Automatic detection of A-line in lung ultrasound images using deep learning and image processing.

BACKGROUND: Auxiliary diagnosis and monitoring of lung diseases based on lung ultrasound (LUS) images is important clinical research. A-line is one of the most common indicators of LUS that can offer support for the assessment of lung diseases. A traditional A-line detection method mainly relies on experienced clinicians, which is inefficient and cannot meet the needs of these areas with backward medical level. Therefore, how to realize the automatic detection of A-line in LUS image is important. PURPOSE: In order to solve the disadvantages of traditional A-line detection methods, realize automatic and accurate detection, and provide theoretical support for clinical application, we proposed a novel A-line detection method for LUS images with different probe types in this paper. METHODS: First, the improved Faster R-CNN model with a selection strategy of localization box was designed to accurately locate the pleural line. Then, the LUS image below the pleural line was segmented for independent analysis excluding the influence of other similar structures. Next, image-processing methods based on total variation, matched filter, and gray difference were applied to achieve the automatic A-line detection. Finally, the "depth" index was designed to verify the accuracy by judging whether the automatic measurement results belong to corresponding manual results (±5%). In experiments, 3000 convex array LUS images were used for training and validating the improved pleural line localization model by five-fold cross validation. 850 convex array LUS images and 1080 linear array LUS images were used for testing the trained pleural line localization model and the proposed image-processing-based A-line detection method. The accuracy analysis, error statistics, and Harsdorff distance were employed to evaluate the experimental results. RESULTS: After 100 epochs, the mean loss value of training and validation set of improved Faster R-CNN model reached 0.6540 and 0.7882, with the validation accuracy of 98.70%. The trained pleural line localization model was applied in the testing set of convex and linear probes and reached the accuracy of 97.88% and 97.11%, respectively, which were 3.83% and 8.70% higher than the original Faster R-CNN model. The accuracy, sensitivity, and specificity of A-line detection reached 95.41%, 0.9244%, 0.9875%, and 94.63%, 0.9230%, and 0.9766% for convex and linear probes, respectively. Compared to the experienced clinicians' results, the mean value and p value of depth error were 1.5342 ± 1.2097 and 0.9021, respectively, and the Harsdorff distance was 5.7305 ± 1.8311. In addition, the accumulated accuracy of the two-stage experiment (pleural line localization and A-line detection) was calculated as the final accuracy of the whole A-line detection system. They were 93.39% and 91.90% for convex and linear probes, respectively, which were higher than these previous methods. CONCLUSIONS: The proposed method combining image processing and deep learning can automatically and accurately detect A-line in LUS images with different probe types, which has important application value for clinical diagnosis.

Maternal-Neonatal Outcomes of Obstetric Deliveries Performed in Negative Pressure Isolation Rooms during the COVID-19 Omicron Variant Pandemic in Taiwan: A Retrospective Cohort Study of a Single Institution.

Objective: To investigate the maternal-neonatal outcomes of obstetric deliveries performed in negative pressure isolated delivery rooms (NPIDRs) during the coronavirus disease 2019 (COVID-19) omicron variant pandemic period in a single tertiary center in northern Taiwan. Methods: Confirmed positive and suspected-positive COVID-19 cases delivered in NPIDRs and COVID-19-negative mothers delivered in conventional delivery rooms (CDRs) in the period of 1 May 2022 to 31 May 2022 during the COVID-19 omicron variant pandemic stage were reviewed. The maternal-neonatal outcomes between the two groups of mothers were analyzed. All deliveries were performed following the obstetric and neonatologic protocols conforming to the epidemic prevention regulations promulgated by the Taiwan Centers for Disease Control (T-CDC). Multiple gestations, deliveries at gestational age below 34 weeks, and major fetal anomalies were excluded from this study. Results: A total of 213 obstetric deliveries were included. Forty-five deliveries were performed in NPIDRs due to a positive COVID-19 polymerase chain reaction (PCR) test (n = 41) or suspected COVID-19 positive status (n = 4). One hundred and sixty-eight deliveries with negative COVID-19 PCR tests were performed in CDRs. There was no statistical difference in maternal characteristics between the two groups of pregnant women. All COVID-19-confirmed cases either presented with mild upper-airway symptoms (78%) or were asymptomatic (22%); none of these cases developed severe acute respiratory syndrome. The total rate of cesarean section was not statistically different between obstetric deliveries in NPIDRs and in CDRs (38.1% vs. 40.0%, p = 0.82, respectively). Regardless of delivery modes, poorer short-term perinatal outcomes were observed in obstetric deliveries in NPIDRs: there were significant higher rates of neonatal respiratory distress (37.8% vs. 10.7%, p < 0.001, respectively), meconium-stained amniotic fluid (22.2% vs. 4.2%, p < 0.001, respectively) and newborn intensive care unit admission (55.6% vs. 8.3%, p < 0.001, respectively) in obstetric deliveries performed in NPIDRs than in CDRs. Maternal surgical outcomes were not significantly different between the two groups of patients. There was no vertical transmission or nosocomial infection observed in COVID-19 confirmed cases in this study period. Conclusions: Our study demonstrates that obstetric deliveries for positive and suspected COVID-19 omicron-variant cases performed in NPIDRs are associated with poorer short-term perinatal outcomes. Reasonable use of personal protective equipment in NPIDRs could effectively prevent nosocomial infection during obstetric deliveries for pregnant women infected with the COVID-19 omicron variant.

