Benchmark methodological approach for the application of artificial intelligence to lung ultrasound data from COVID-19 patients: From frame to prognostic-level.

Automated ultrasound imaging assessment of the effect of CoronaVirus disease 2019 (COVID-19) on lungs has been investigated in various studies using artificial intelligence-based (AI) methods. However, an extensive analysis of state-of-the-art Convolutional Neural Network-based (CNN) models for frame-level scoring, a comparative analysis of aggregation techniques for video-level scoring, together with a thorough evaluation of the capability of these methodologies to provide a clinically valuable prognostic-level score is yet missing within the literature. In addition to that, the impact on the analysis of the posterior probability assigned by the network to the predicted frames as well as the impact of temporal downsampling of LUS data are topics not yet extensively investigated. This paper takes on these challenges by providing a benchmark analysis of methods from frame to prognostic level. For frame-level scoring, state-of-the-art deep learning models are evaluated with additional analysis of best performing model in transfer-learning settings. A novel cross-correlation based aggregation technique is proposed for video and exam-level scoring. Results showed that ResNet-18, when trained from scratch, outperformed the existing methods with an F1-Score of 0.659. The proposed aggregation method resulted in 59.51%, 63.29%, and 84.90% agreement with clinicians at the video, exam, and prognostic levels, respectively; thus, demonstrating improved performances over the state of the art. It was also found that filtering frames based on the posterior probability shows higher impact on the LUS analysis in comparison to temporal downsampling. All of these analysis were conducted over the largest standardized and clinically validated LUS dataset from COVID-19 patients.

Lung Ultrasound in COVID-19 and Post-COVID-19 Patients, an Evidence-Based Approach.

OBJECTIVES: Worldwide, lung ultrasound (LUS) was utilized to assess coronavirus disease 2019 (COVID-19) patients. Often, imaging protocols were however defined arbitrarily and not following an evidence-based approach. Moreover, extensive studies on LUS in post-COVID-19 patients are currently lacking. This study analyses the impact of different LUS imaging protocols on the evaluation of COVID-19 and post-COVID-19 LUS data. METHODS: LUS data from 220 patients were collected, 100 COVID-19 positive and 120 post-COVID-19. A validated and standardized imaging protocol based on 14 scanning areas and a 4-level scoring system was implemented. We utilized this dataset to compare the capability of 5 imaging protocols, respectively based on 4, 8, 10, 12, and 14 scanning areas, to intercept the most important LUS findings. This to evaluate the optimal trade-off between a time-efficient imaging protocol and an accurate LUS examination. We also performed a longitudinal study, aimed at investigating how to eventually simplify the protocol during follow-up. Additionally, we present results on the agreement between AI models and LUS experts with respect to LUS data evaluation. RESULTS: A 12-areas protocol emerges as the optimal trade-off, for both COVID-19 and post-COVID-19 patients. For what concerns follow-up studies, it appears not to be possible to reduce the number of scanning areas. Finally, COVID-19 and post-COVID-19 LUS data seem to show differences capable to confuse AI models that were not trained on post-COVID-19 data, supporting the hypothesis of the existence of LUS patterns specific to post-COVID-19 patients. CONCLUSIONS: A 12-areas acquisition protocol is recommended for both COVID-19 and post-COVID-19 patients, also during follow-up.

Multi-objective automatic analysis of lung ultrasound data from COVID-19 patients by means of deep learning and decision trees.

