Articolo in rivista, 2023, ENG, 10.3390/bioengineering10050555

Efficient lung ultrasound classification

Bruno A.; Ignesti G.; Salvetti O.; Moroni D.; Martinelli M.

CNR-ISTI, Pisa, Italy; CNR-ISTI, Pisa, Italy; CNR-ISTI, Pisa, Italy; CNR-ISTI, Pisa, Italy; CNR-ISTI, Pisa, Italy

A machine learning method for classifying Lung UltraSound is here proposed to provide a point of care tool for supporting a safe, fast and accurate diagnosis, that can also be useful during a pandemic like as SARS-CoV-2. Given the advantages (e.g. safety, rapidity, portability, cost-effectiveness) provided by the ultrasound technology over other methods (e.g. X-ray, computer tomography, magnetic resonance imaging), our method was validated on the largest LUS public dataset. Focusing on both accuracy and efficiency, our solution is based on an efficient adaptive ensembling of two EfficientNet-b0 models reaching 100% of accuracy, which, to our knowledge, outperforms the previous state-of-the-art. The complexity of this solution keeps the number of parameters in the same order as an EfficientNet-b0 by adopting specific design choices that are adaptive ensembling with a combination layer, ensembling performed on the deep features, minimal ensemble only two weak models. Moreover, a visual analysis of the saliency maps on sample images of all the classes of the dataset reveals where the focus is on an inaccurate weak model versus an accurate model.

Bioengineering (Basel) 10 (5)

Keywords

Convolutional Neural Networks, EfficientNet, Lung Ultrasound, SARS-CoV-2, COVID-19, Pneumonia, Ensemble, Computer Vision, Supervised Learning, Deep Learning

CNR authors

Salvetti Ovidio, Bruno Antonio, Ignesti Giacomo, Martinelli Massimo, Moroni Davide

CNR institutes

ISTI – Istituto di scienza e tecnologie dell'informazione "Alessandro Faedo"

ID: 478917

Year: 2023

Type: Articolo in rivista

Creation: 2023-03-09 15:23:20.000

Last update: 2023-06-13 09:36:07.000

External IDs

CNR OAI-PMH: oai:it.cnr:prodotti:478917

DOI: 10.3390/bioengineering10050555

ISI Web of Science (WOS): 000995582700001

Scopus: 2-s2.0-85160710809