Today's landscape of human digital twinning is evolving and brings a wide array of different technologies and approaches. In its essence, a digital twin involves creating a digital replica of a physical entity. Simulation in industry is not a new concept; NASA has been creating digital twins since the Apollo 13 mission1.
In healthcare, the scale of the model and interactions are more complex, but the aims are the same: to prevent disease and monitor patients more effectively. We create detailed models of patient-specific pathophysiology that can interact and respond to new input data as it arises. This technology holds immense potential for personalizing and enhancing patient care in a safe environment, enabling doctors to predict outcomes, plan and monitor treatments, and understand complex medical conditions much more effectively.
At one extreme, we see traditional models dominating the space, particularly in physiology engines2. These models, based on a top-down approach, are used throughout medical education, providing trainees with a foundational understanding of human physiology in the context of simulated scenarios. However, their application to clinical practice is limited. The top-down approach to these models, alongside their inability to adapt to the unique physiology of individual patients, restricts their utility.
On the other end of the spectrum, we find ambitious research and development initiatives aimed at constructing highly detailed models of human physiology. These models are characterized by their complexity and the inclusion of millions of parameters3. Yet, their application in everyday healthcare is hindered by the immense computational power required to run these models in real-world clinical settings. This makes them currently impractical for widespread clinical use.
There is a need for practical, scalable digital twinning solutions in healthcare that work in a fast-paced environment. AIBODY's approach is to strike an optimal balance between specific digital twin models and usability. We take 2D trans-thoracic echocardiogram data and produce anatomically accurate patient-specific models combined with a tailored haemodynamic model unique to the patient.
One of the standout features of AIBODY's digital twins is the simple integration into existing clinical workflows by providing access to our digital twins through the web browser. In the fast-paced environment of a hospital, where time is of the essence, it is essential for any healthcare technology application. AIBODY leverages simple diagnostic methods that clinicians can readily access, ensuring that the adoption of our digital twins doesn't disrupt existing processes but rather enhances them.
The digital twin approach, performance, and integration simplicity of AIBODY's solutions are what set us apart. By focusing on these aspects, we ensure that our digital twins are not just theoretical models but practical tools that clinicians can use to make informed decisions, tailor treatments to individual patient needs, and ultimately improve patient outcomes.
We are excited to see where the future of digital twins takes us; at AIBODY, we are at the forefront, bridging the gap between theoretical models and practical clinical applications. Our digital twinning solutions represent a significant leap towards a future where personalized, data-driven healthcare is not just a possibility but a reality.
Dr Aaron Smith
Chief Medical Officer
MBChB BSc MRCP
1. Allen BD. Digital Twins and Living Models at NASA [Internet]. [cited 2023 Dec 4]. Available from: https://ntrs.nasa.gov/citations/20210023699
2. Bray A, Webb JB, Enquobahrie A, Vicory J, Heneghan J, Hubal R, et al. Pulse Physiology Engine: an Open-Source Software Platform for Computational Modeling of Human Medical Simulation. SN Compr Clin Med. 2019 May 1;1(5):362–77.
3. Vázquez M, Arís R, Aguado-Sierra J, Houzeaux G, Santiago A, López M, et al. Alya Red CCM: HPC-Based Cardiac Computational Modelling. In: Klapp J, Ruíz Chavarría G, Medina Ovando A, López Villa A, Sigalotti LDG. Selected Topics of Computational and Experimental Fluid Mechanics. Cham: Springer International Publishing; 2015. p. 189–207.