Automatic deep learning-based pleural effusion segmentation in lung ultrasound images

Damjan Vukovic, Andrew Wang, Maria Antico, Marian Steffens, Igor Ruvinov, Ruud J.G. van Sloun, David Canty, Alistair Royse, Colin Royse, Kavi Haji, Jason Dowling, Girija Chetty, Davide Fontanarosa

Research output: Contribution to journalArticlepeer-review


Point-of-care lung ultrasound (LUS) allows real-time patient scanning to help diagnose pleural effusion (PE) and plan further investigation and treatment. LUS typically requires training and experience from the clinician to accurately interpret the images. To address this limitation, we previously demonstrated a deep-learning model capable of detecting the presence of PE on LUS at an accuracy greater than 90%, when compared to an experienced LUS operator.

This follow-up study aimed to develop a deep-learning model to provide segmentations for PE in LUS. Three thousand and forty-one LUS images from twenty-four patients diagnosed with PE were selected for this study. Two LUS experts provided the ground truth for training by reviewing and segmenting the images. The algorithm was then trained using ten-fold cross-validation. Once training was completed, the algorithm segmented a separate subset of patients.

Comparing the segmentations, we demonstrated an average Dice Similarity Coefficient (DSC) of 0.70 between the algorithm and experts. In contrast, an average DSC of 0.61 was observed between the experts.
Original languageEnglish
Article number274
Pages (from-to)1-14
Number of pages14
JournalBMC Medical Informatics and Decision Making
Issue number2023
Publication statusPublished - 29 Nov 2023


Dive into the research topics of 'Automatic deep learning-based pleural effusion segmentation in lung ultrasound images'. Together they form a unique fingerprint.

Cite this