Sputum deposition classification for mechanically ventilated patients using LSTM method based on airflow signals.

A novel sputum deposition classification method for mechanically ventilated patients based on the long-short-term memory network (LSTM) method was proposed in this study. A wireless ventilation airflow signals collection system was designed and used in this study. The ventilation airflow signals were collected wirelessly and used for sputum deposition classification. Two hundred sixty data groups from 15 patients in the intensive care unit were compiled and analyzed. A two-layer LSTM framework and 11 features extracted from the airflow signals were used for the model training. The cross-validations were adopted to test the classification performance. The sensitivity, specificity, precision, accuracy, F1 score, and G score were calculated. The proposed method has an accuracy of 84.7 ± 4.1% for sputum and non-sputum deposition classification. Moreover, compared with other classifiers (logistic regression, random forest, naive Bayes, support vector machine, and K-nearest neighbor), the proposed LSTM method is superior. In addition, the other advantages of using ventilation airflow signals for classification are its convenience and low complexity. Intelligent devices such as phones, laptops, or ventilators can be used for data processing and reminding medical staff to perform sputum suction. The proposed method could significantly reduce the workload of medical staff and increase the automation and efficiency of medical care, especially during the COVID-19 pandemic.

Application of deep learning-based diagnostic systems in screening asymptomatic COVID-19 patients among oversea returnees.

INTRODUCTION: Our study aimed to investigate the performance of deep learning (DL)-based diagnostic systems in alerting against COVID-19, especially among asymptomatic individuals coming from overseas, and to analyze the features of identified asymptomatic patients in detail. METHODOLOGY: DL diagnostic systems were deployed to assist in the screening of COVID-19, including the pneumonia system and pulmonary nodules system. 1,917 overseas returnees who underwent CT examination and rRT-PCR tests were enrolled. DL pneumonia system promptly alerted clinicians to suspected COVID-19 after CT examinations while the performance was evaluated with rRT-PCR results as the reference. The radiological features of asymptomatic COVID-19 cases were described according to the Nomenclature of the Fleischner Society. RESULTS: Fifty-three cases were confirmed as COVID-19 patients by rRT-PCR tests, including 5 asymptomatic cases. DL pneumonia system correctly alerted 50 cases as suspected COVID-19 with a sensitivity of 0.9434 and specificity of 0.9592 (within 2 minutes per case); while the pulmonary nodules system alerted 2 of the 3 missed asymptomatic cases. Additionally, five asymptomatic patients presented different characteristics such as elevated creatine kinase level and prolonged prothrombin time, as well as atypical radiological features. CONCLUSIONS: DL diagnostic systems are promising complementary approaches for prompt screening of imported COVID-19 patients, even the imported asymptomatic cases. Unique clinical and radiological characteristics of asymptomatic cases might be of great value in screening as well. ADVANCES IN KNOWLEDGE: DL-based systems are practical, efficient, and reliable to assist radiologists in screening COVID-19 patients. Differential features of asymptomatic patients might be useful to clinicians in the frontline to differentiate asymptomatic cases.

Dynamics of Detection of Anti-SARS-CoV-2 IgG/IgM and SARS-CoV-2 RNA in Respiratory Tract Specimens in 48 COVID-19 Patients.

