IGU-Aug: Information-guided unsupervised augmentation and pixel-wise contrastive learning for medical image analysis.

Contrastive learning (CL) is a form of self-supervised learning and has been widely used for various tasks. Different from widely studied instance-level contrastive learning, pixel-wise contrastive learning mainly helps with pixel-wise dense prediction tasks. The counter-part to an instance in instance-level CL is a pixel, along with its neighboring context, in pixel-wise CL. Aiming to build better feature representation, there is a vast literature about designing instance augmentation strategies for instance-level CL; but there is little similar work on pixel augmentation for pixel-wise CL with a pixel granularity. In this paper, we attempt to bridge this gap. We first classify a pixel into three categories, namely low-, medium-, and high-informative, based on the information quantity the pixel contains. We then adaptively design separate augmentation strategies for each category in terms of augmentation intensity and sampling ratio. Extensive experiments validate that our information-guided pixel augmentation strategy succeeds in encoding more discriminative representations and surpassing other competitive approaches in unsupervised local feature matching. Furthermore, our pretrained model improves the performance of both one-shot and fully supervised models. To the best of our knowledge, we are the first to propose a pixel augmentation method with a pixel granularity for enhancing unsupervised pixel-wise contrastive learning. Code is available at https: //github.com/Curli-quan/IGU-Aug.

Analysis of postoperative recurrence-free survival in non-small cell lung cancer patients based on consensus clustering.

AIMS: This study aims to assess whether consensus clustering, based on computed tomography (CT) radiomics from both intratumoral and peritumoral regions, can effectively stratify the risk of non-small cell lung cancer (NSCLC) patients and predict their postoperative recurrence-free survival (RFS). MATERIALS AND METHODS: A retrospective analysis was conducted on the data of surgical patients diagnosed with NSCLC between December 2014 and April 2020. After preprocessing CT images, radiomic features were extracted from a 9-mm region encompassing both the tumor and its peritumoral area. Consensus clustering was utilized to analyze the radiomics features and categorize patients into distinct clusters. A comparison of the differences in clinical pathological characteristics was conducted among the clusters. Kaplan-Meier survival analysis was employed to investigate differences in survival among the clusters. RESULTS: A total of 266 patients were included in this study, and consensus clustering identified three clusters (Cluster 1: n=111, Cluster 2: n=61, Cluster 3: n=94). Multiple clinical risk factors, including pathological TNM staging, programmed cell death ligand 1 (PD-L1), and epidermal growth factor receptor (EGFR) expression status exhibit significant differences among the three clusters. Kaplan-Meier survival analysis demonstrated significant variations in RFS across the clusters (P<0.001). The 3-year cumulative recurrence-free survival rates were 76.5% (95% CI: 68.6-84.4) for Cluster 1, 45.9% (95% CI: 33.4-58.4) for Cluster 2, and 41.5% (95% CI: 31.6-51.5) for Cluster 3. CONCLUSIONS: Consensus clustering of CT radiomics based on intratumoral and peritumoral regions can stratify the risk of postoperative recurrence in patients with NSCLC.

[Clinical features of anti-neurofascin-155 antibody positive autoimmune nodopathy in 6 children].

