Evaluation of Deep Learning Model Architectures for Point-of-Care Ultrasound Diagnostics
Open Access
- 18 April 2024
- journal article
- research article
- Published by MDPI AG in Bioengineering
- Vol. 11 (4) , 392
Abstract
Point-of-care ultrasound imaging is a critical tool for patient triage during trauma for diagnosing injuries and prioritizing limited medical evacuation resources. Specifically, an eFAST exam evaluates if there are free fluids in the chest or abdomen but this is only possible if ultrasound scans can be accurately interpreted, a challenge in the pre-hospital setting. In this effort, we evaluated the use of artificial intelligent eFAST image interpretation models. Widely used deep learning model architectures were evaluated as well as Bayesian models optimized for six different diagnostic models: pneumothorax (i) B- or (ii) M-mode, hemothorax (iii) B- or (iv) M-mode, (v) pelvic or bladder abdominal hemorrhage and (vi) right upper quadrant abdominal hemorrhage. Models were trained using images captured in 27 swine. Using a leave-one-subject-out training approach, the MobileNetV2 and DarkNet53 models surpassed 85% accuracy for each M-mode scan site. The different B-mode models performed worse with accuracies between 68% and 74% except for the pelvic hemorrhage model, which only reached 62% accuracy for all model architectures. These results highlight which eFAST scan sites can be easily automated with image interpretation models, while other scan sites, such as the bladder hemorrhage model, will require more robust model development or data augmentation to improve performance. With these additional improvements, the skill threshold for ultrasound-based triage can be reduced, thus expanding its utility in the pre-hospital setting.Keywords
Funding Information
- U.S. Department of War (IS220007)
- U.S. Department of Energy Oak Ridge Institute for Science and Education (IS220007)
This publication has 35 references indexed in Scilit:
- Putting Medical Boots on the Ground: Lessons from the War in Ukraine and Applications for Future Conflict with Near-Peer AdversariesJournal of the American College of Surgeons, 2023
- Detection of pneumothorax on ultrasound using artificial intelligenceJournal of Trauma and Acute Care Surgery, 2022
- An image classification deep-learning algorithm for shrapnel detection from ultrasound imagesScientific Reports, 2022
- Use of eFAST in Patients with Injury to the Thorax or AbdomenNew England Journal of Medicine, 2022
- AI in health and medicineNature Medicine, 2022
- A review of medical image data augmentation techniques for deep learning applicationsJournal of Medical Imaging and Radiation Oncology, 2021
- Deep Neural Networks for Human Activity Recognition With Wearable Sensors: Leave-One-Subject-Out Cross-Validation for Model SelectionIEEE Access, 2020
- SonoNet: Real-Time Detection and Localisation of Fetal Standard Scan Planes in Freehand UltrasoundIEEE Transactions on Medical Imaging, 2017
- M-Mode Ultrasound Applications for the Emergency Medicine PhysicianThe Journal of Emergency Medicine, 2015
- Triage in Medicine, Part I: Concept, History, and TypesAnnals of Emergency Medicine, 2007