Regional quality estimation for echocardiography using deep learning

2024-08-01

arXiv.org

AI Summary

Existing cardiac ultrasound image quality assessment methods yield only global scores, failing to distinguish between anatomical view correctness and intrinsic image quality, and lacking support for segment-level analysis—limiting their clinical utility. To address this, we propose the first myocardial-segment-oriented regional quality assessment paradigm. Our method integrates U-Net–based segmentation, generalized contrast-to-noise ratio (gCNR) computation, B-mode local coherence modeling, and an end-to-end convolutional network to enable fine-grained, interpretable quality quantification. Evaluated against expert annotations, our approach achieves a Spearman correlation coefficient of 0.69—surpassing conventional metrics and approaching the inter-observer agreement among three cardiologists (0.63). The framework has been open-sourced as part of the arqee toolkit, enabling real-time sonographer guidance and enhancing the reliability of quantitative echocardiographic measurements.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Segmentation
• Knowledge Representation and Reasoning: Qualitative Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success

Abstract

Automatic estimation of cardiac ultrasound image quality can be beneficial for guiding operators and ensuring the accuracy of clinical measurements. Previous work often fails to distinguish the view correctness of the echocardiogram from the image quality. Additionally, previous studies only provide a global image quality value, which limits their practical utility. In this work, we developed and compared three methods to estimate image quality: 1) classic pixel-based metrics like the generalized contrast-to-noise ratio (gCNR) on myocardial segments as region of interest and left ventricle lumen as background, obtained using a U-Net segmentation 2) local image coherence derived from a U-Net model that predicts coherence from B-Mode images 3) a deep convolutional network that predicts the quality of each region directly in an end-to-end fashion. We evaluate each method against manual regional image quality annotations by three experienced cardiologists. The results indicate poor performance of the gCNR metric, with Spearman correlation to the annotations of rho = 0.24. The end-to-end learning model obtains the best result, rho = 0.69, comparable to the inter-observer correlation, rho = 0.63. Finally, the coherence-based method, with rho = 0.58, outperformed the classical metrics and is more generic than the end-to-end approach. The image quality prediction tool is available as an open source Python library at https://github.com/GillesVanDeVyver/arqee.

Problem

• Distinguish echocardiogram view correctness from image quality
• Provide regional image quality estimation instead of global
• Evaluate deep learning methods for automatic quality assessment

Innovation

• U-Net segmentation for myocardial regions
• Local image coherence via U-Net prediction
• End-to-end convolutional network quality prediction