Performance evaluation of antigen detection rapid diagnostic test (Ag-RDT) for COVID-19 diagnosis in a primary healthcare center during the Shanghai COVID-19 quarantine period.

BACKGROUND: Rapid and accurate detection of SARS-CoV-2 infection is the cornerstone of prompt patient care. However, the reliability of the antigen rapid diagnostic test (Ag-RDT) in the diagnosis of SARS-CoV-2 infection remains inconclusive. METHODS: We conducted a field evaluation of Ag-RDT performance during the Shanghai Coronavirus disease 2019 (COVID-19) quarantine and screened 7225 individuals visiting our Emergency Department. 83 asymptomatic SARS-CoV-2 (+) individuals were enrolled in the current study. Simultaneously, Ag-RDT was performed to evaluate its testing performance. RESULTS: For the Ag-RDT(-) cases, the average cycle threshold (Ct) values of the N gene were 27.26 ± 4.59, which were significantly higher than the Ct value (21.9 ± 4.73) of the Ag-RDT(+) individuals (p < 0.0001). The overall sensitivity of Ag-RDT versus that of RT-PCR was 43.37%. The Ag-RDT(+) individuals regarding the N gene's Ct value were 16 cases in the < 20 range, 12 in 20-25, 5 in 25-30, and 3 in 30-35. The corresponding sensitivity was 84.21%, 52.17%, 21.74% and 16.67%, respectively. Meanwhile, sampling had a straight specificity of 100% regardless of the Ct value. CONCLUSIONS: The Ag-RDT were extremely sensitive in asymptomatic COVID-19 individuals with a Ct value < 20.

Evaluation of Pulmonary Edema Using Ultrasound Imaging in Patients With COVID-19 Pneumonia Based on a Non-local Channel Attention ResNet.

Recent research has revealed that COVID-19 pneumonia is often accompanied by pulmonary edema. Pulmonary edema is a manifestation of acute lung injury (ALI), and may progress to hypoxemia and potentially acute respiratory distress syndrome (ARDS), which have higher mortality. Precise classification of the degree of pulmonary edema in patients is of great significance in choosing a treatment plan and improving the chance of survival. Here we propose a deep learning neural network named Non-local Channel Attention ResNet to analyze the lung ultrasound images and automatically score the degree of pulmonary edema of patients with COVID-19 pneumonia. The proposed method was designed by combining the ResNet with the non-local module and the channel attention mechanism. The non-local module was used to extract the information on characteristics of A-lines and B-lines, on the basis of which the degree of pulmonary edema could be defined. The channel attention mechanism was used to assign weights to decisive channels. The data set contains 2220 lung ultrasound images provided by Huoshenshan Hospital, Wuhan, China, of which 2062 effective images with accurate scores assigned by two experienced clinicians were used in the experiment. The experimental results indicated that our method achieved high accuracy in classifying the degree of pulmonary edema in patients with COVID-19 pneumonia by comparison with previous deep learning methods, indicating its potential to monitor patients with COVID-19 pneumonia.

