Showing papers by "Jun Xia published in 2022"

Significant histological disease of patients with chronic hepatitis B virus infection in the grey zone

Abstract: Many patients with chronic hepatitis B (CHB) do not meet the definitions of the traditional natural phases and are classified as being in the grey zone (GZ).

Explainable COVID-19 Infections Identification and Delineation Using Calibrated Pseudo Labels

Abstract: The upheaval brought by the arrival of the COVID-19 pandemic has continued to bring fresh challenges over the past two years. During this COVID-19 pandemic, there has been a need for rapid identification of infected patients and specific delineation of infection areas in computed tomography (CT) images. Although deep supervised learning methods have been established quickly, the scarcity of both image-level and pixel-level labels as well as the lack of explainable transparency still hinder the applicability of AI. Can we identify infected patients and delineate the infections with extreme minimal supervision? Semi-supervised learning has demonstrated promising performance under limited labelled data and sufficient unlabelled data. Inspired by semi-supervised learning, we propose a model-agnostic calibrated pseudo-labelling strategy and apply it under a consistency regularization framework to generate explainable identification and delineation results. We demonstrate the effectiveness of our model with the combination of limited labelled data and sufficient unlabelled data or weakly-labelled data. Extensive experiments have shown that our model can efficiently utilize limited labelled data and provide explainable classification and segmentation results for decision-making in clinical routine.

Value of MRI-based semi-quantitative structural neuroimaging in predicting the prognosis of patients with idiopathic normal pressure hydrocephalus after shunt surgery

Abstract: To explore the value of structural neuroimaging in predicting the prognosis of shunt surgery for idiopathic normal-pressure hydrocephalus (iNPH) using two different standard semi-quantitative imaging scales.A total of 47 patients with iNPH who underwent shunt surgery at our hospital between 2018 and 2020 were included in this study. The modified Rankin Scale (mRS) and iNPH grading scale (iNPHGS) were used to evaluate and quantify the clinical symptoms before and after shunt surgery. The disproportionately enlarged subarachnoid space hydrocephalus (DESH) and iNPH Radscale scores were used to evaluate the preoperative MR images. The primary endpoint was improvement in the mRS score a year after surgery, and the secondary endpoint was the iNPHGS after 1 year. The preoperative imaging features of the improved and non-improved groups were compared.The rates of the primary and secondary outcomes were 59.6% and 61.7%, respectively, 1 year after surgery. There were no significant differences in preoperative DESH score, iNPH Radscale, Evans' index (EI), or callosal angle (CA) between the improved and non-improved groups. Significant correlations were observed between the severity of gait disorder and EI and the CA.The value of structural neuroimaging in predicting the prognosis of shunt surgery is limited, and screening for shunt surgery candidates should not rely only on preoperative imaging findings.• Early shunt surgery can significantly improve the clinical symptoms and prognosis of patients with idiopathic normal-pressure hydrocephalus (iNPH). • Structural imaging findings have limited predictiveness for the prognosis of patients with iNPH after shunt surgery. • Patients should not be selected for shunt surgery based on only structural imaging findings.

HDL: Hybrid Deep Learning for the Synthesis of Myocardial Velocity Maps in Digital Twins for Cardiac Analysis

Abstract: Synthetic digital twins based on medical data accelerate the acquisition, labelling and decision making procedure in digital healthcare. A core part of digital healthcare twins is modelbased data synthesis, which permits the generation of realistic medical signals without requiring to cope with the modelling complexity of anatomical and biochemical phenomena producing them in reality. Unfortunately, algorithms for cardiac data synthesis have been so far scarcely studied in the literature. An important imaging modality in the cardiac examination is three-directional CINE multislice myocardial velocity mapping (3Dir MVM), which provides a quantitative assessment of cardiac motion in three orthogonal directions of the left ventricle. The long acquisition time and complex acquisition produce make it more urgent to produce synthetic digital twins of this imaging modality. In this study, we propose a hybrid deep learning (HDL) network, especially for synthetic 3Dir MVM data. Our algorithm is featured by a hybrid UNet and a Generative Adversarial Network with a foreground-background generation scheme. The experimental results show that from temporally down-sampled magnitude CINE images (six times), our proposed algorithm can still successfully synthesise high temporal resolution 3Dir MVM CMR data (PSNR=42.32) with precise left ventricle segmentation (DICE=0.92). These performance scores indicate that our proposed HDL algorithm can be implemented in real-world digital twins for myocardial velocity mapping data simulation. To the best of our knowledge, this work is the first one in the literature investigating digital twins of the 3Dir MVM CMR, which has shown great potential for improving the efficiency of clinical studies via synthesised cardiac data.

