What question did this study set out to answer?

This study aims to enhance pleural effusion diagnosis by developing a deep-learning model that accurately segments lung ultrasound images.

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

View Full Paper

Ask AI

Bookmark

Authors

Follow-up study develops a deep learning model for segmenting pleural effusion in lung ultrasound images, highlighting its diagnostic potential.

This study aims to enhance pleural effusion diagnosis by developing a deep-learning model that accurately segments lung ultrasound images.
Utilized 3041 lung ultrasound images collected from 24 patients diagnosed with pleural effusion.
Ground truth for training was established by two lung ultrasound experts reviewing and segmenting the images.
The algorithm was trained using ten-fold cross-validation and evaluated against a separate patient subset.
Achieved an average Dice Similarity Coefficient (DSC) of 0.70 between the algorithm and expert segmentations.
Experts demonstrated an average DSC of 0.61 among themselves.
The trained algorithm showed promise in improving the accuracy of pleural effusion segmentation.

Vukovic et al. (2023) studied this question.

View Full Paper

Ask AI

Bookmark

Follow-up study develops a deep learning model for segmenting pleural effusion in lung ultrasound images, highlighting its diagnostic potential.