BACKGROUND: The rapid spread of pneumonia caused by SARS-CoV-2 has seriously threatened people. In this study, we detected the expression of anti-SARS-CoV-2 IgG/IgM and respiratory tract SARS-CoV-2 RNA in patients with COVID-19 and explored the correlation and clinical significance between SARS-CoV-2 antibody and respiratory SARS-CoV-2 RNA. METHODS: From March 5, 2020 to April 28, 2020, 48 cases with COVID-19 diagnosed in Beijing Xiaotangshan Hospital were enrolled. SARS-CoV-2 RNAs were detected by real-time fluorescence RT-PCR method. Serum SARS-CoV-2 IgG/IgM antibodies were determined by colloidal gold immunochromatography. The statistical analysis was performed using chi-squared test. RESULTS: In all the patients, SARS-CoV-2 RNA among 270 upper respiratory tract (nasal or throat swabs) samples, 71 lower respiratory tract (sputum) samples, and anti-SARS-CoV-2 IgM/IgG antibodies in 123 serum samples were detected during the hospitalization period. The positive rate of anti-SARS-CoV-2 IgG was significantly higher than that of anti-SARS-CoV-2 IgM within the first week after symptom onset (p < 0.05). The positive rate of anti-SARS-CoV-2 IgG was also significantly higher than that of anti-SARS-CoV-2 IgM during day 8 - 30 after symptom onset (p < 0.01). The positive rate of SARS-CoV-2 RNA in the lower respiratory tract specimens (64.8%, 46/71) was significantly higher than that in the upper respiratory tract (46.7%, 126/270) (p < 0.05). The positive rate (100%, 4/4) of SARS-CoV-2 RNA detection in the lower respiratory tract specimens before IgG seroconversion was significantly higher than that of the positive rate (59.3%, 32/54) after IgG seroconversion (p < 0.01). The positive rate (72.2%, 57/79) of SARS-CoV-2 RNA detection in the upper respiratory tract specimens before IgG seroconversion was significantly higher than that of the positive rate (30.7%, 39/127) after IgG seroconversion (p < 0.01). CONCLUSIONS: Anti-SARS-CoV-2 IgG might be detected within the first week after symptom onset. The application of SARS-CoV-2 antibody (IgG/IgM) detection is important for the suspected cases of SARS-CoV-2 infection with negative SARS-CoV-2 RNA results. The positive rate of SARS-CoV-2 RNA detection in the lower respiratory tract specimens was significantly higher than that in the upper respiratory tract. Sputum detection is recommended for the detection of SARS-CoV-2 RNA. Using lower respiratory tract specimens may reduce the false negative PCR tests. The detection of SARS-CoV-2 RNA can be improved by investigating follow-up specimens over time.

Prevention of SARS-CoV-2 transmission from international arrivals: Xiaotangshan Designated Hospital, China.

A surge in the number of international arrivals awaiting coronavirus disease 2019 (COVID-19) screening overwhelmed health-care workers and depleted medical resources in designated hospitals in Beijing, China in March 2020. The People's Government of Beijing Municipality therefore issued a policy which required the mandatory transfer of all asymptomatic passengers arriving from a foreign country to designated quarantine hotels, and the transfer of passengers with fever or respiratory symptoms to designated hospitals. Xiaotangshan Designated Hospital, a severe acute respiratory syndrome hospital in 2003, was rapidly renovated and put into operation with the main tasks of screening and isolating symptomatic international arrivals at Beijing Capital International Airport, providing basic medical care for mild to moderate COVID-19-positive cases, and rapidly referring severe to critical COVID-19-positive cases to higher-level hospitals. During the month-long period of its operation, 2171 passengers were screened and 53 were confirmed as having COVID-19 (six severe to critical). We describe how the use of Xiaotangshan Designated Hospital in this way enabled the efficient grouping and assessment of passengers arriving from a foreign country, the provision of optimal patient care without compromising public safety and the prioritization of critically ill patients requiring life-saving treatment. The designated hospital is a successful example of the World Health Organization's recommendation to renovate existing medical infrastructures to improve the COVID-19 response capacity. The flexible design of Xiaotangshan Designated Hospital means that it can be repurposed and reopened at any time to respond to the changing pandemic conditions.

En mars 2020, la brusque hausse du nombre d'arrivées internationales en attente de dépistage de la maladie à coronavirus 2019 (COVID-19) a submergé les professionnels de la santé et épuisé les ressources médicales dans les hôpitaux de référence à Beijing, en Chine. Le gouvernement populaire de la municipalité de Beijing a réagi en ordonnant que tous les passagers asymptomatiques en provenance d'un pays étranger soient transférés vers des hôtels reconvertis en centres de quarantaine, et que ceux manifestant de la fièvre ou des symptômes respiratoires soient envoyés dans des hôpitaux de référence. L'hôpital de référence Xiaotangshan, construit en 2003 pour lutter contre le syndrome respiratoire aigu sévère, a rapidement été rénové et mis en service. Ses tâches principales: dépister et isoler les passagers internationaux symptomatiques débarquant au Beijing Capital International Airport, prodiguer les soins médicaux de base aux cas positifs de COVID-19 souffrant d'une forme légère à modérée, et adresser dès que possible les cas positifs de COVID-19 dans un état grave ou critique aux hôpitaux spécialisés. En l'espace d'un mois, 2171 passagers ont été testés et 53 se sont révélés positifs à la COVID-19 (6 étant dans un état grave ou critique). Nous décrivons la façon dont l'hôpital de référence Xiaotangshan a ainsi permis de regrouper et d'évaluer efficacement les arrivées en provenance de l'étranger, d'offrir une prise en charge optimale des patients sans compromettre la sécurité publique, et d'établir des priorités afin que les malades gravement atteints puissent bénéficier d'un traitement dans les plus brefs délais. Cet hôpital de référence est un exemple réussi de la mise en œuvre de la recommandation formulée par l'Organisation mondiale de la Santé: rénover les infrastructures médicales existantes afin d'améliorer les capacités de lutte contre la COVID-19. Grâce à sa conception flexible, l'hôpital Xiaotangshan peut être réutilisé et rouvert à n'importe quel moment pour réagir à un contexte pandémique en perpétuelle évolution.

