RESUMEN
BACKGROUND: Upper gastrointestinal cancers (including oesophageal cancer and gastric cancer) are the most common cancers worldwide. Artificial intelligence platforms using deep learning algorithms have made remarkable progress in medical imaging but their application in upper gastrointestinal cancers has been limited. We aimed to develop and validate the Gastrointestinal Artificial Intelligence Diagnostic System (GRAIDS) for the diagnosis of upper gastrointestinal cancers through analysis of imaging data from clinical endoscopies. METHODS: This multicentre, case-control, diagnostic study was done in six hospitals of different tiers (ie, municipal, provincial, and national) in China. The images of consecutive participants, aged 18 years or older, who had not had a previous endoscopy were retrieved from all participating hospitals. All patients with upper gastrointestinal cancer lesions (including oesophageal cancer and gastric cancer) that were histologically proven malignancies were eligible for this study. Only images with standard white light were deemed eligible. The images from Sun Yat-sen University Cancer Center were randomly assigned (8:1:1) to the training and intrinsic verification datasets for developing GRAIDS, and the internal validation dataset for evaluating the performance of GRAIDS. Its diagnostic performance was evaluated using an internal and prospective validation set from Sun Yat-sen University Cancer Center (a national hospital) and additional external validation sets from five primary care hospitals. The performance of GRAIDS was also compared with endoscopists with three degrees of expertise: expert, competent, and trainee. The diagnostic accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of GRAIDS and endoscopists for the identification of cancerous lesions were evaluated by calculating the 95% CIs using the Clopper-Pearson method. FINDINGS: 1â036â496 endoscopy images from 84â424 individuals were used to develop and test GRAIDS. The diagnostic accuracy in identifying upper gastrointestinal cancers was 0·955 (95% CI 0·952-0·957) in the internal validation set, 0·927 (0·925-0·929) in the prospective set, and ranged from 0·915 (0·913-0·917) to 0·977 (0·977-0·978) in the five external validation sets. GRAIDS achieved diagnostic sensitivity similar to that of the expert endoscopist (0·942 [95% CI 0·924-0·957] vs 0·945 [0·927-0·959]; p=0·692) and superior sensitivity compared with competent (0·858 [0·832-0·880], p<0·0001) and trainee (0·722 [0·691-0·752], p<0·0001) endoscopists. The positive predictive value was 0·814 (95% CI 0·788-0·838) for GRAIDS, 0·932 (0·913-0·948) for the expert endoscopist, 0·974 (0·960-0·984) for the competent endoscopist, and 0·824 (0·795-0·850) for the trainee endoscopist. The negative predictive value was 0·978 (95% CI 0·971-0·984) for GRAIDS, 0·980 (0·974-0·985) for the expert endoscopist, 0·951 (0·942-0·959) for the competent endoscopist, and 0·904 (0·893-0·916) for the trainee endoscopist. INTERPRETATION: GRAIDS achieved high diagnostic accuracy in detecting upper gastrointestinal cancers, with sensitivity similar to that of expert endoscopists and was superior to that of non-expert endoscopists. This system could assist community-based hospitals in improving their effectiveness in upper gastrointestinal cancer diagnoses. FUNDING: The National Key R&D Program of China, the Natural Science Foundation of Guangdong Province, the Science and Technology Program of Guangdong, the Science and Technology Program of Guangzhou, and the Fundamental Research Funds for the Central Universities.
