Fully Automated Deep Learning-enabled Detection for Hepatic Steatosis on Computed Tomography: A Multicenter International Validation Study (preprint)

Despite high global prevalence of hepatic steatosis, no automated diagnostics demonstrated generalizability in detecting steatosis on multiple international datasets. Traditionally, hepatic steatosis detection relies on clinicians selecting the region of interest (ROI) on computed tomography (CT) to measure liver attenuation. ROI selection demands time and expertise, and therefore is not routinely performed in populations. To automate the process, we validated an existing artificial intelligence (AI) system for 3D liver segmentation and used it to purpose a novel method: AI-ROI, which could automatically select the ROI for attenuation measurements. AI segmentation and AI-ROI method were evaluated on 1,014 non-contrast enhanced chest CT images from eight international datasets: LIDC-IDRI, NSCLC-Lung1, RIDER, VESSEL12, RICORD-1A, RICORD-1B, COVID-19-Italy, and COVID-19-China. AI segmentation achieved a mean dice coefficient of 0.957. Attenuations measured by AI-ROI showed no significant differences (p = 0.545) and a reduction of 71% time compared to expert measurements. The area under the curve (AUC) of the steatosis classification of AI-ROI is 0.921 (95% CI: 0.883 - 0.959). If performed as a routine screening method, our AI protocol could potentially allow early non-invasive, non-pharmacological preventative interventions for hepatic steatosis. 1,014 expert-annotated liver segmentations of patients with hepatic steatosis annotations can be downloaded here: https://drive.google.com/drive/folders/1-g_zJeAaZXYXGqL1OeF6pUjr6KB0igJX.

Trend of respiratory pathogens during the COVID-19 epidemic: comparison between 2020 and the last 5 years (preprint)

In China, the first SARS-CoV-2 infection was diagnosed in Wuhan on December 8. Spreads in other regions have occurred since the end of January, happens to be the start of Lunar New Year holiday. In this study, we analyzed the prevalence of common respiratory pathogens in children with respiratory infections during the SARS-CoV-2 pandemic and compared them with the time trends from 2016 to 2019. Overall, results obtained indicate that the time trend of other respiratory infections were significantly different from previous years, especially the pattern of influenza and Mycoplasma pneumonia. Therefore, in the current scenario of COVID-19 pandemic, other common pathogens testing should not be excluded. The natural home isolation period in new year holiday may weaken the transmission of common respiratory viruses.

Evaluation of two commercial multiplex PCR tests for the diagnosis of acute respiratory infections in hospitalized children (preprint)

Background Acute respiratory tract infections (ARTI), including the common cold, pharyngitis, sinusitis, otitis media, tonsillitis, bronchiolitis and pneumonia are the most common diagnoses among pediatric patients and account for the majority of antibiotic prescriptions. A clear and rapid diagnosis is the key to preventing antibiotic abuse. Recently, based on different detection principles, many multi-target molecular analyses that can simultaneously detect dozens of pathogens have been developed, thereby greatly improving sensitivity and shortening turnaround time. In this work, we conducted a head-to-head comparative study between melting curve analysis (MCA) and capillary electrophoresis assay (CE) in the detection of nine respiratory pathogens in sputum samples collected from hospitalized ARTI childre.Methods Through MCA and CE analysis, nine common respiratory pathogens were tested on hospitalized children under the age of 13 who met the ARTI criteria.Results A total of 237 children with sputum specimens were tested. For all the targets combined, the positive detection rate of XYRes-MCA was significantly higher than ResP-CE (72.2% vs. 63.7%, p = .002). Some pathogens were detected more often with MCA, such as parainfluenza virus, influenza B and coronavirus, and some pathogens do the opposite, such as adenovirus and influenza A (all p < .01). Very good kappa values for most of pathogens were observed, except for Influenza B and coronavirus (both κ = .39).Conclusions Multiplex melting curve and capillary electrophoresis assays performed similarly for the detection of common respiratory pathogens in hospitalized children, except for Influenza B and coronavirus. Higher sensitivity was observed in the melting curve assay. By using this sensitive and rapid test, it may improv patient prognosis and antimicrobial management.

Evaluation of two commercial multiplex PCR tests for the diagnosis of acute respiratory infections in hospitalized children (preprint)

Background Acute respiratory tract infections (ARTI), including the common cold, pharyngitis, sinusitis, otitis media, tonsillitis, bronchiolitis and pneumonia are the most common diagnoses in pediatric patients, and account for most antibiotic prescriptions. A confirmed and rapid ARTI diagnosis is key to preventing antibiotic abuse. Recently, based on different detection principles, many multi-target molecular analyses that can detect dozens of pathogens at the same time have been developed, greatly improving sensitivity and shortening turnaround time. In this work, we performed a head-to-head comparative study between melting curve analysis (MCA) and capillary electrophoresis assay (CE) in the detection of nine respiratory pathogens in sputum samples collected from hospitalized children with ARTI. Methods By MCA and CE analysis, nine common respiratory pathogens were tested in hospitalized children< 13 years of age who met the ARTI criteria respectively. Results A total of 237 children with sputum specimens were tested. For all the targets combined, the positive detection rate of XYRes-MCA was significantly higher than that of ResP-CE (72.2% vs. 63.7%, p=.002). Some pathogens were detected more often with MCA, such as parainfluenza virus, influenza B and coronavirus, and some pathogens do the opposite, such as adenovirus and influenza A (all p<.01). Very good kappa values for most of pathogens were observed, except for Influenza B and coronavirus (both κ=.39). Conclusions Multiplex melting curve and capillary electrophoresis assays performed similarly for the detection of common respiratory pathogens in hospitalized children, except for Influenza B and coronavirus. A higher sensitivity was observed in the melting curve assay. By using this sensitive and rapid test, it may be possible to achieve improved patient prognosis and antimicrobial management.