颈动脉结构与功能超声评估一体化软件平台：开发与可行性评价

BACKGROUND: The extensive assessment of vascular health requires the integration of both structural and functional (hemodynamic) parameters. These are usually obtained by fragmented technologies and different signals (e.g., imaging, tonometry). In this study, we aimed to (i) develop a novel software platform that enables a combined ultrasound-based analysis of longitudinal scans of the carotid; (ii) provide an initial evaluation of its technical performance and usability.

METHODS: The development of the new platform was based on two validated software medical devices, the ARCSolver and the Carotid Studio, with novel components and integration efforts. The integration was achieved through the design of a unified processing pipeline and a shared data architecture, allowing seamless interoperability between the two systems. Moreover, besides the structural ultrasound-based analysis derived by Carotid Studio (i.e., Intima-Media-Thickness and diameter), a new generalized transfer function (newGTF) was developed in order to obtain ultrasound-based aortic waveforms and the ARCSolver-based algorithms were applied to obtain hemodynamic parameters (i.e., Pulse Wave Analysis). A study was conducted to evaluate the performance of the integrated system, and of the new GTF, with repeated acquisitions on 14 healthy volunteers with pairwise ultrasound and reference tonometric acquisitions at the right common carotid artery. Accuracy of the new GTF was assessed against central pulse waveforms measured by the tonometric reference method, and precision was evaluated through the repeatability of central hemodynamic parameters. Finally, a small group of medical operators tested usability of the integrated solution.

RESULTS: According to the identified aims: (i) The resulting software provides an intuitive graphical user interface that enables comprehensive carotid analysis from a single acquisition. It processes longitudinal ultrasound scans to derive local structural parameters, including diameter, and applies mathematical models combined with a machine-learning-based approach to transform diameter waveforms into aortic pressure curves; (ii) the findings demonstrated good repeatability of the functional derived parameters, correlation with the reference approach and excellent usability, confirming the system's potential for clinical and research applications.

CONCLUSION: The developed integrated solution may facilitate more comprehensive vascular assessment by enabling multiple measurements to be obtained within a single software environment.