Asunto(s)
Algoritmos , Inteligencia Artificial , Endoscopía/métodos , Neoplasias Gastrointestinales/diagnóstico , Procesamiento de Imagen Asistido por Computador/métodos , Adolescente , Adulto , Anciano , Anciano de 80 o más Años , Estudios de Casos y Controles , Niño , Femenino , Estudios de Seguimiento , Humanos , Masculino , Persona de Mediana Edad , Pronóstico , Estudios Prospectivos , Curva ROC , Estudios Retrospectivos , Adulto JovenRESUMEN
BACKGROUND: Non-pharmaceutical measures and travel restrictions have halted the spread of coronavirus disease 2019 (COVID-19) and influenza. Nonetheless, with COVID-19 restrictions lifted, an unanticipated outbreak of the influenza B/Victoria virus in late 2021 and another influenza H3N2 outbreak in mid-2022 occurred in Guangdong, southern China. The mechanism underlying this phenomenon remains unknown. To better prepare for potential influenza outbreaks during COVID-19 pandemic, we studied the molecular epidemiology and phylogenetics of influenza A(H3N2) and B/Victoria that circulated during the COVID-19 pandemic in this region. METHODS: From January 1, 2018 to December 31, 2022, we collected throat swabs from 173,401 patients in Guangdong who had acute respiratory tract infections. Influenza viruses in the samples were tested using reverse transcription-polymerase chain reaction, followed by subtype identification and sequencing of hemagglutinin (HA) and neuraminidase (NA) genes. Phylogenetic and genetic diversity analyses were performed on both genes from 403 samples. A rigorous molecular clock was aligned with the phylogenetic tree to measure the rate of viral evolution and the root-to-tip distance within strains in different years was assessed using regression curve models to determine the correlation. RESULTS: During the early period of COVID-19 control, various influenza viruses were nearly undetectable in respiratory specimens. When control measures were relaxed in January 2020, the influenza infection rate peaked at 4.94% (39/789) in December 2021, with the influenza B/Victoria accounting for 87.18% (34/39) of the total influenza cases. Six months later, the influenza infection rate again increased and peaked at 11.34% (255/2248) in June 2022; influenza A/H3N2 accounted for 94.51% (241/255) of the total influenza cases in autumn 2022. The diverse geographic distribution of HA genes of B/Victoria and A/H3N2 had drastically reduced, and most strains originated from China. The rate of B/Victoria HA evolution (3.11 × 10-3, P < 0.05) was 1.7 times faster than before the COVID-19 outbreak (1.80 × 10-3, P < 0.05). Likewise, the H3N2 HA gene's evolution rate was 7.96 × 10-3 (P < 0.05), which is 2.1 times faster than the strains' pre-COVID-19 evolution rate (3.81 × 10-3, P < 0.05). CONCLUSIONS: Despite the extraordinarily low detection rate of influenza infection, concealed influenza transmission may occur between individuals during strict COVID-19 control. This ultimately leads to the accumulation of viral mutations and accelerated evolution of H3N2 and B/Victoria viruses. Monitoring the evolution of influenza may provide insights and alerts regarding potential epidemics in the future.
Asunto(s)
COVID-19 , Subtipo H3N2 del Virus de la Influenza A , Virus de la Influenza B , Gripe Humana , Epidemiología Molecular , Filogenia , SARS-CoV-2 , Humanos , COVID-19/epidemiología , COVID-19/virología , COVID-19/transmisión , Subtipo H3N2 del Virus de la Influenza A/genética , Subtipo H3N2 del Virus de la Influenza A/aislamiento & purificación , China/epidemiología , Gripe Humana/epidemiología , Gripe Humana/virología , Virus de la Influenza B/genética , Virus de la Influenza B/aislamiento & purificación , Virus de la Influenza B/clasificación , SARS-CoV-2/genética , Adulto , Persona de Mediana Edad , Masculino , Femenino , Pandemias , Adulto Joven , Anciano , Glicoproteínas Hemaglutininas del Virus de la Influenza/genética , Adolescente , Neuraminidasa/genética , Niño , PreescolarRESUMEN
BACKGROUND: Recent outbreaks of avian influenza and ongoing virus reassortment have drawn focus on spill-over infections. The increase in human infections with highly pathogenic avian influenza H5N6 virus and its high fatality rate posed a potential threat, necessitating the search for a more effective treatment. METHODS: Longitudinal clinical data and specimens were collected from five H5N6 patients after admission. All patients received antiviral treatment of either sequential monotherapy of oseltamivir and baloxavir or the two drugs in combination. Severity of illness; viral load in sputum, urine, and blood; and cytokine levels in serum and sputum were serially analyzed. FINDINGS: All patients developed acute respiratory distress syndrome (ARDS) and viral sepsis within 1 week after disease onset. When delayed oseltamivir showed poor effects, baloxavir was administered and rapidly decreased viral load. In addition, levels of IL-18, M-CSF, IL-6, and HGF in sputum and Mig and IL-18 in serum that reflected ARDS and sepsis deterioration, respectively, were also reduced with baloxavir usage. However, three patients eventually died from exacerbation of underlying disease and secondary bacterial infection. Nonsurvivors had more severe extrapulmonary organ dysfunction and insufficient H5N6 virus-specific antibody response. CONCLUSIONS: For critical human cases of H5N6 infection, baloxavir demonstrated effects on viral load and pulmonary/extrapulmonary cytokines, even though treatment was delayed. Baloxavir could be regarded as a first-line treatment to limit continued viral propagation, with potential future application in avian influenza human infections and poultry workers exhibiting influenza-like illness. FUNDING: This work was funded by the National Natural Science Foundation of China (81761128014).