Un aumento del número de llegadas de vuelos internacionales en espera de la detección del coronavirus 2019 (COVID-19) sobrecargó al personal sanitario y agotó los recursos médicos en los hospitales designados de Pekín (China) en marzo de 2020. Por lo tanto, la policía del Gobierno Popular del municipio de Pekín se tuvo que hacer cargo del traslado obligatorio de todos los pasajeros asintomáticos que llegaran de un país extranjero a los hoteles de cuarentena designados, y el traslado de los pasajeros con fiebre o síntomas respiratorios a los hospitales designados. El hospital designado de Xiaotangshan, un hospital especializado en el síndrome respiratorio agudo severo en 2003, se rehabilitó rápidamente y se puso en funcionamiento con las tareas principales de examinar y aislar a los sintomáticos que llegaban al Aeropuerto Internacional de Pekín, proporcionando atención médica básica a los casos positivos de COVID-19 de leves a moderados, y derivando rápidamente los casos positivos de COVID-19 de graves a críticos a hospitales de nivel superior. Durante el mes que duró su funcionamiento, se examinó a 2.171 pasajeros y se confirmó que 53 tenían la COVID-19 (6 de ellos con intensidad de grave a crítica). Describimos cómo el uso del hospital designado de Xiaotangshan permitió agrupar y evaluar eficazmente a los pasajeros que llegaban de un país extranjero, prestar una atención óptima a los pacientes sin comprometer la seguridad pública y priorizar a los pacientes en estado crítico que requerían tratamiento para salvar su vida. El hospital designado es un ejemplo de éxito de la recomendación de la Organización Mundial de la Salud de renovar las infraestructuras médicas existentes para mejorar la capacidad de respuesta ante la COVID-19. El diseño flexible del hospital designado de Xiaotangshan significa que puede utilizarse y volver a habilitarse en cualquier momento para responder a las condiciones cambiantes de la pandemia.

Moderate vs. mild cases of overseas-imported COVID-19 in Beijing: a retrospective cohort study.

This study compared the differences in the clinical manifestations, treatment courses and clinical turnover between mild and moderate coronavirus disease 2019 (COVID-19). Clinical data of the patients with imported COVID-19 admitted to Beijing Xiaotangshan Designated Hospital between March 15 and April 30, 2020, were retrospectively analysed. A total of 53 COVID-19 patients were included, with 21 mild and 32 moderate cases. Compared with the mild group, the moderate group showed significant differences in breathing frequency, lymphocyte count, neutrophil percentage, neutrophil/lymphocyte ratio, procalcitonin, C-reactive protein, and dynamic erythrocyte sedimentation rate. In the moderate group, 87.5% exhibited ground-glass opacities, 14% exhibited consolidative opacities, 53.1% exhibited local lesions and 68.8% exhibited unilateral lesions. The proportion of patients who received antiviral or antibiotic treatment in the moderate group was higher than that in the mild group, and the number of cases that progressed to severe disease in the moderate group was also significantly higher (18.7% vs. 0%, p = 0.035). Compared with patients with mild COVID-19, those with moderate COVID-19 exhibited more noticeable inflammatory reactions, more severe pulmonary imaging manifestations and earlier expression of protective antibodies. The overall turnover of the moderate cases was poorer than that of the mild cases.

Mechanical ventilation strategy for pulmonary rehabilitation based on patient-ventilator interaction.

