Deep Learning-Based Algorithms for Real-Time Lung Ultrasound Assisted Diagnosis

Lung ultrasound is an increasingly utilized non-invasive imaging modality for assessing lung condition but interpreting it can be challenging and depends on the operator’s experience. To address these challenges, this work proposes an approach that combines artificial intelligence (AI) with feature-...

Full description

Saved in:
Bibliographic Details
Main Authors: Mario Muñoz, Adrián Rubio, Guillermo Cosarinsky, Jorge F. Cruza, Jorge Camacho
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/14/24/11930
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850036375016964096
author Mario Muñoz
Adrián Rubio
Guillermo Cosarinsky
Jorge F. Cruza
Jorge Camacho
author_facet Mario Muñoz
Adrián Rubio
Guillermo Cosarinsky
Jorge F. Cruza
Jorge Camacho
author_sort Mario Muñoz
collection DOAJ
description Lung ultrasound is an increasingly utilized non-invasive imaging modality for assessing lung condition but interpreting it can be challenging and depends on the operator’s experience. To address these challenges, this work proposes an approach that combines artificial intelligence (AI) with feature-based signal processing algorithms. We introduce a specialized deep learning model designed and trained to facilitate the analysis and interpretation of lung ultrasound images by automating the detection and location of pulmonary features, including the pleura, A-lines, B-lines, and consolidations. Employing Convolutional Neural Networks (CNNs) trained on a semi-automatically annotated dataset, the model delineates these pulmonary patterns with the objective of enhancing diagnostic precision. Real-time post-processing algorithms further refine prediction accuracy by reducing false-positives and false-negatives, augmenting interpretational clarity and obtaining a final processing rate of up to 20 frames per second with accuracy levels of 89% for consolidation, 92% for B-lines, 66% for A-lines, and 92% for detecting normal lungs compared with an expert opinion.
format Article
id doaj-art-6a8a948927c1496bbf658134e5ddc42d
institution DOAJ
issn 2076-3417
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-6a8a948927c1496bbf658134e5ddc42d2025-08-20T02:57:12ZengMDPI AGApplied Sciences2076-34172024-12-0114241193010.3390/app142411930Deep Learning-Based Algorithms for Real-Time Lung Ultrasound Assisted DiagnosisMario Muñoz0Adrián Rubio1Guillermo Cosarinsky2Jorge F. Cruza3Jorge Camacho4Institute for Physical and Information Technologies, Spanish National Research Council, 28006 Madrid, SpainInstitute for Physical and Information Technologies, Spanish National Research Council, 28006 Madrid, SpainInstitute for Physical and Information Technologies, Spanish National Research Council, 28006 Madrid, SpainInstitute for Physical and Information Technologies, Spanish National Research Council, 28006 Madrid, SpainInstitute for Physical and Information Technologies, Spanish National Research Council, 28006 Madrid, SpainLung ultrasound is an increasingly utilized non-invasive imaging modality for assessing lung condition but interpreting it can be challenging and depends on the operator’s experience. To address these challenges, this work proposes an approach that combines artificial intelligence (AI) with feature-based signal processing algorithms. We introduce a specialized deep learning model designed and trained to facilitate the analysis and interpretation of lung ultrasound images by automating the detection and location of pulmonary features, including the pleura, A-lines, B-lines, and consolidations. Employing Convolutional Neural Networks (CNNs) trained on a semi-automatically annotated dataset, the model delineates these pulmonary patterns with the objective of enhancing diagnostic precision. Real-time post-processing algorithms further refine prediction accuracy by reducing false-positives and false-negatives, augmenting interpretational clarity and obtaining a final processing rate of up to 20 frames per second with accuracy levels of 89% for consolidation, 92% for B-lines, 66% for A-lines, and 92% for detecting normal lungs compared with an expert opinion.https://www.mdpi.com/2076-3417/14/24/11930lung ultrasound (LUS)artificial intelligence (AI)convolutional neural networkdeep learningpleuraB-line
spellingShingle Mario Muñoz
Adrián Rubio
Guillermo Cosarinsky
Jorge F. Cruza
Jorge Camacho
Deep Learning-Based Algorithms for Real-Time Lung Ultrasound Assisted Diagnosis
Applied Sciences
lung ultrasound (LUS)
artificial intelligence (AI)
convolutional neural network
deep learning
pleura
B-line
title Deep Learning-Based Algorithms for Real-Time Lung Ultrasound Assisted Diagnosis
title_full Deep Learning-Based Algorithms for Real-Time Lung Ultrasound Assisted Diagnosis
title_fullStr Deep Learning-Based Algorithms for Real-Time Lung Ultrasound Assisted Diagnosis
title_full_unstemmed Deep Learning-Based Algorithms for Real-Time Lung Ultrasound Assisted Diagnosis
title_short Deep Learning-Based Algorithms for Real-Time Lung Ultrasound Assisted Diagnosis
title_sort deep learning based algorithms for real time lung ultrasound assisted diagnosis
topic lung ultrasound (LUS)
artificial intelligence (AI)
convolutional neural network
deep learning
pleura
B-line
url https://www.mdpi.com/2076-3417/14/24/11930
work_keys_str_mv AT mariomunoz deeplearningbasedalgorithmsforrealtimelungultrasoundassisteddiagnosis
AT adrianrubio deeplearningbasedalgorithmsforrealtimelungultrasoundassisteddiagnosis
AT guillermocosarinsky deeplearningbasedalgorithmsforrealtimelungultrasoundassisteddiagnosis
AT jorgefcruza deeplearningbasedalgorithmsforrealtimelungultrasoundassisteddiagnosis
AT jorgecamacho deeplearningbasedalgorithmsforrealtimelungultrasoundassisteddiagnosis