Asunto(s)
Dibenzotiepinas , Virus de la Influenza A , Gripe Aviar , Gripe Humana , Morfolinas , Piridonas , Síndrome de Dificultad Respiratoria , Sepsis , Triazinas , Animales , Humanos , Gripe Aviar/tratamiento farmacológico , Gripe Aviar/epidemiología , Oseltamivir/uso terapéutico , Subtipo H5N6 del Virus de la Influenza A , Interleucina-18/uso terapéutico , Gripe Humana/tratamiento farmacológico , Gripe Humana/epidemiología , Síndrome de Dificultad Respiratoria/tratamiento farmacológico , Sepsis/tratamiento farmacológicoRESUMEN
Background: The coronavirus disease 2019 (COVID-19) pandemic caused by severe acute respiratory coronavirus-2 (SARS-CoV-2) has placed enormous diagnostic burden on hospitals and testing laboratories. It is thus critical for such facilities to optimize the diagnostic process to enable maximum testing on minimum resources. The current standard of diagnosis is the detection of the viral nucleic acid in clinical specimens. Methods: In order to optimize the laboratory's nucleic acid testing system for COVID-19, we performed a Discrete-Event-Simulation using the Arena Simulation Software to model the detection process based on the data obtained from the First Affiliated Hospital of Guangzhou Medical University (FAHGMU). The maximum of total time that specimens spent and the equipment consumption was compared under different scenarios in the model. Results: Seven scenarios were performed to simulate actual situation and improved situations. We analyzed conditions that adding a new nucleic acid extraction system (NAES), shifting a member from night duty to morning duty, using specimen tubes containing guanidine isothiocyanate (GITC), then tested the maximum testing capacity in the current number of technicians. In addition, the costs including personal protective equipment (PPE) and testing kits was calculated. Conclusions: A work schedule based on specimen-load improves efficiency without incurring additional costs, while using the specimen tubes containing GITC could reduce testing time by 30 min. In contrast, adding new NAESs or polymerase chain reaction (PCR) instruments has minimal impact on testing efficiency.
RESUMEN
To inform seroepidemiological studies, we characterized the IgG- responses in COVID-19 patients against the two major SARS-CoV-2 viral proteins, spike (S) and nucleocapsid (N). We tested 70 COVID-19 sera collected up to 85 days post-symptom onset and 230 non-COVID-19 sera, including 27 SARS sera from 2003. Although the average SARS-CoV-2 S and N-IgG titers were comparable, N-responses were more variable among individuals. S- and N-assay specificity tested with non-COVID-19 sera were comparable at 97.5% and 97.0%, respectively. Therefore, S will make a better target due to its lower cross-reactive potential and its' more consistent frequency of detection compared to N.
Asunto(s)
Anticuerpos Antivirales/sangre , Proteínas de la Nucleocápside de Coronavirus/inmunología , SARS-CoV-2/inmunología , Glicoproteína de la Espiga del Coronavirus/inmunología , Adulto , Anciano , Anciano de 80 o más Años , Reacciones Cruzadas , Humanos , Persona de Mediana Edad , Fosfoproteínas/inmunología , Coronavirus Relacionado al Síndrome Respiratorio Agudo Severo/inmunologíaRESUMEN
China implemented stringent non-pharmaceutical interventions (NPIs) in spring 2020, which has effectively suppressed SARS-CoV-2. In this study, we utilized data from routine respiratory virus testing requests from physicians and examined circulation of 11 other respiratory viruses in Southern China, from January 1, 2018 to December 31, 2020. A total of 58,169 throat swabs from patients with acute respiratory tract infections (ARTIs) were collected and tested. We found that while the overall activity of respiratory viruses was lower during the period with stringent NPIs, virus activity rebounded shortly after the NPIs were relaxed and social activities resumed. Only influenza was effectively suppressed with very low circulation which extended to the end of 2020. Circulation of other respiratory viruses in the community was maintained even during the period of stringent interventions, especially for rhinovirus. Our study shows that NPIs against COVID-19 have different impacts on respiratory viruses.
RESUMEN
BACKGROUND: Severe COVID-19 patients typically test positive for SARS-CoV-2 RNA for extended periods of time, even after recovery from severe disease. Due to the timeframe involved, these patients may have developed humoral immunity to SARS-CoV-2 while still testing positive for viral RNA in swabs. Data are lacking on exposure risks in these situations. Here, we studied SARS-CoV-2 environmental contamination in an ICU and an isolation ward caring for such COVID-19 patients. METHODS: We collected air and surface samples in a hospital caring for critical and severe COVID-19 cases from common areas and areas proximal to patients. RESULTS: Of the 218 ICU samples, an air sample contained SARS-CoV-2 RNA. Of the 182 isolation ward samples, nine contained SARS-CoV-2 RNA. These were collected from a facemask, the floor, mobile phones, and the air in the patient room and bathroom. Serum antibodies against SARS-CoV-2 were detected in these patients at the beginning of the study. CONCLUSIONS: While there is a perception of increased risk in the ICU, our study demonstrates that isolation wards may pose greater risks to healthcare workers and exposure risks remain with clinically improved patients, weeks after their initial diagnoses. As these patients had serum antibodies, further studies may be warranted to study the utility of serum antibodies as a surrogate of viral clearance in allowing people to return to work. We recommend continued vigilance even with patients who appear to have recovered from COVID-19.