Research | Deep Breathe

Featured

Publications.

Self-supervised pretraining improves performance and inference efficiency in multiple lung ultrasound interpretation tasks

Derek Wu, Delaney Smith, Blake VanBerlo, Amir Roshankar, Hoseok Lee, Brian Li, Faraz Ali, Marwan Rahman, John Basmaji, Jared Tschirhart, Alex Ford, Bennett VanBerlo, Ashritha Durvasula, Claire Vannelli, Chintan Dave, Jason Deglint, Jordan Ho, Rushil Chaudhary, Hans Clausdorff, Ross Prager, Scott Millington, Samveg Shah, Brian Buchanan, Robert Arntfield

In this study, we investigated whether self-supervised pretraining could produce a neural network feature extractor applicable to multiple classification tasks in B-mode lung ultrasound analysis.

10.1109/access.2023.3337398

Research

Latest science.

Diagnostic accuracy of an automated classifier for the detection of pleural effusions in patients undergoing lung ultrasound

Rushil Chaudhary, Jordan Ho, Delaney Smith, Saad Hossain, Jaswin Hargun, Blake VanBerlo, Niall Murphy, Ross Prager, Kiran Rikhraj, Jared Tschirhart, Robert Arntfield

This study aims to develop and validate a deep learning model for detecting pleural effusions in lung ultrasound images, with adaptable performance characteristics tailored to specific clinical scenarios.

The American Journal of Emergency Medicine

https://doi.org/10.1016/j.ajem.2025.01.041

Read publication

Automated real-time detection of lung sliding using artificial intelligence: a prospective diagnostic accuracy study

Hans Clausdorff Fiedler, Ross Prager, Delaney Smith, Derek Wu, Chintan Dave, Jared Tschirhart, Benjamin Wu, Blake Van Berlo, Richard Malthaner, Robert Arntfield

In patients with suspected pneumothorax, what is the real-time diagnostic accuracy of AI-LUS to recognize the absence of lung sliding?

10.1016/j.chest.2024.02.011

Read publication

Intra-video positive pairs in self-supervised learning for ultrasound

Blake VanBerlo, Alexander Wong, Jesse Hosey, Robert Arntfield

In this study, we investigated the effect of utilizing proximal, distinct images from the same B-mode ultrasound video as pairs for SSL.

10.48550/ARXIV.2403.07715

Read publication

Improving the generalizability and performance of an ultrasound deep learning model using limited multicenter data for lung sliding artifact identification

Derek Wu, Delaney Smith, Blake VanBerlo, Amir Roshankar, Hoseok Lee, Brian Li, Faraz Ali, Marwan Rahman, John Basmaji, Jared Tschirhart, Alex Ford, Bennett VanBerlo, Ashritha Durvasula, Claire Vannelli, Chintan Dave, Jason Deglint, Jordan Ho, Rushil Chaudhary, Hans Clausdorff, Ross Prager, Scott Millington, Samveg Shah, Brian Buchanan, Robert Arntfield

We report a multicenter study that exploits limited available external data to improve the generalizability and performance of our lung sliding model while identifying poorly performing subgroups to inform future iterative improvements.

10.3390/diagnostics14111081

Read publication

Exploring the utility of self-supervised pretraining strategies for the detection of absent lung sliding in M-mode lung ultrasound

Blake VanBerlo, Brian Li, Alexander Wong, Jesse Hoey, Robert Arntfield

This study explores the efficacy of self-supervised pretraining in improving performance for lung sliding classification in M-mode lung ultrasound images, highlighting its superiority over full supervision.

IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops

10.1109/cvprw59228.2023.00309

Read publication

Prospective real-time validation of a lung ultrasound deep learning model in the ICU

Chintan Dave, Derek Wu, Jared Tschirhart, Delaney Smith, Blake VanBerlo, Jason Deglint, Faraz Ali, Rushil Chaudhary, Bennett VanBerlo, Alex Ford, Marwan A. Rahman, Joseph McCauley, Benjamin Wu, Jordan Ho, Brian Li, Robert Arntfield

This prospective study assessed the accuracy of a deep learning model deployed in real-time at the bedside to differentiate between normal and abnormal lung parenchyma on lung ultrasound in critically ill patients.

Critical Care Medicine

10.1097/01.ccm.0000906084.96293.1b

Read publication

Introducing medical students to deep learning through image labelling: a new approach to meet calls for greater artificial intelligence fluency among medical trainees

Jared Tschirhart, Amadene Woolsey, Jamila Skinner, Khadija Ahmed, Courtney Fleming, Justin Kim, Chintan Dave, Robert Arntfield

Our approach bridges the gap between medical education and AI by empowering medical students to excel in medical imaging and labeling, fostering competence while generating valuable data for AI integration in clinical practice.

Canadian Medical Education Journal

10.36834/cmej.75074

Read publication

Enhancing annotation efficiency with machine learning: automated partitioning of a lung ultrasound dataset by view

Bennett VanBerlo, Delaney Smith, Jared Tschirhart, Blake VanBerlo, Derek Wu, Alex Ford, Joseph McCauley, Benjamin Wu, Rushil Chaudhary, Chintan Dave, Jordan Ho, Jason Deglint, Brian Li, Robert Arntfield

We propose a method leveraging hierarchical organization of annotating tasks to increase efficiency in annotating lung ultrasound images, demonstrating significant time savings and higher throughput in view-specific annotation tasks.

10.3390/diagnostics12102351

Read publication

Accurate assessment of the lung sliding artefact on lung ultrasonography using a deep learning approach

Blake VanBerlo, Derek Wu, Brian Li, Marwan A. Rahman, Gregory Hogg, Bennett VanBerlo, Jared Tschirhart, Alex Ford, Jordan Ho, Joseph McCauley, Benjamin Wu, Jason Deglint, Jaswin Hargun, Rushil Chaudhary, Chintan Dave, Robert Arntfield

Using deep learning methods, our study developed a binary classifier capable of accurately distinguishing between the presence and absence of the lung sliding artifact on lung ultrasound images.

Computers in Biology and Medicine

10.1016/j.compbiomed.2022.105953

Read publication

Automation of lung ultrasound interpretation via deep learning for the classification of normal versus abnormal lung parenchyma: A multicenter study

Robert Arntfield, Derek Wu, Jared Tschirhart, Blake VanBerlo, Alex Ford, Jordan Ho, Joseph McCauley, Benjamin Wu, Jason Deglint, Rushil Chaudhary, Chintan Dave, Bennett VanBerlo, John Basmaji, Scott Millington

This study introduces automated LUS interpretation using deep learning to address these challenges. By training a neural network on a large dataset of labeled LUS images, the study aimed to distinguish between normal and abnormal lung patterns.

10.3390/diagnostics11112049

Read publication

Development of a convolutional neural network to differentiate among the etiology of similar appearing pathological B lines on lung ultrasound: a deep learning study

Robert Arntfield, Blake VanBerlo, Thamer Alaifan, Nathan Phelps, Matthew White, Rushil Chaudhary, Jordan Ho, Derek Wu

This study investigated whether deep learning techniques could enhance diagnostic specificity in lung ultrasound imaging. A convolutional neural network was trained on ultrasound images depicting B lines associated with various respiratory pathologies.

10.1136/bmjopen-2020-045120

Read publication