COVID-19 raised the need for automatic medical diagnosis, to increase the physicians' efficiency in managing the pandemic. Among all the techniques for evaluating the status of the lungs of a patient with COVID-19, lung ultrasound (LUS) offers several advantages: portability, cost-effectiveness, safety. Several works approached the automatic detection of LUS imaging patterns related COVID-19 by using deep neural networks (DNNs). However, the decision processes based on DNNs are not fully explainable, which generally results in a lack of trust from physicians. This, in turn, slows down the adoption of such systems. In this work, we use two previously built DNNs as feature extractors at the frame level, and automatically synthesize, by means of an evolutionary algorithm, a decision tree (DT) that aggregates in an interpretable way the predictions made by the DNNs, returning the severity of the patients' conditions according to a LUS score of prognostic value. Our results show that our approach performs comparably or better than previously reported aggregation techniques based on an empiric combination of frame-level predictions made by DNNs. Furthermore, when we analyze the evolved DTs, we discover properties about the DNNs used as feature extractors. We make our data publicly available for further development and reproducibility.

Combining L-Arginine with vitamin C improves long-COVID symptoms: The LINCOLN Survey.

INTRODUCTION: Recent evidence suggests that oxidative stress and endothelial dysfunction play critical roles in the pathophysiology of COVID-19 and Long-COVID. We hypothesized that a supplementation combining L-Arginine (to improve endothelial function) and Vitamin C (to reduce oxidation) could have favorable effects on Long-COVID symptoms. METHODS: We designed a survey (LINCOLN: L-Arginine and Vitamin C improves Long-COVID), assessing several symptoms that have been associated with Long-COVID to be administered nationwide to COVID-19 survivors; the survey also included effort perception, measured using the Borg scale. Patients receiving the survey were divided in two groups, with a 2:1 ratio: the first group included patients that received L-Arginine + Vitamin C, whereas the second group received a multivitamin combination (alternative treatment). RESULTS: 1390 patients successfully completed the survey. Following a 30-day treatment in both groups, the survey revealed that patients in the L-Arginine + Vitamin C treatment arm had significantly lower scores compared to patients who had received the multivitamin combination. There were no other significant differences between the two groups. When examining effort perception, we observed a significantly lower value (p < 0.0001) in patients receiving L-Arginine + Vitamin C compared to the alternative-treatment arm. CONCLUSIONS: Our survey indicates that the supplementation with L-Arginine + Vitamin C has beneficial effects in Long-COVID, in terms of attenuating its typical symptoms and improving effort perception.

Predictors of Helmet CPAP Failure in COVID-19 Pneumonia: A Prospective, Multicenter, and Observational Cohort Study.

Background: Continuous positive airway pressure (CPAP) can be beneficial in acute respiratory failure (ARF) due to coronavirus (COVID-19) pneumonia, but delaying endotracheal intubation (ETI) in nonresponders may increase mortality. We aimed at investigating the performance of composite respiratory indexes as possible predictors of CPAP failure in ARF due to COVID-19. Methods: This was a multicenter, prospective, observational, and cohort study conducted in the respiratory units of three University hospitals in Milan and in a secondary care hospital in Codogno (Italy), on consecutive adult patients with ARF due to COVID-19 pneumonia that underwent CPAP between March 2020 and March 2021. ETI transfer to the intensive care unit or death is defined CPAP failure. Predictors of CPAP failure were assessed before T0 and 1 hour after T1 CPAP initiation and included mROX index (ratio of PaO2/FiO2 to respiratory rate), alveolar-to-arterial (A-a) O2 gradient, and the HACOR (heart rate, acidosis, consciousness, oxygenation, and respiratory rate) score. Results: Three hundred and fifty four patients (mean age 64 years, 73% males) were included in the study; 136 (38.4%) satisfied criteria for CPAP failure. A-a O2 gradient, mROX, and HACOR scores were worse in patients who failed CPAP, both at T0 and T1 (p < 0.001 for all parameters). The HACOR score was associated with CPAP failure (odds ratio-OR-for every unit increase in HACOR = 1.361; 95%CI: 1.103-1.680; p=0.004; AUROC = 0.742; p < 0.001). CPAP failure was best predicted by a threshold of 4.50 (sensitivity = 53% and specificity = 87%). Conclusions: The HACOR score may be a reliable and early predictor of CPAP failure in patients treated for ARF in COVID-19 pneumonia.

