Rapid Lung Ultrasound COVID-19 Severity Scoring with Resource-Efficient Deep Feature Extraction (preprint)

Artificial intelligence-based analysis of lung ultrasound imaging has been demonstrated as an effective technique for rapid diagnostic decision support throughout the COVID-19 pandemic. However, such techniques can require days- or weeks-long training processes and hyper-parameter tuning to develop intelligent deep learning image analysis models. This work focuses on leveraging 'off-the-shelf' pre-trained models as deep feature extractors for scoring disease severity with minimal training time. We propose using pre-trained initializations of existing methods ahead of simple and compact neural networks to reduce reliance on computational capacity. This reduction of computational capacity is of critical importance in time-limited or resource-constrained circumstances, such as the early stages of a pandemic. On a dataset of 49 patients, comprising over 20,000 images, we demonstrate that the use of existing methods as feature extractors results in the effective classification of COVID-19-related pneumonia severity while requiring only minutes of training time. Our methods can achieve an accuracy of over 0.93 on a 4-level severity score scale and provides comparable per-patient region and global scores compared to expert annotated ground truths. These results demonstrate the capability for rapid deployment and use of such minimally-adapted methods for progress monitoring, patient stratification and management in clinical practice for COVID-19 patients, and potentially in other respiratory diseases.

Lung Ultrasound Segmentation and Adaptation between COVID-19 and Community-Acquired Pneumonia (preprint)

Lung ultrasound imaging has been shown effective in detecting typical patterns for interstitial pneumonia, as a point-of-care tool for both patients with COVID-19 and other community-acquired pneumonia (CAP). In this work, we focus on the hyperechoic B-line segmentation task. Using deep neural networks, we automatically outline the regions that are indicative of pathology-sensitive artifacts and their associated sonographic patterns. With a real-world data-scarce scenario, we investigate approaches to utilize both COVID-19 and CAP lung ultrasound data to train the networks; comparing fine-tuning and unsupervised domain adaptation. Segmenting either type of lung condition at inference may support a range of clinical applications during evolving epidemic stages, but also demonstrates value in resource-constrained clinical scenarios. Adapting real clinical data acquired from COVID-19 patients to those from CAP patients significantly improved Dice scores from 0.60 to 0.87 (p < 0.001) and from 0.43 to 0.71 (p < 0.001), on independent COVID-19 and CAP test cases, respectively. It is of practical value that the improvement was demonstrated with only a small amount of data in both training and adaptation data sets, a common constraint for deploying machine learning models in clinical practice. Interestingly, we also report that the inverse adaptation, from labelled CAP data to unlabeled COVID-19 data, did not demonstrate an improvement when tested on either condition. Furthermore, we offer a possible explanation that correlates the segmentation performance to label consistency and data domain diversity in this point-of-care lung ultrasound application.

Lung Ultrasound to Predict COVID-19 Pneumonia: A Multicenter Descriptive Study (preprint)

Background: The pandemic surge of Coronavirus disease 2019 (COVID-19) is posing the unprecedent challenge of rapidly identifying and isolating probable cases and diagnosing the main respiratory complications. We aimed to describe the application of a lung ultrasound (LUS)-based diagnostic approach, combining the LUS likelihood of COVID-19 pneumonia with patient’s symptoms and clinical history.Methods: This is an international multicenter prospective observational study on patients suspected for COVID-19, presenting to 22 different US and European hospitals. Patients underwent LUS and reverse transcription-polymerase chain reaction (RT-PCR) swab test. We identified 3 different clinical phenotypes based on pre-existing chronic cardiac or respiratory diseases (mixed phenotype), and on the presence (severe phenotype) or absence (mild phenotype) of signs and/or symptoms of respiratory failure at presentation. We defined the LUS likelihood of COVID-19 pneumonia according to 4 different patterns, characterized by the presence and distribution of typical and atypical LUS signs: high (HPLUS), intermediate (IPLUS), alternative (APLUS) and low (LPLUS) probability patterns. The association between the combination of patterns and phenotypes with RT-PCR results was described and analyzed.Findings: We studied 1462 patients, classified in mild (n=400), severe (n=727) and mixed (n=335) phenotypes. In the overall population, the HPLUS corresponded to a positive RT-PCR in 92.6% of cases, with similarly high percentages in all clinical phenotypes ranging from 87.5% (mild) to 90.3% (mixed) and 96.5% (severe). The IPLUS yielded a lower match with positive RT-PCR (65.7%). In patients with respiratory failure, the LPLUS predicted a negative RT-PCR in 100% of cases. In the overall population, the APLUS indicated an alternative pulmonary condition in 81.1% of patients. At multivariate analysis the HPLUS strongly predicted RT-PCR positivity (odds ratio 4.173, interquartile range 2.595-6.712, p<0.0001), independently from age, low oxygen saturation and dyspnea.Interpretation: Combining LUS patterns of probability for interstitial pneumonia with clinical phenotypes at presentation could facilitate the early diagnosis of COVID-19 or suggest an alternative pulmonary condition. This approach may be useful to rapidly guide and support patient’s allocation for a wiser use of hospital resources during a pandemic surge.Funding: None.Conflict of Interest: The authors declare no conflicts of interest. Ethical Approval: The local Ethical Committee Boards of each center approved the study, and the study was conducted following the ethical standards of the 1964 Helsinki declaration and its later amendments and with local guidelines for good clinical practice.

Image quality assessment for closed-loop computer-assisted lung ultrasound (preprint)

We describe a novel, two-stage computer assistance system for lung anomaly detection using ultrasound imaging in the intensive care setting to improve operator performance and patient stratification during coronavirus pandemics. The proposed system consists of two deep-learning-based models: a quality assessment module that automates predictions of image quality, and a diagnosis assistance module that determines the likelihood-oh-anomaly in ultrasound images of sufficient quality. Our two-stage strategy uses a novelty detection algorithm to address the lack of control cases available for training the quality assessment classifier. The diagnosis assistance module can then be trained with data that are deemed of sufficient quality, guaranteed by the closed-loop feedback mechanism from the quality assessment module. Using more than 25000 ultrasound images from 37 COVID-19-positive patients scanned at two hospitals, plus 12 control cases, this study demonstrates the feasibility of using the proposed machine learning approach. We report an accuracy of 86% when classifying between sufficient and insufficient quality images by the quality assessment module. For data of sufficient quality - as determined by the quality assessment module - the mean classification accuracy, sensitivity, and specificity in detecting COVID-19-positive cases were 0.95, 0.91, and 0.97, respectively, across five holdout test data sets unseen during the training of any networks within the proposed system. Overall, the integration of the two modules yields accurate, fast, and practical acquisition guidance and diagnostic assistance for patients with suspected respiratory conditions at point-of-care.