Association of Coexistent Hepatitis B Surface Antigen and Antibody With Severe Liver Fibrosis and Cirrhosis in Treatment-Naive Patients With Chronic Hepatitis B

Abstract: Key Points Question Is the coexistence of hepatitis B surface antigen (HBsAg) and antibody against HBsAg (anti-HBs) associated with severe liver fibrosis and cirrhosis in patients with chronic hepatitis B (CHB)? Findings In this cross-sectional study of 6534 patients with CHB, the coexistence of HBsAg and anti-HBs was independently associated with severe liver fibrosis and cirrhosis in patients with CHB, especially among those who had hepatitis B e antigen negativity. Meaning This study’s findings suggest that coexistent HBsAg and anti-HBs is a rare serological pattern that may reflect a special status of infection and that close monitoring for liver fibrosis and cirrhosis is warranted in patients with CHB who have this serological profile.

AI-based medical e-diagnosis for fast and automatic ventricular volume measurement in patients with normal pressure hydrocephalus

Abstract: Based on CT and MRI images acquired from normal pressure hydrocephalus (NPH) patients, using machine learning methods, we aim to establish a multimodal and high-performance automatic ventricle segmentation method to achieve an efficient and accurate automatic measurement of the ventricular volume. First, we extract the brain CT and MRI images of 143 definite NPH patients. Second, we manually label the ventricular volume (VV) and intracranial volume (ICV). Then, we use the machine learning method to extract features and establish automatic ventricle segmentation model. Finally, we verify the reliability of the model and achieved automatic measurement of VV and ICV. In CT images, the Dice similarity coefficient (DSC), intraclass correlation coefficient (ICC), Pearson correlation, and Bland-Altman analysis of the automatic and manual segmentation result of the VV were 0.95, 0.99, 0.99, and 4.2 ± 2.6, respectively. The results of ICV were 0.96, 0.99, 0.99, and 6.0 ± 3.8, respectively. The whole process takes 3.4 ± 0.3 s. In MRI images, the DSC, ICC, Pearson correlation, and Bland-Altman analysis of the automatic and manual segmentation result of the VV were 0.94, 0.99, 0.99, and 2.0 ± 0.6, respectively. The results of ICV were 0.93, 0.99, 0.99, and 7.9 ± 3.8, respectively. The whole process took 1.9 ± 0.1 s. We have established a multimodal and high-performance automatic ventricle segmentation method to achieve efficient and accurate automatic measurement of the ventricular volume of NPH patients. This can help clinicians quickly and accurately understand the situation of NPH patient's ventricles.

Development of a panel of three multiplex allele-specific qRT-PCR assays for quick differentiation of recombinant variants and Omicron subvariants of SARS-CoV-2

Abstract: Quick differentiation of the circulating variants and the emerging recombinant variants of SARS-CoV-2 is essential to monitor their transmission. However, the widely used gene sequencing method is time-consuming and costly when facing the viral recombinant variants, because partial or whole genome sequencing is required. Allele-specific real time RT-PCR (qRT-PCR) represents a quick and cost-effective method in SNP genotyping and has been successfully applied for SARS-CoV-2 variant screening. In the present study, we developed a panel of 3 multiplex allele-specific qRT-PCR assays targeting 12 key differential mutations for quick differentiation of SARS-CoV-2 recombinant variants (XD and XE) and Omicron subvariants (BA.1 and BA.2). Two parallel multiplex qRT-PCR reactions were designed to separately target the protype allele and the mutated allele of the four mutations in each allele-specific qRT-PCR assay. The variation of Cp values (ΔCp) between the two multiplex qRT-PCR reactions was applied for mutation determination. The developed multiplex allele-specific qRT-PCR assays exhibited outstanding analytical sensitivities (with limits of detection [LoDs] of 2.97-27.43 copies per reaction), wide linear detection ranges (107-100 copies per reaction), good amplification efficiencies (82% to 95%), good reproducibility (Coefficient of Variations (CVs) < 5% in both intra-assay and inter-assay tests) and clinical performances (99.5%-100% consistency with Sanger sequencing). The developed multiplex allele-specific qRT-PCR assays in this study provide an alternative tool for quick differentiation of SARS-CoV-2 recombinant variants (XD and XE) and Omicron subvariants (BA.1 and BA.2).