Pleuropulmonary Ultrasound in Pediatrics: Proposal of a Reporting Model From the Academy of Thoracic Ultrasound.

With the emergence of the Covid-19 pandemic, pleuropulmonary ultrasound has become a very common tool in clinical practice, even in the pediatric field. Therefore, the clinicians' need to speak a common ultrasound language becomes increasingly necessary. The Italian scientific society AdET (Academy of Thoracic Ultrasound) has been carrying out the study and dissemination of pulmonary ultrasound in medical practice in Italy for years. With this article, the pediatric AdET group wants to propose a report model of pediatric pulmonary ultrasound as a useful tool in daily clinical practice to interpret the images and reach a diagnostic conclusion, aiming to share a standardized approach that may also support the sharing of research findings.

On the Impact of Different Lung Ultrasound Imaging Protocols in the Evaluation of Patients Affected by Coronavirus Disease 2019: How Many Acquisitions Are Needed?

Lung ultrasound (LUS) is currently being extensively used for the evaluation of patients affected by coronavirus disease 2019. In the past months, several imaging protocols have been proposed in the literature. However, how the different protocols would compare when applied to the same patients had not been investigated yet. To this end, in this multicenter study, we analyzed the outcomes of 4 different LUS imaging protocols, respectively based on 4, 8, 12, and 14 LUS acquisitions, on data from 88 patients. Results show how a 12-area acquisition system seems to be a good tradeoff between the acquisition time and accuracy.

A New Lung Ultrasound Protocol Able to Predict Worsening in Patients Affected by Severe Acute Respiratory Syndrome Coronavirus 2 Pneumonia.

OBJECTIVES: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection can generate severe pneumonia associated with high mortality. A bedside lung ultrasound (LUS) examination has been shown to have a potential role in this setting. The purpose of this study was to evaluate the potential prognostic value of a new LUS protocol (evaluation of 14 anatomic landmarks, with graded scores of 0-3) in patients with SARS-CoV-2 pneumonia and the association of LUS patterns with clinical or laboratory findings. METHODS: A cohort of 52 consecutive patients with laboratory-confirmed SARS-CoV-2 underwent LUS examinations on admission in an internal medicine ward and before their discharge. A total LUS score as the sum of the scores at each explored area was computed. We investigated the association between the LUS score and clinical worsening, defined as a combination of high-flow oxygen support, intensive care unit admission, or 30-day mortality as the primary end point. RESULTS: Twenty (39%) patients showed a worse outcome during the observation period; the mean LUS scores ± SDs were 20.4 ± 8.5 and 29.2 ± 7.3 in patients without and with worsening, respectively (P < .001). In a multivariable analysis, adjusted for comorbidities (>2), age (>65 years), sex (male), and body mass index (≥25 kg/m2 ), the association between the LUS score and worsening (odds ratio, 1.17; 95% confidence interval, 1.05 to 1.29; P = .003) was confirmed, with good discrimination of the model (area under the receiver operating characteristic curve, 0.82). A median LUS score higher than 24 was associated with an almost 6-fold increase in the odds of worsening (odds ratio, 5.67; 95% confidence interval, 1.29 to 24.8; P = .021). CONCLUSIONS: Lung ultrasound can represent an effective tool for monitoring and stratifying the prognosis of patients with SARS-CoV-2 pulmonary involvement.

Changes in quality of life and dyspnoea after hospitalization in COVID-19 patients discharged at home.

