A method for constructing arterial digital twins for virtual vascular clinical trials

The method constructs arterial digital twins using CT angiography and advanced models to simulate vascular clinical trials, addressing the limitations of traditional trials by enabling efficient and accurate simulation of vascular pathologies and treatment outcomes, thus reducing costs and improving trial reliability.

CN118350298BActive Publication Date: 2025-03-28NANTONG UNIV
2 Cites 1 Cited by

Patent Information

Application Number
CN202410363150.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-03-28
Publication Date
2025-03-28
Estimated Expiration
2044-03-28

AI Technical Summary

Technical Problem

Traditional clinical trials are time-consuming and costly, with limited sample sizes leading to unreliable results and delayed development of new treatments, while requiring extensive resources and adherence to strict ethical and legal regulations, which can be challenging for research institutions. Additionally, existing virtual clinical trials face limitations in accurately simulating blood flow dynamics for vascular studies.

Method used

A method for constructing an arterial digital twin for virtual vascular clinical trials using CT angiography images, involving cycleGAN models for image augmentation, SWIN Transformer for automatic blood vessel segmentation, and PINN models for building reduced-order models, enabling the simulation of blood flow and treatment outcomes.

Benefits of technology

This approach allows for the construction of dynamic arterial digital twins that support virtual clinical trials with reduced data requirements, enabling the simulation of various pathologies and treatment schemes, predicting treatment effectiveness and reducing surgical risks while minimizing human and temporal costs.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present application discloses a method for constructing an arterial digital twin for a virtual vascular clinical trial, comprising the following steps: based on some sample CTA images, collecting vascular imaging information; based on the cycleGAN model, amplifying the CTA images of virtual patients to obtain a first training set; performing fully automatic vascular segmentation on the CTA images in the first training set to obtain vascular segmentation images; performing centerline extraction and grid division on the vascular images to obtain vascular grids; setting boundary conditions to construct a vascular model; constructing a full-order CFD model to obtain an input training set; and constructing a vascular twin reduced-order model. The present application can realize virtual clinical trials based on a small amount of clinical data by constructing an arterial digital twin model supported by real dynamic data. The method can simulate different pathological conditions and treatment plans, predict treatment effects and the probability of complications, reduce surgical risks, and reduce manpower and time costs.
Need to check novelty before this filing date? Find Prior Art