Objective: To analyze the clinical features of anti-neurofascin-155 (NF155) antibody positive autoimmune nodopathy in children. Methods: This was a case series study. A total of 6 children who were diagnosed accurately as anti-NF155 antibody positive autoimmune nodopathy by cell immunofluorescence assay at the Children's Hospital of Fudan University from January 2020 to December 2023 were collected. This study retrospectively analyzed 6 pediatric children's clinical manifestations, laboratory and electrophysiological examination results, and treatment outcomes. Results: Among 6 children with anti-NF155 antibody positive autoimmune nodopathy, there were 4 boys and 2 girls. The onset age of 6 children ranged from 3 years and 8 months to 12 years. All 6 children had extremity weakness (more severe in the distal and the lower extremities than in the upper extremities), 5 children had sensory deficits such as numbness or pain in the extremities, 4 children had tremors and ataxia, 3 children had cranial nerve involvement. Among the 6 children, 4 children had protein-cell separation in cerebrospinal fluid examinations. Among the 6 children, 1 child had central nervous system demyelination, the brain magnetic resonance imaging showed multiple abnormal signals in the bilateral cerebral hemispheres. Four children showed motor and sensory nerve damage in electrophysiological examination, and 2 children only showed motor nerve damage. Three children showed myelin and axonal damage, and 3 children only showed axonal damage. Among the 6 children, 5 children were treated with intravenous immunoglobulin and steroids. Among them, 2 children underwent plasma exchange due to poor efficacy, and subsequently, rituximab was added. There was 1 child changed the treatment with olfatomumab since the symptoms did not significantly improve after using rituximab. After treatment for 4-15 months, 2 children had no clinical symptoms, 1 child had improvement in clinical symptoms, 2 children had no significant improvement in clinical symptoms, and 1 child who did not receive the immunotherapy had no significant change in clinical symptoms. Conclusions: Anti-NF155 antibody positive autoimmune nodopathy in children presents with varying degrees of clinical manifestations. It is mainly characterized by extremity weakness, numbness and pain, often accompanied bytremorsand ataxia. Some pediatric patients may also have central nervous system demyelination. Cerebrospinal fluid and electrophysiological examination are important auxiliary examination methods. If steroid therapy is not effective, plasma exchange and rituximab treatment should be used as soon as possible.

[The efficacy and safety of protein A immunoadsorption combined with rituximab treatment for highly sensitized patients undergoing haplo-hematopoietic stem cell transplantation].

Objective: To investigate the efficacy and safety of protein A immunoadsorption (PAIA) combined with rituximab (RTX) in highly sensitized patients who underwent haplo-hematopoietic stem cell transplantation (haplo-HSCT) . Methods: The clinical data of 56 highly sensitized patients treated with PAIA and RTX before haplo-HSCT at the First Affiliated Hospital of Soochow University and Soochow Hopes Hematonosis Hospital between March 2021 and June 2023 were retrospectively analyzed. The number of human leukocyte antigen (HLA) antibody types and the mean fluorescence intensity (MFI), humoral immunity, adverse reactions during adsorption, and survival within 100 days before and after adsorption were measured. Results: After receiving the PAIA treatment, the median MFI of patients containing only HLA Ⅰ antibodies decreased from 7 859 (3 209-12 444) to 3 719 (0-8 275) (P<0.001), and the median MFI of HLA Ⅰ+Ⅱ antibodies decreased from 5 476 (1 977-12 382) to 3 714 (0-11 074) (P=0.035). The median MFI of patients with positive anti-donor-specific antibodies decreased from 8 779 (2 697-18 659) to 4 524 (0-15 989) (P<0.001). The number of HLA-A, B, C, DR, and DQ antibodies in all patients decreased after the PAIA treatment, and the differences were statistically significant (A, B, C, DR: P<0.001, DQ: P<0.01). The humoral immune monitoring before and after the PAIA treatment showed a significant decrease in the number of IgG and complement C3 (P<0.001 and P=0.002, respectively). Forty-four patients underwent HLA antibody monitoring after transplantation, and the overall MFI and number of antibody types decreased. However, five patients developed new antibodies with low MFI, and nine patients continued to have high MFI. The overall survival, disease-free survival, non-recurrent mortality, and cumulative recurrence rates at 100 days post-transplantation were 83.8%, 80.2%, 16.1%, and 4.5%, respectively. Conclusions: The combination of PAIA and RTX has a certain therapeutic effect and good safety in the desensitization treatment of highly sensitive patients before haplo-HSCT.

Unified Multi-Modal Image Synthesis for Missing Modality Imputation.

