AI Digital Twins Aim to Shift Vascular Care From Episodic Diagnosis to Continuous Monitoring

Researchers have proposed a new framework designed to transform vascular medicine from a system of episodic diagnoses into a model of continuous, adaptive care. By integrating artificial intelligence with "digital twins"—computational representations of a patient's circulatory system—the approach treats panvascular disease as a dynamic network rather than a series of isolated blockages.

Led by Lingsen You and published in Light: Science & Applications, the research introduces a system that combines 3D anatomy, blood-flow dynamics, and tissue properties into a personalized physiological model. To feed these AI models, the framework utilizes intelligent photonics, including photoacoustic imaging, near-infrared spectroscopy, and optical coherence tomography (OCT). These tools provide the high-resolution data necessary to create an accurate digital replica of a patient's vessels. The ultimate goal is the integration of diverse data into clinically meaningful predictions rather than simple automated image reading.

The Panvascular Challenge

Traditional cardiovascular care has long relied on "snapshots," using occasional scans that can miss critical biological changes occurring between patient visits. While intravascular imaging tools like OCT and ultrasound provide detailed views of plaque and vessel walls, the sheer volume of data often creates "cognitive bottlenecks" for specialists attempting to interpret the results manually.

Furthermore, the proposed framework addresses the reality of "panvascular" disease. This perspective recognizes that inflammation and atherosclerosis do not occur in isolation but affect the entire circulatory system simultaneously, including the cerebral, renal, carotid, peripheral, and coronary vessels.

Implications for Patient Care

This shift toward adaptive vascular care could fundamentally change how clinicians manage high-risk patients. By utilizing a digital twin, doctors could monitor disease progression in real-time and test the efficacy of various interventions virtually before performing them on the actual patient. This capability would potentially reduce the need for invasive trial-and-error procedures and allow for more precise adjustments to medication.

The Path Forward

If successfully implemented, the system would enable the earlier detection of dangerous plaques that might otherwise go unnoticed between scheduled screenings. While the framework provides a roadmap for integrating photonics and AI, the transition from theoretical modeling to standard clinical practice will depend on the scalability of these digital twins across diverse patient populations and the seamless integration of high-resolution imaging into routine care workflows.