MultiFlow: A unified deep learning framework for multi-vessel classification, segmentation and clustering of phase-contrast MRI validated on a multi-site single ventricle patient cohort

2025-02-17

AI Summary

Prognostic assessment in single-ventricle congenital heart disease remains challenging due to anatomical complexity and heterogeneous hemodynamics. Method: We propose the first unified deep learning framework integrating multi-vessel classification, flow-image segmentation, and dynamic time-series clustering. Specifically: (1) MultiFlowSeg—a U-Net variant—achieves precise classification and segmentation of aorta (AO), superior/inferior vena cava (SVC/IVC), and left/right pulmonary arteries (LPA/RPA); (2) MultiFlowDTC, enhanced by dynamic time warping (DTW), enables robust phenotypic clustering on phase-contrast MRI (PC-MRI) time series; (3) validation is performed on the multicenter FORCE dataset, accommodating low-quality images and anatomical variants (e.g., dextrocardia). Results: Classification accuracy reaches 100% for AO/SVC/IVC and 94% for LPA/RPA; segmentation achieves a median Dice score of 0.91. Five hemodynamic subtypes with statistically significant prognostic divergence are identified—each associated with ejection fraction, exercise capacity, hepatic disease progression, and mortality—establishing a novel paradigm for personalized intervention.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Segmentation
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

This study presents a unified deep learning (DL) framework, MultiFlowSeg, for classification and segmentation of velocity-encoded phase-contrast magnetic resonance imaging data, and MultiFlowDTC for temporal clustering of flow phenotypes. Applied to the FORCE registry of Fontan procedure patients, MultiFlowSeg achieved 100% classification accuracy for the aorta, SVC, and IVC, and 94% for the LPA and RPA. It demonstrated robust segmentation with a median Dice score of 0.91 (IQR: 0.86-0.93). The automated pipeline processed registry data, achieving high segmentation success despite challenges like poor image quality and dextrocardia. Temporal clustering identified five distinct patient subgroups, with significant differences in clinical outcomes, including ejection fraction, exercise tolerance, liver disease, and mortality. These results demonstrate the potential of combining DL and time-varying flow data for improved CHD prognosis and personalized care.

Problem

• Unified DL framework for MRI data analysis.
• High accuracy in vessel classification and segmentation.
• Identifies patient subgroups for personalized CHD care.

Innovation

• Unified deep learning framework
• Automated multi-vessel segmentation
• Temporal clustering of phenotypes