Automated lung ultrasound scoring for evaluation of coronavirus disease 2019 pneumonia using two-stage cascaded deep learning model.

Coronavirus disease 2019 (COVID-19) pneumonia has erupted worldwide, causing massive population deaths and huge economic losses. In clinic, lung ultrasound (LUS) plays an important role in the auxiliary diagnosis of COVID-19 pneumonia. However, the lack of medical resources leads to the low using efficiency of the LUS, to address this problem, a novel automated LUS scoring system for evaluating COVID-19 pneumonia based on the two-stage cascaded deep learning model was proposed in this paper. 18,330 LUS images collected from 26 COVID-19 pneumonia patients were successfully assigned scores by two experienced doctors according to the designed four-level scoring standard for training the model. At the first stage, we made a secondary selection of these scored images through five ResNet-50 models and five-fold cross validation to obtain the available 12,949 LUS images which were highly relevant to the initial scoring results. At the second stage, three deep learning models including ResNet-50, Vgg-19, and GoogLeNet were formed the cascaded scored model and trained using the new dataset, whose predictive result was obtained by the voting mechanism. In addition, 1000 LUS images collected another 5 COVID-19 pneumonia patients were employed to test the model. Experiments results showed that the automated LUS scoring model was evaluated in terms of accuracy, sensitivity, specificity, and F1-score, being 96.1%, 96.3%, 98.8%, and 96.1%, respectively. They proved the proposed two-stage cascaded deep learning model could automatically score an LUS image, which has great potential for application to the clinics on various occasions.

Quantitative Analysis and Automated Lung Ultrasound Scoring for Evaluating COVID-19 Pneumonia With Neural Networks.

As being radiation-free, portable, and capable of repetitive use, ultrasonography is playing an important role in diagnosing and evaluating the COVID-19 Pneumonia (PN) in this epidemic. By virtue of lung ultrasound scores (LUSS), lung ultrasound (LUS) was used to estimate the excessive lung fluid that is an important clinical manifestation of COVID-19 PN, with high sensitivity and specificity. However, as a qualitative method, LUSS suffered from large interobserver variations and requirement for experienced clinicians. Considering this limitation, we developed a quantitative and automatic lung ultrasound scoring system for evaluating the COVID-19 PN. A total of 1527 ultrasound images prospectively collected from 31 COVID-19 PN patients with different clinical conditions were evaluated and scored with LUSS by experienced clinicians. All images were processed via a series of computer-aided analysis, including curve-to-linear conversion, pleural line detection, region-of-interest (ROI) selection, and feature extraction. A collection of 28 features extracted from the ROI was specifically defined for mimicking the LUSS. Multilayer fully connected neural networks, support vector machines, and decision trees were developed for scoring LUS images using the fivefold cross validation. The model with 128×256 two fully connected layers gave the best accuracy of 87%. It is concluded that the proposed method could assess the ultrasound images by assigning LUSS automatically with high accuracy, potentially applicable to the clinics.

Impact of Shenfu injection on a composite of organ dysfunction development in critically ill patients with coronavirus disease 2019 (COVID-19): A structured summary of a study protocol for a randomized controlled trial.

