Abstract | ||
---|---|---|
Lung ultrasound (LUS) imaging is used to assess lung abnormalities, including the presence of B-line artefacts due to fluid leakage into the lungs caused by a variety of diseases. However, manual detection of these artefacts is challenging. In this paper, we propose a novel methodology to automatically detect and localize B-lines in LUS videos using deep neural networks trained with weak labels. To this end, we combine a convolutional neural network (CNN) with a long short-term memory (LSTM) network and a temporal attention mechanism. Four different models are compared using data from 60 patients. Results show that our best model can determine whether one-second clips contain B-lines or not with an F1 score of 0.81, and extracts a representative frame with B-lines with an accuracy of 87.5%. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1109/ISBI48211.2021.9434006 | 2021 IEEE 18TH INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING (ISBI) |
Keywords | DocType | ISSN |
Lung ultrasound (LUS), video analysis, classification | Conference | 1945-7928 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
9 |
Name | Order | Citations | PageRank |
---|---|---|---|
Hamideh Kerdegari | 1 | 0 | 0.68 |
Phung Tran Huy Nhat | 2 | 0 | 0.34 |
Angela McBride | 3 | 0 | 0.34 |
VITAL Consortium | 4 | 0 | 0.34 |
Reza Razavi | 5 | 0 | 0.34 |
Nguyen Van Hao | 6 | 0 | 0.34 |
Louise Thwaites | 7 | 0 | 0.34 |
Sophie Yacoub | 8 | 0 | 0.34 |
Alberto Gomez | 9 | 50 | 12.62 |