Mechanical ventilation is an effective medical means in the treatment of patients with critically ill, COVID-19 and other pulmonary diseases. During the mechanical ventilation and the weaning process, the conduct of pulmonary rehabilitation is essential for the patients to improve the spontaneous breathing ability and to avoid the weakness of respiratory muscles and other pulmonary functional trauma. However, inappropriate mechanical ventilation strategies for pulmonary rehabilitation often result in weaning difficulties and other ventilator complications. In this article, the mechanical ventilation strategies for pulmonary rehabilitation are studied based on the analysis of patient-ventilator interaction. A pneumatic model of the mechanical ventilation system is established to determine the mathematical relationship among the pressure, the volumetric flow, and the tidal volume. Each ventilation cycle is divided into four phases according to the different respiratory characteristics of patients, namely, the triggering phase, the inhalation phase, the switching phase, and the exhalation phase. The control parameters of the ventilator are adjusted by analyzing the interaction between the patient and the ventilator at different phases. A novel fuzzy control method of the ventilator support pressure is proposed in the pressure support ventilation mode. According to the fuzzy rules in this research, the plateau pressure can be obtained by the trigger sensitivity and the patient's inspiratory effort. An experiment prototype of the ventilator is established to verify the accuracy of the pneumatic model and the validity of the mechanical ventilation strategies proposed in this article. In addition, through the discussion of the patient-ventilator asynchrony, the strategies for mechanical ventilation can be adjusted accordingly. The results of this research are meaningful for the clinical operation of mechanical ventilation. Besides, these results provide a theoretical basis for the future research on the intelligent control of ventilator and the automation of weaning process.

Comparison of the clinical manifestations between different age groups of patients with overseas imported COVID-19.

The current study investigated the clinical manifestations and outcomes of different age groups of patients with overseas imported COVID-19. In total, 53 COVID-19 patients admitted to the designated Beijing Xiaotangshan Hospital between March 16 and April 15 of 2020 were included. Based on the percentage of disease aggravation during hospital stay according to CT, the patients were divided into two groups: ≤40 years (group A; n = 41) and >40 years (group B; n = 12). The demographic data, epidemiological history, disease courses, potential complications, clinical symptoms, lab indices, chest CT outcomes, treatment protocols and turnovers of the two groups were compared. According to clinical typing, compared with group A, group B had a significantly greater proportion of the common type of COVID-19 (P<0.05) and greater comorbidity of type 2 diabetes (P<0.001). The two groups presented significantly different lab indices. Group B showed significantly more frequent CT abnormalities, with greater proportions of multiple lesions and bilateral lung involvement (P<0.05). During hospitalization, group B had a greater proportion of disease aggravation according to CT (P<0.01). Compared with group A, group B received a significantly greater proportion of antiviral therapy and presented a significantly greater occurrence of adverse drug reactions (P<0.05). The two groups did not significantly differ in time from admission to clinical symptom improvement or from disease onset to negative outcomes according to nucleic acid testing, the appearance of IgG or the appearance of IgM. They also did not significantly differ in length of stay. Older imported COVID-19 patients, particularly those with type 2 diabetes, showed a broader pulmonary extent and faster development of the disease, more severe pathogenetic conditions and a greater risk of developing a critically severe type. Increased attention should be given to this population in clinical practice.

A Novel Method to Evaluate Patient-Ventilator Synchrony during Mechanical Ventilation

The synchrony of patient-ventilator interaction affects the process of mechanical ventilation which is clinically applied for respiratory support. The occurrence of patient-ventilator asynchrony (PVA) not only increases the risk of ventilator complications but also affects the comfort of patients. To solve the problem of uncertain patient-ventilator interaction in the mechanical ventilation system, a novel method to evaluate patient-ventilator synchrony is proposed in this article. Firstly, a pneumatic model is established to simulate the mechanical ventilation system, which is verified to be accurate by the experiments. Then, the PVA phenomena are classified and detected based on the analysis of the ventilator waveforms. On this basis, a novel synchrony index SIhao is established to evaluate the patient-ventilator synchrony. It not only solves the defects of previous evaluation indexes but also can be used as the response parameter in the future research of ventilator control algorithms. The accurate evaluation of patient-ventilator synchrony can be applied to the adjustment of clinical strategies and the pathological analyses of patients. This research can also reduce the burden on clinicians and help to realize the adaptive control of the mechanical ventilation and weaning process in the future.