COVID-BLUeS -- A Prospective Study on the Value of AI in Lung Ultrasound Analysis

2025-09-09

AI Summary

Existing lung ultrasound (LUS) AI models suffer from poor generalizability due to low-quality data, frame-level modeling that disregards temporal dynamics, and insufficient integration of clinical context. Method: We prospectively collected high-fidelity LUS video sequences and matched complete blood count (CBC) data from 63 suspected COVID-19 patients, establishing the first publicly available multimodal LUS-CBC benchmark dataset. We further propose an end-to-end video–clinical multimodal fusion framework enabling zero-shot transfer. Contribution/Results: Our multimodal model achieves 79% accuracy in COVID-19 identification—significantly outperforming unimodal baselines and clinician annotations. The study demonstrates the feasibility of AI-assisted LUS analysis for both infection detection and severity assessment, while revealing critical cross-site generalization bottlenecks. It provides a foundational dataset, architectural paradigm, and empirical validation for clinically deployable LUS AI systems.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

As a lightweight and non-invasive imaging technique, lung ultrasound (LUS) has gained importance for assessing lung pathologies. The use of Artificial intelligence (AI) in medical decision support systems is promising due to the time- and expertise-intensive interpretation, however, due to the poor quality of existing data used for training AI models, their usability for real-world applications remains unclear. In a prospective study, we analyze data from 63 COVID-19 suspects (33 positive) collected at Maastricht University Medical Centre. Ultrasound recordings at six body locations were acquired following the BLUE protocol and manually labeled for severity of lung involvement. Several AI models were applied and trained for detection and severity of pulmonary infection. The severity of the lung infection, as assigned by human annotators based on the LUS videos, is not significantly different between COVID-19 positive and negative patients (p = 0.89). Nevertheless, the predictions of image-based AI models identify a COVID-19 infection with 65% accuracy when applied zero-shot (i.e., trained on other datasets), and up to 79% with targeted training, whereas the accuracy based on human annotations is at most 65%. Multi-modal models combining images and CBC improve significantly over image-only models. Although our analysis generally supports the value of AI in LUS assessment, the evaluated models fall short of the performance expected from previous work. We find this is due to 1) the heterogeneity of LUS datasets, limiting the generalization ability to new data, 2) the frame-based processing of AI models ignoring video-level information, and 3) lack of work on multi-modal models that can extract the most relevant information from video-, image- and variable-based inputs. To aid future research, we publish the dataset at: https://github.com/NinaWie/COVID-BLUES.

Problem

• Evaluating AI accuracy in diagnosing COVID-19 via lung ultrasound
• Assessing limitations of existing AI models for real-world LUS applications
• Identifying data heterogeneity and multimodal integration challenges in LUS analysis

Innovation

• AI models for lung infection detection
• Multi-modal combining images and CBC
• Frame-based processing ignoring video information