BACKGROUND: To date, the effects of COVID-19 pneumonia on health-related quality of life (HRQoL) and dyspnoea are unknown. METHODS: In a real-life observational study, 20 patients with COVID-19-related pneumonia received usual care plus erdosteine (300 mg twice daily) for 15 days after hospital discharge following local standard operating procedures. At discharge (T0) and on Day 15 (T1), participants completed the St George's Respiratory Questionnaire (SGRQ), the modified Medical Research Council (mMRC) scale of dyspnoea during daily activity, the BORG scale for dyspnoea during exertion, and Visual Analogue Scale (VAS) for dyspnoea at rest. Paired t-tests compared scores at T0 and T1. RESULTS: The mean (SD) SGRQ total score decreased from 25.5 (15.5) at T0 to 16.9 (13.2) at T1 (p<0.01); 65% of patients achieved a clinically important change of ≥4 points. SGRQ domain scores (symptoms, activity, and impact) were also significantly reduced (all p<0.01). The mean (SD) VAS score decreased from 1.6 (1.7) to 1.4 (2.5); p<0.01. The mean mMRC score decreased significantly (p=0.031) and 30% of patients achieved a clinically important change of ≥1 point. The mean (SD) Borg score increased from 12.8 (4.2) to 14.3 (2.4); p<0.01. CONCLUSION: The present proof of concept study is the first to report HRQoL in patients with COVID-19. During 15 days after hospital discharge, patients reported significant improvements in HRQoL and dyspnoea at rest and during daily activities.

Severity of respiratory failure at admission and in-hospital mortality in patients with COVID-19: a prospective observational multicentre study.

OBJECTIVES: COVID-19 causes lung parenchymal and endothelial damage that lead to hypoxic acute respiratory failure (hARF). The influence of hARF severity on patients' outcomes is still poorly understood. DESIGN: Observational, prospective, multicentre study. SETTING: Three academic hospitals in Milan (Italy) involving three respiratory high dependency units and three general wards. PARTICIPANTS: Consecutive adult hospitalised patients with a virologically confirmed diagnosis of COVID-19. Patients aged <18 years or unable to provide informed consent were excluded. INTERVENTIONS: Anthropometrical, clinical characteristics and blood biomarkers were assessed within the first 24 hours from admission. hARF was graded as follows: severe (partial pressure of oxygen to fraction of inspired oxygen ratio (PaO2/FiO2) <100 mm Hg); moderate (PaO2/FiO2 101-200 mm Hg); mild (PaO2/FiO2 201-300 mm Hg) and normal (PaO2/FiO2 >300 mm Hg). PRIMARY AND SECONDARY OUTCOME MEASURES: The primary outcome was the assessment of clinical characteristics and in-hospital mortality based on the severity of respiratory failure. Secondary outcomes were intubation rate and application of continuous positive airway pressure during hospital stay. RESULTS: 412 patients were enrolled (280 males, 68%). Median (IQR) age was 66 (55-76) years with a PaO2/FiO2 at admission of 262 (140-343) mm Hg. 50.2% had a cardiovascular disease. Prevalence of mild, moderate and severe hARF was 24.4%, 21.9% and 15.5%, respectively. In-hospital mortality proportionally increased with increasing impairment of gas exchange (p<0.001). The only independent risk factors for mortality were age ≥65 years (HR 3.41; 95% CI 2.00 to 5.78, p<0.0001), PaO2/FiO2 ratio ≤200 mm Hg (HR 3.57; 95% CI 2.20 to 5.77, p<0.0001) and respiratory failure at admission (HR 3.58; 95% CI 1.05 to 12.18, p=0.04). CONCLUSIONS: A moderate-to-severe impairment in PaO2/FiO2 was independently associated with a threefold increase in risk of in-hospital mortality. Severity of respiratory failure is useful to identify patients at higher risk of mortality. TRIAL REGISTRATION NUMBER: NCT04307459.

Proposal for International Standardization of the Use of Lung Ultrasound for Patients With COVID-19: A Simple, Quantitative, Reproducible Method.

Growing evidence is showing the usefulness of lung ultrasound in patients with the 2019 new coronavirus disease (COVID-19). Severe acute respiratory syndrome coronavirus 2 has now spread in almost every country in the world. In this study, we share our experience and propose a standardized approach to optimize the use of lung ultrasound in patients with COVID-19. We focus on equipment, procedure, classification, and data sharing.