Multi-modal medical images provide complementary soft-tissue characteristics that aid in the screening and diagnosis of diseases. However, limited scanning time, image corruption and various imaging protocols often result in incomplete multi-modal images, thus limiting the usage of multi-modal data for clinical purposes. To address this issue, in this paper, we propose a novel unified multi-modal image synthesis method for missing modality imputation. Our method overall takes a generative adversarial architecture, which aims to synthesize missing modalities from any combination of available ones with a single model. To this end, we specifically design a Commonality- and Discrepancy-Sensitive Encoder for the generator to exploit both modality-invariant and specific information contained in input modalities. The incorporation of both types of information facilitates the generation of images with consistent anatomy and realistic details of the desired distribution. Besides, we propose a Dynamic Feature Unification Module to integrate information from a varying number of available modalities, which enables the network to be robust to random missing modalities. The module performs both hard integration and soft integration, ensuring the effectiveness of feature combination while avoiding information loss. Verified on two public multi-modal magnetic resonance datasets, the proposed method is effective in handling various synthesis tasks and shows superior performance compared to previous methods.

[Analysis of tumor regression after concurrent radiochemotherapy for locally advanced hypopharyngeal carcinoma].

Objective: To investigate the pathological characteristics of tumor regression and the expression level of chemoradiotherapy resistance-related molecular markers after preoperative concurrent radiochemotherapy in patients with locally advanced hypopharyngeal carcinoma. Methods: The clinical data of 44 patients with locally advanced hypopharyngeal carcinoma who underwent preoperative concurrent radiochemotherapy in the Department of Head and Neck Surgery of Shandong Otolaryngology Hospital from August 2016 to August 2020 were retrospectively analyzed. All patients received preoperative concurrent chemotherapy and radiotherapy. After radiochemotherapy, electronic laryngoscopy and imaging examination were performed to assess the tumor regression status. After 4 weeks, surgical resection was performed, and the specimens of the primary focus were processed as continuous pathological sections. After operation, HE staining and TdT-mediated dUTP nick-end labeling (TUNEL) method were used to detect the distribution characteristics and apoptosis of the remaining cancer focus, and immunohistochemistry was performed to determine the proliferation of the remaining cancer focus and the expression of radiation resistance-related molecular markers [signal transducer and activator of transcription 3 (STAT3), hypoxia-inducible factor-1alpha (HIF-1α), sex determining region Y-box 2 (SOX2), and P53]. Results: A total of 44 patients were included, all of whom were male, with a mean age of (58.3±3.5) years. There were 40 cases of pyriform sinus carcinoma and 4 cases of posterior pharyngeal wall carcinoma. Twenty-nine cases were in stage T3 and 15 cases were in stage T4. There were 6 stage Ⅲ cases and 38 stage Ⅳ cases. According to the response evaluation criteria in solid tumors (RECIST), 13 patients achieved complete response (CR), 22 patients had partial response (PR), and 9 patients achieved stable disease (SD) after concurrent radiochemotherapy. The primary lesion resection methods included 19 cases of hypopharyngeal circumferential resection and 2 cases of total laryngectomy and partial hypopharyngeal resection. Twenty-three cases underwent supracricoid cartilage subtotal laryngectomy cricoid tongue fixation (CHP). Among 22 patients with PR, 10 had large PR (remission rate ≥70%) and 12 had small PR (remission rate <70%). The residual tumor was found in 30 patients (68.2%) after resection of all primary lesions by HE staining of pathological sections, of which 3 patients (3/13) with CR had residual cancer, all of which were focal residues. In large PR patients, residual cancer was detected in 6 cases (6/10), scattered in 4 cases, and focal residual in 2 cases, respectively. Large residual tumors were detected in small PR and SD patients. TUNEL method did not show any sign of apoptosis in 30 specimens with residual cancer focus, and the positive expression rate of Ki-67 was less than 10%. The expression of STAT3 (3.40±2.49 vs 5.23±3.02, t=-2.932, P=0.007) in 19 cases (63.3%) and HIF-1α (3.73±2.66 vs 6.97±3.05, t=-4.45, P<0.001) in 22 cases (73.3%) of residual cancer were significantly higher than those before radiochemotherapy. Other molecular markers showed no significant changes. All patients were followed up for 3 years. The 2-year survival rate was 59.3%, and the 3-year survival rate was 54.1%. Conclusions: Preoperative radiochemotherapy can make some patients with locally advanced hypopharyngeal carcinoma achieve complete or significant remission in clinical evaluation, but pathological detection still shows some residual cancer lesions with enhanced anti-apoptosis ability and decreased proliferation activity.

