holistic, individual, and comprehensive preventive or treatment regime.

For practical implementation of a digital patient twin, there are four essential requirements:

The hospitals or healthcare practices must be sufficiently digitally networked.
The data must be structured and annotated.
Patients must be able to decide about the use of their data at all times.
Medical professionals must have access to the information processed in the digital patient twin and be able to use this digital interface in everyday clinical practice.

There is still a long way to go before a digital patient twin could be fully realized in healthcare. But the technology itself is no longer science fiction. Powerful AI-based solutions that will most likely contribute to the concept of a digital patient twin are already in use:

: With the help of artificial intelligence, routine radiology processes can be fully automated, which may reduce image errors. Body and organ movements could be monitored with additional sensors, such as BioMatrix MR and Fast3D Cam CT, and evaluated with algorithms. Using this data, the computed tomography image may be (semi-)automatically planned to ensure optimal patient positions for image acquisition. This is particularly helpful for supporting a reliable diagnosis of clinicians.

Artificial intelligence also supports radiologists in evaluating large amounts of image data. Algorithms make it possible to systematize large numbers of image cases. Anatomies are automatically labeled, and the results are compared with reference values to highlight abnormalities (e.g., Syngo Carbon²). This allows the automatic extraction of “imaging biomarkers” and, for example, enables doctors to distinguish between benign and malignant structures. Artificial intelligence can routinely detect radiological abnormalities that are sometimes difficult or even impossible to see with the human eye and provide valuable assistance.
: Existing digital patient twin technology can also support doctors in preparing radiotherapies. With the help of artificial intelligence, tumor contrasts, borders, and volumes as well as tissue changes can be determined automatically. This saves up to one hour per patient when contouring the organs at risk. In fact, the best AI-based contouring results use specially derived raw data from the CT system that is not optimized for diagnoses by the human eye (e.g., ADMIRE CT, MR Deep Resolve).

Data models can be used to calculate whether and how patients will respond to radiation treatments. This allows the duration, dose, and frequency of the individual radiation treatments to be optimized, making the treatment of tumors even more efficient and effective. In a clinical study published in the journal Lancet in 2019, these advantages were demonstrated for the radiotherapy of lung cancer. The study showed that this method achieves a result that is up to 45 percent better than standard radiation treatments (Lou et. al. 2019).

AI and robot-assisted neuro interventions, based on digital patient twins, open new dimensions in the field of cerebrovascular therapy (e.g., Artis icono and Corindus robots). For the emergency treatment of strokes in the future, a catheter-based therapy such as a thrombectomy or coiling could be performed remotely to minimize the time between the first signs of the stroke and the therapeutic intervention.
: The translation and visualization of data into understandable findings, using artificial intelligence, is an important capability of digital patient twins. With cinematic rendering, a patient’s state of health can be presented realistically, and the treatment can be better prepared and carried out with the help of 3D organ models.

Thanks to virtual reality (VR) and augmented reality (AR), available clinical data can be interactively viewed, used, and edited. Surgeons already work with 3D glasses. Researchers have found that CT images created with cinematic rendering ensure a faster and more accurate understanding of a patient’s anatomy. With this method, surgeons were able to complete their work with greater accuracy and in roughly half the time. It is expected that this technique will also benefit surgical planning and related decision-making.

The added value of digital patient twins

The basis for using the digital patient twin is a running digital infrastructure. Although the technology already exists, it will take many years to translate these concepts into clinical applications. So, it is especially important to pave the way in hospitals today and continuously implement partial solutions as they become available.

The added value of the digital patient twin concept for hospitals will come from the multitude of possible applications in medicine and administration that optimize the healthcare pathway. This technology will be indispensable in clinical routines for translating the growing amount of data into decision-relevant knowledge. Ultimately, it is expected to ensure that the process of care before, during, and after a clinical treatment will be even more individualized and precise.

Learn more about digital patient twins

Portrait of Chloé Audigier

Digital Twin

Modeling the human liver

Digital Twin

A patient twin to protect John’s heart

Digital Twin

“Digital Twins will be essential to make precision medicine a reality”

Digital Twin

How my digital self helped me survive cancer

#Futureshaper

Shikha's journey to the "patient twin"

Digital Twin

Humanizing medtech

Sources: Lou B, Doken S, Zhuang T, Wingerter D, Gidwani M, Mistry N, Ladic L, Kamen A, Abazeed ME (2019) An image-based deep learning framework for individualising radiotherapy dose: a retrospective analysis of outcome prediction. The Lancet Digital Health 1, 136-147.

By Peter Aulbach

The products/features mentioned herein are not commercially available in all countries. Their future availability cannot be guaranteed.

Syngo Carbon consists of several products which are (medical) devices in their own right. Some products are under development and not commercially available. Future availability cannot be ensured.