Lung ultrasound (LUS) is an accurate thoracic imaging technique distinguished by its handheld size, low-cost, and lack of radiation. User dependence and poor access to training have limited the impact and dissemination of LUS outside of acute care hospital environments. Automated interpretation of LUS using deep learning can overcome these barriers by increasing accuracy while allowing point-of-care use by non-experts. In this multicenter study, we seek to automate the clinically vital distinction between A line (normal parenchyma) and B line (abnormal parenchyma) on LUS by training a customized neural network using 272,891 labelled LUS images. After external validation on 23,393 frames, pragmatic clinical application at the clip level was performed on 1162 videos. The trained classifier demonstrated an area under the receiver operating curve (AUC) of 0.96 (±0.02) through 10-fold cross-validation on local frames and an AUC of 0.93 on the external validation dataset. Clip-level inference yielded sensitivities and specificities of 90% and 92% (local) and 83% and 82% (external), respectively, for detecting the B line pattern. This study demonstrates accurate deep-learning-enabled LUS interpretation between normal and abnormal lung parenchyma on ultrasound frames while rendering diagnostically important sensitivity and specificity at the video clip level.
Background: Annotating large medical imaging datasets is an arduous and expensive task, especially when the datasets in question are not organized according to deep learning goals. Here, we propose a method that exploits the hierarchical organization of annotating tasks to optimize efficiency. Methods: We trained a machine learning model to accurately distinguish between one of two classes of lung ultrasound (LUS) views using 2908 clips from a larger dataset. Partitioning the remaining dataset by view would reduce downstream labelling efforts by enabling annotators to focus on annotating pathological features specific to each view. Results: In a sample view-specific annotation task, we found that automatically partitioning a 780-clip dataset by view saved 42 min of manual annotation time and resulted in 55±6 additional relevant labels per hour. Conclusions: Automatic partitioning of a LUS dataset by view significantly increases annotator efficiency, resulting in higher throughput relevant to the annotating task at hand. The strategy described in this work can be applied to other hierarchical annotation schemes.
To evaluate the accuracy of a bedside, real-time deployment of a deep learning (DL) model capable of distinguishing between normal (A line pattern) and abnormal (B line pattern) lung parenchyma on lung ultrasound (LUS) in critically ill patients. DESIGN:Prospective, observational study evaluating the performance of a previously trained LUS DL model. Enrolled patients received a LUS examination with simultaneous DL model predictions using a portable device. Clip-level model predictions were analyzed and compared with blinded expert review for A versus B line pattern. Four prediction thresholding approaches were applied to maximize model sensitivity and specificity at bedside. SETTING: Academic ICU. PATIENTS:One-hundred critically ill patients admitted to ICU, receiving oxygen therapy, and eligible for respiratory imaging were included. Patients who were unstable or could not undergo an LUS examination were excluded. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS:A total of 100 unique ICU patients (400 clips) were enrolled from two tertiary-care sites. Fifty-six patients were mechanically ventilated. When compared with gold standard expert annotation, the real-time inference yielded an accuracy of 95%, sensitivity of 93%, and specificity of 96% for identification of the B line pattern. Varying prediction thresholds showed that real-time modification of sensitivity and specificity according to clinical priorities is possible. CONCLUSIONS:A previously validated DL classification model performs equally well in real-time at the bedside when platformed on a portable device. As the first study to test the feasibility and performance of a DL classification model for LUS in a dedicated ICU environment, our results justify further inquiry into the impact of employing real-time automation of medical imaging into the care of the critically ill.
INTRODUCTION:Since the early pandemic, prone positioning (PP) has been broadly utilized for non-intubated COVID-19 patients, but results from recently published randomized controlled trials (RCTs) are contradictory. We aimed to systematically synthesize the outcomes associated with PP for non-intubated COVID-19 patients. METHODS:Two independent groups of researchers searched MEDLINE, Embase, PubMed, Web of Science, Scopus, and ClinicalTrials.gov for RCTs of PP in nonintubated adult patients with COVID-19 and published in English from January 1st, 2020 to July 1st, 2022. The same two independent groups extracted the data and assessed the risk of bias. We used a random-effects meta-analysis to pool individual studies and the GRADE approach to assess certainty/quality of the evidence. The primary outcome was the reported cumulative intubation risk, while secondary outcomes included mortality, need for escalating respiratory support, hospital length of stay, ICU admission, and adverse events. The study protocol was prospectively registered with PROSPERO, CRD42022343625. RESULTS:12 RCTs with 2,886 patients were included. For non-intubated COVID-19 patients, PP significantly reduced the intubation risk (risk ratio [RR] 0.85, 95%CI 0.75 to 0.96), compared to supine position. Subgroup analysis showed a significant reduction in intubation risk among patients supported by high-flow nasal cannula (HFNC) or noninvasive ventilation (NIV) (RR 0.83, 95%CI 0.73 to 0.94) but not in patients with conventional oxygen therapy (RR 1.02, 95%CI 0.67 to 1.56). No significant reduction was seen in mortality (RR 0.96, 95%CI 0.82 to 1.13), need for escalating respiratory support (RR 1.03, 95%CI 0.77 to 1.37), hospital length of stay (MD 0.35 days, 95%CI -0.57 to 1.26), ICU admission (RR 0.75, 95%CI 0.51 to 1.10), and adverse events. No obvious risk of bias and publication bias was found for the primary outcome. CONCLUSIONS:In non-intubated COVID-19 patients, PP reduced the need for intubation, in particular among those requiring respiratory support with HFNC or NIV, but did not reduce mortality, need for escalating respiratory support, hospital length of stay, and ICU admission.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.