Individual and area-level socioeconomic status, Life's Simple 7, and comorbid cardiovascular disease and cancer: a prospective analysis of the UK Biobank cohort.

OBJECTIVES: We aimed to investigate the associations of individual and area-level socioeconomic status (SES) with incident cardiovascular diseases (CVD) alone, cancer alone, and comorbid CVD and cancer, and the mediation role of cardiovascular health score in these associations. STUDY DESIGN: This was a population-based prospective cohort study. METHODS: We used data from the UK Biobank, a population-based prospective cohort study. Latent class analysis was used to create an individual-level SES index based on three indicators (household income, education level, and employment status), and the Townsend Index was defined as the area-level socioeconomic status. We used the American Heart Association's (AHA) Life's Simple 7 (smoking, body weight, physical activity, diet, blood pressure, blood glucose, and total cholesterol) to calculate the cardiovascular health score. We used Cox proportional hazard regression models to estimate the hazard ratio (HR) and 95% confidence interval (CI) adjusted for demographic, environmental, and genetic factors. RESULTS: Compared with high SES, the HRs in participants with low individual and area-level SES were 1.33 (95% confidence interval [CI] 1.29 to 1.38) and 1.24 (95% CI 1.20 to 1.29) for incident CVD, 0.96 (95% CI 0.93 to 0.99) and 0.95 (95%CI 0.92 to 0.98) for incident cancer, 1.32 (95%CI 1.24 to 1.40) and 1.15 (95%CI 1.08 to 1.22) for incident comorbid CVD and cancer, respectively. Additionally, the mediation proportion of CVD score for individual and area-level SES was 47.93% and 48.87% for incident CVD, 44.83% and 59.93% for incident comorbid CVD and cancer. The interactions between individual-level SES and CVD scores were significant on incident CVD, and comorbid CVD and cancer, and the protective associations were stronger in participants with high individual-level SES. CONCLUSIONS: Life's Simple 7 significantly mediated the associations between SES and comorbid CVD and cancer, while almost half of the associations remained unclear.

[Single-cell transcriptomic sequencing coupled with Mendelian randomization analysis elucidates the pivotal role of CTSC in chronic rhinosinusitis].

Objective: To investigate the molecular mechanisms of chronic rhinosinusitis (CRS), to identify key cell subgroups and genes, to construct effective diagnostic models, and to screen for potential therapeutic drugs. Methods: Key cell subgroups in CRS were identified through single-cell transcriptomic sequencing data. Essential genes associated with CRS were selected and diagnostic models were constructed by hdWGCNA (high dimensional weighted gene co-expression network analysis) and various machine learning algorithms. Causal inference analysis was performed using Mendelian randomization and colocalization analysis. Potential therapeutic drugs were identified using molecular docking technology, and the results of bioinformatics analysis were validated by immunofluorescence staining. Graphpad Prism, R, Python, and Adobe Illustrator software were used for data and image processing. Results: An increased proportion of basal and suprabasal cells was observed in CRS, especially in eosinophilic CRS with nasal polyps (ECRSwNP), with P=0.001. hdWGCNA revealed that the "yellow module" was closely related to basal and suprabasal cells in CRS. Univariate logistic regression and LASSO algorithm selected 13 key genes (CTSC, LAMB3, CYP2S1, TRPV4, ARHGAP21, PTHLH, CDH26, MRPS6, TENM4, FAM110C, NCKAP5, SAMD3, and PTCHD4). Based on these 13 genes, an effective CRS diagnostic model was developed using various machine learning algorithms (AUC=0.958). Mendelian randomization analysis indicated a causal relationship between CTSC and CRS (inverse variance weighted: OR=1.06, P=0.006), and colocalization analysis confirmed shared genetic variants between CTSC and CRS (PPH4/PPH3>2). Molecular docking results showed that acetaminophen binded well with CTSC (binding energy:-5.638 kcal/mol). Immunofluorescence staining experiments indicated an increase in CTSC+cells in CRS. Conclusion: This study integrates various bioinformatics methods to identify key cell types and genes in CRS, constructs an effective diagnostic model, underscores the critical role of the CTSC gene in CRS pathogenesis, and provides new targets for the treatment of CRS.