OBJECTIVES: This study aims to determine the protection provided by Shenfu injection (a traditional Chinese medicine) against development of organ dysfunction in critically ill patients with coronavirus disease 2019 (COVID-19). TRIAL DESIGN: This study is a multicenter, randomized, controlled, open-label, two-arm ratio 1:1, parallel group clinical trial. PARTICIPANTS: The patients, who are aged from 18 to 75 years old, with a confirmed or suspected diagnosis of severe or critical COVID-19, will be consecutively recruited in the study, according to the guideline on diagnosis and treatment of COVID-19 (the 7th version) issued by National Health Commission of the People's Republic of China. Exclusion criteria include pregnant and breastfeeding women, atopy or allergies to Shenfu Injection (SFI), severe underlying disease (malignant tumor with multiple metastases, uncontrolled hemopathy, cachexia, severe malnutrition, HIV), active bleeding, obstructive pneumonia caused by lung tumor, severe pulmonary interstitial fibrosis, alveolar proteinosis and allergic alveolitis, continuous use of immunosuppressive drugs in last 6 months, organ transplantation, expected death within 48 hours, the patients considered unsuitable for this study by researchers. The study is conducted in 11 ICUs of designated hospitals for COVID-19, located in 5 cities of China. INTERVENTION AND COMPARATOR: The enrolled patients will randomly receive 100 ml SFI (study group) or identical volume of saline (control group) twice a day for seven consecutive days. Patients in the both groups will be given usual care and the necessary supportive therapies as recommended by the latest edition of the management guidelines for COVID-19 (the 7th version so far). MAIN OUTCOMES: The primary endpoint is a composite of newly developed or exacerbated organ dysfunction. This is defined as an increase in the sequential organ failure assessment (SOFA) score of two or more, indicating sepsis and involvement of at least one organ. The SOFA score will be measured for the 14 days after enrolment from the baseline (the score at randomization). The secondary endpoints are shown below: • SOFA score in total • Pneumonia severity index score • Dosage of vasoactive drugs • Ventilation free days within 28 days • Length of stay in intensive care unit • Total hospital costs to treat the patient • 28-day mortality • The incidence of adverse drug events related to SFI RANDOMISATION: The block randomization codes were generated by SAS V.9.1 for allocation of participants in this study. The ratio of random distribution is 1:1. The sealed envelope method is used for allocation concealment. BLINDING (MASKING): The patients and statistical personnel analyzing study data are both blinded. The blinding of group assignment is not adopted for the medical staff. NUMBERS TO BE RANDOMISED (SAMPLE SIZE): This study is expected to recruit 300 patients with COVID-19, (150 in each group). TRIAL STATUS: Protocol version 2.0, February 15, 2020. Patient recruitment started on February 25, and will end on August 31, 2020. TRIAL REGISTRATION: Chinese Clinical Trial Registry: ChiCTR2000030043. Registered February 21, 2020, http://www.chictr.org.cn/showprojen.aspx?proj=49866 FULL PROTOCOL: The full protocol is attached as an additional file, accessible from the Trials website (Additional file 1). In the interest in expediting dissemination of this material, the familiar formatting has been eliminated; this letter serves as a summary of the key elements of the full protocol.

Antiviral agents, glucocorticoids, antibiotics, and intravenous immunoglobulin in 1142 patients with coronavirus disease 2019: a systematic review and meta-analysis.

INTRODUCTION: The treatment effects of antiviral agents, glucocorticoids, antibiotics, and intravenous immunoglobulin are controversial in patients with coronavirus disease 2019 (COVID­19). OBJECTIVES: This study aimed to evaluate the impact of drug therapy on the risk of death in patients with COVID­19. PATIENTS AND METHODS: The PubMed, Embase, Web of Science, Cochrane Library, and major preprint platforms were searched to retrieve articles published until April 7, 2020. Subsequently, the effects of specific drug interventions on mortality of patients with COVID­19 were assessed. Odds ratios (ORs) and relative risks (RRs) with corresponding 95% CIs were pooled using random effects models. RESULTS: Of 3421 references, 6 studies were included. Pooled results from retrospective studies revealed that antiviral agents may contribute to survival benefit (OR, 0.42; 95% CI, 0.17-0.99; P = 0.048; I2 = 82.8%), whereas a single randomized controlled trial found no effects of an antiviral agent on mortality (RR, 0.77; 95% CI, 0.45-1.3; P = 0.33). Glucocorticoid use led to an increased risk of death (OR, 2.43; 95% CI, 1.44-4.1; P = 0.001; I2 = 61.9%). Antibiotics did not significantly affect mortality (OR, 1.13; 95% CI, 0.67-1.89; P = 0.64; I2 = 0%). Similarly, intravenous immunoglobulin had a nonsignificant effect on mortality (OR, 2.66; 95% CI, 0.72-9.89; P = 0.14; I2 = 93.1%). CONCLUSIONS: With the varied heterogeneities across interventions, the current evidence indicated a probable survival benefit from antiviral agent use and a harmful effect of glucocorticoids in patients with COVID­19. Neither any of antibiotics nor intravenous immunoglobulin were associated with survival benefit in this population.