Are preoperative phase-contrast CSF flow parameters ideal for predicting the outcome of shunt surgery in patients with idiopathic normal pressure hydrocephalus?

Abstract: Purpose Phase-contrast magnetic resonance (PC-MR) is widely used in patients with idiopathic normal pressure hydrocephalus (iNPH), but its role in predicting prognosis remains controversial. To evaluate the effectiveness of preoperative PC-MR CSF flow measurement in predicting the clinical response to shunt surgery in patients with iNPH. Methods Forty-six patients with definite iNPH were included between January 2018 and January 2022. PC-MR was used to evaluate CSF peak velocity (PV), average velocity, aqueductal stroke volume (ASV), net ASV, and net flow. The modified Rankin Scale (mRS), iNPH grading scale (iNPHGS), Mini-Mental State Examination (MMSE), and Timed 3-m Up and Go Test (TUG) were used for clinical assessment. The primary endpoint was the improvement in the mRS score 1 year after surgery, and the secondary endpoints were the iNPHGS, MMSE, and TUG scores at 1 year. Differences between shunt improvement and non-improvement groups, based on the clinical outcomes, were compared using the Mann-Whitney U-test, logistic regression models, and receiver operating characteristic curves. Correlations between CSF flow parameters and the baseline clinical outcomes were assessed using Spearman's correlation coefficient. Results No CSF parameters significantly differed between shunt improvement and non-improvement groups based on mRS and secondary outcomes. And all CSF parameters showed significant overlap in both shunt improvement and non-improvement groups based on mRS and secondary outcomes. Significant correlations between the mRS and iNPHGS scores, and PV, ASV, and net ASV were observed. Conclusion While some preoperative PC-MR CSF flow parameters reflected the symptom severity of iNPH to a certain extent, they alone might not be ideal markers of shunt responsiveness.

Random Sampling and Reconstruction of Sparse Time- and Band-Limited Signals

Abstract: Abstract The random sampling and reconstruction algorithm of sparse time- and band-limited signals, which stem from bandlimited functions, is studied in this paper. For time- and band-limited signals, considering the projection on a finite-dimensional solution space, the reconstruction algorithm can therefore be described by a matrix. Especially, we study the case of the finite-dimensional spaces with regard to time- and band-limited signals. First, as the cardinality of the sample value is not less than the dimension of solution space, we prove that the coefficient matrix has full rank. Second, as sample value is insufficient, especially spare time- and band-limited signals, based on the Restricted Isometry Property (RIP) condition and the concentration inequality of a probability distribution for random sampling, we present a reconstruction from few randomly chosen samples on Finally, numerical experiments show that with high probability the time- and band-limited signals can be approximately recovered provided τ-sparsity of non-vanishing coefficients ak .

Association between serum uric acid concentration and radiographic axial spondylarthritis: a cross-sectional study of 202 patients

Abstract: Department of Rheumatology and Immunology, Shenzhen Second People’s Hospital, Shenzhen, Guangdong 510000, China; Department of Laboratory, Shenzhen Second People’s Hospital, Shenzhen, Guangdong 510000, China; Department of Neurology, Shenzhen Second People’s Hospital, Shenzhen, Guangdong 510000, China; Department of Cardiology, Shenzhen University General Hospital, Shenzhen, Guangdong 510000, China; Department of Radiology, Cancer Hospital & Shenzhen Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Shenzhen, Guangdong 510000,