Abstract
The presence of B-line artefacts, the main artefact reflecting lung abnormalities in dengue patients, is often assessed using lung ultrasound (LUS) imaging. Inspired by human visual attention that enables us to process videos efficiently by paying attention to where and when it is required, we propose a spatiotemporal attention mechanism for B-line detection in LUS videos. The spatial attention allows the model to focus on the most task relevant parts of the image by learning a saliency map. The temporal attention generates an attention score for each attended frame to identify the most relevant frames from an input video. Our model not only identifies videos where B-lines show, but also localizes, within those videos, B-line related features both spatially and temporally, despite being trained in a weakly-supervised manner. We evaluate our approach on a LUS video dataset collected from severe dengue patients in a resource-limited hospital, assessing the B-line detection rate and the model’s ability to localize discriminative B-line regions spatially and B-line frames temporally. Experimental results demonstrate the efficacy of our approach for classifying B-line videos with an F1 score of up to 83.2% and localizing the most salient B-line regions both spatially and temporally with a correlation coefficient of 0.67 and an IoU of 69.7%, respectively.
Highlights
Introduction published maps and institutional affilDengue is a viral disease that can transmit to humans through the bites of an infected female Aedes genus mosquito [1]
We carried out five types of experiments: first, experiments to evaluate our proposed method on the B-line classification task using the lung ultrasound (LUS) video dataset described earlier; second, experiments to evaluate the spatial attention mechanism on the B-line spatial localization task; third, experiments to evaluate the temporal attention mechanism on the
B-line temporal localization task; fourth, experiments to investigate the effect of different video lengths on model performance during the training and testing phase; and fifth, several ablation experiments were conducted to evaluate the performance of components presented in our model
Summary
Introduction published maps and institutional affilDengue is a viral disease that can transmit to humans through the bites of an infected female Aedes genus mosquito [1]. Dengue can cause a wide range of symptoms from a mild febrile illness through to severe and life threatening manifestations such as shock, bleeding and organ dysfunction [2]. A hallmark feature of severe dengue is increased capillary permeability causing plasma leakage, which manifests as intravascular volume depletion and fluid accumulation such as pleural effusions, pulmonary edema and ascites [3]. Accurate and dynamic monitoring of fluid leak into the lungs of patients with severe dengue admitted to ICU is critical for optimal care. Lung ultrasound (LUS) imaging is a fast, portable and safe imaging technique, employed as a reference modality in intensive care units (ICU) for rapid real-time lung assessment. LUS imaging captures image artefacts such as B-lines, which are caused by fluid, that indicate a pulmonary abnormality such as edema and effusions [4]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
