# A comprehensive workflow for CCTA and OCT data fusion with 3D printing validation: advancing patient-specific testing environments for percutaneous coronary intervention devices

## TL;DR

This paper introduces a new method to create detailed 3D models of coronary arteries using medical imaging data, validated through 3D printing for testing heart catheterization devices.

## Contribution

A novel workflow for fusing CCTA and OCT data to produce high-resolution, patient-specific coronary artery phantoms for device testing.

## Key findings

- 15 patient-specific 3D coronary artery models were successfully created with detailed anatomical features.
- 13 out of 15 3D printed phantoms were validated with a median vertex deviation of 0.15 mm.
- The median absolute stenosis difference between models was 3%.

## Abstract

To create high-resolution, patient-specific 3D coronary artery models aimed at developing digital models and functional phantoms for the testing of cardiac catheterization devices.

Using coronary computed tomography angiography (CCTA) and optical coherence tomography (OCT), coronary artery lesions were identified and quantified. Imaging data were fused using a custom-made workflow to create highly accurate digital 3D models. For validation of the workflow, coronary artery phantoms were fabricated using additive manufacturing. An OCT was then conducted on the 3D printed phantom, and the developed workflow was applied to generate a derivative model, which was subsequently compared to the original.

CCTA and OCT datasets from 15 patients were successfully collected and used to develop patient-specific 3D coronary artery models, including detailed inner shells, calcifications, outer wall structures, and side branches. Of these, 13 out of 15 3D printed phantoms were successfully validated and compared to their corresponding original model. The median vertex deviation of the derivative model was 0.15 (0.14\documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$-$$\end{document}-0.17) mm. The median absolute stenosis difference between the derivative model and the original model was 3 (1–5)%AS.

We present a novel workflow to produce high-resolution patient-specific phantoms of coronary arteries.