Revolutionizing Surgical Preparedness: Personalized Preoperative Planning with Digital Twins of an Organ

Dear Editor,

In general surgical practice, traditional methods of planning and risk assessment have remained critical and astoundingly relevant. However, it has had its set of challenges and limitations. Surgical interventions, often lifesaving, inherently carry risks ranging from the spread of infection to more severe complications. ¹ Literature has shown that the increasing number of elderly individuals, who often have many simultaneous medical conditions, makes preoperative decision-making more complex, which ultimately poses a higher risk of an unsuccessful procedure.^1,2 While acknowledging that not all complications are avoidable through advancing technology, there exists a method that may contribute towards decreasing the likelihood of such adverse events.

Advanced medical imaging, which involves commonly used modalities for surgical planning such as magnetic resonance imaging (MRI) and computed tomography (CT), is unable to accurately portray the complex anatomical structures, leaving physicians to conceptualize two dimensional (2D) reconstructions into three-dimensional (3D). Virtual 3D renderings have made advancements in the efficacy of preoperative planning yet have often failed to replicate the reality of complete patient anatomy. ³ In addition, this intangibility makes surgical simulation difficult, especially for complicated surgical procedures. ² Considering these shortcomings, surgical training has started integrating the use of machine learning, such as artificial intelligence (AI) and virtual reality (VR). However, latest research has shown immersive VR technology to have minimal effect on short-term patient outcomes. Furthermore, the current body of evidence is restricted by low-quality and heterogeneous studies, hindering our understanding of its effects on preoperative planning, and limiting the conclusions that can be drawn where it helps surgeons improve anatomy visualization. ⁴

Precision medicine, on the other hand, is a pioneering approach that leverages data on an individual’s genomic, environmental, and lifestyle factors for personalized medical care. A digital twin, within the framework of precision medicine, is a 3D virtual representation or model of an individual patient, organ, or biological system created using the above-mentioned information along with medical imaging, electronic health records (EHRs), and physiological measurements.

We propose the application of precision medicine through digital twins to enhance personalized preoperative surgical planning. Their potential to predict situations in a virtual environment before any real implementation promises to reduce risks and save costs. ² A unified virtual model may be formed by integrating complementary information from multiple imaging modalities, such as MRI, CT, ultrasound, and molecular imaging, to form a more comprehensive and holistic 3D view of a patient’s anatomy and physiology. 3D models are already in use in various procedures. Computer facilitation is increasingly being used in arthroplasty as 3D imaging during preoperative planning exhibits the potential of tailoring bone cuts to meet every patient’s individual needs. This could especially contribute to the more extended durability of implants in younger people. ⁵ However, in addition to 3D visualization, digital twins offer a much more fluid depiction, allowing simulations, predictive analysis, and risk assessment, processes that are not possible with static 3D images.

Using this technology, surgeons can run simulations on a digital twin before surgery, enabling them to explore different surgical approaches, techniques, and scenarios while also assessing their viability, practicality, and potential outcomes specific to a particular patient in a risk-free virtual environment. It has been demonstrated that a patient-specific mathematical model could correctly predict the risk of post-hepatectomy portal hypertension, a key determinant of liver failure and, hence, post-operative mortality. ⁶ Most notably, a recent open-label experimental study demonstrated that mixed-reality anatomic cardiac digital twins significantly influenced and improved surgical team planning, with no adverse effects on post-surgery outcomes. ⁷

The application of digital twins in surgery, however, raises essential ethical considerations. Patient data if securely anonymized before integration into digital twin models can possibly reduce patient privacy concerns. A ‘process oriented ethical map’ can help developers of digital twins see an overview of all major ethical implications before starting. ⁸ Datasets can be routinely checked for diversity to enhance model accuracy across different populations to potentially mitigate bias. Technically, creating and sustaining digital twins requires high data fidelity and substantial computational resources. As this technology advances, future research will likely focus on incorporating AI to refine precision and enhance simulation capabilities, establishing new standards for tailored approaches in patient care and surgical planning.