Effects of feed conditioning temperature on pellet quality, growth performance, intestinal development and blood parameters of geese from 1 to 28 d of age.

A 28-d experiment was conducted to investigate the effects of feed-conditioning temperature on the pellet quality, growth performance, intestinal development, and blood parameters of geese. A total of 180 one-day-old White Yuzhou goslings were randomly allotted to 5 treatment groups, with 6 replicates containing 6 birds each. Five diets were conditioned at 65, 70, 75, 80, and 85°C. Body weight and feed intake per pen basis were recorded from the arrival to the end of the trial. Blood and small intestine samples were collected on d 28 for analysis. The results showed that the pellet durability index (PDI), pellet hardness, and gelatinisation degree of starch (GDS) increased with increasing conditioning temperature (P < 0.05). The final body weight (FBW), average daily gain (ADG) and average daily feed intake (ADFI) of goslings significantly increased when conditioning temperature increased from 65 or 70°C to 80 or 85°C (P < 0.05), accompanied by unaffected feed conversion ratio (FCR) (P > 0.05). The villus height to crypt depth ratio (VH/CD) in the duodenum and ileum improved with increasing conditioning temperature (P < 0.05). Additionally, trypsin and amylase activity were enhanced when the conditioning temperature increased from 65 to 85°C (P < 0.05). No significant differences in the carcass traits and blood parameters of goslings were observed among the groups (P > 0.05). Overall, under the present experimental conditions, increasing the steam-conditioning temperature of pelleted feed improved pellet quality, growth performance, intestinal morphology, and digestive enzyme activity in goslings. Based on broken-line regression analysis, the lower critical conditioning temperature for ADG in geese from 1 to 28 d of age was 80.95°C.

[Epidemiological characteristics of leptospirosis in China from 2010 to 2022].

OBJECTIVE: To analyze the epidemiological characteristics of leptospirosis in China from 2010 to 2022, so as to provide insights into formulation of the leptospirosis control strategy. METHODS: All data pertaining to clinically diagnosed cases and confirmed cases of leptospirosis reported in China from January 1, 2010 to December 31, 2022 was collected from Chinese Disease Prevention and Control Information Management System. The spatial, temporal and population distributions, and report and diagnosis institutions of leptospirosis cases were analyzed using a descriptive epidemiological method. RESULTS: A total of 4 559 leptospirosis cases were reported in China from 2010 to 2022, with an annual average number of 351 cases, and the number of reported leptospirosis cases reduced from 679 cases in 2010 to 158 cases in 2018. A total of 4 276 leptospirosis cases were reported in Sichuan Province, Yunnan Province, Guangdong Province, Hunan Province, Fujian Province, Zhejiang Province, Guangxi Zhuang Autonomous Region, Anhui Province, Jiangxi Province and Guizhou Province, accounting for 93.79% of the total number of leptospirosis cases in China. The number of leptospirosis cases had recently appeared a remarkable decline in Yunnan Province, while a significant rise was seen in the number of leptospirosis cases in two provinces of Zhejiang and Guangdong. No leptospirosis cases were reported in Henan Province from 2010 to 2020; however, there were 5 cases and 2 cases reported in 2021 and 2022, respectively. There was only one leptospirosis case reported in Shaanxi Province from 2010 to 2017; however, leptospirosis cases were reported in the province for 5 consecutive years since 2018. Leptospirosis cases were reported throughout the year in China from 2010 to 2022, with the peak of incidence found during the period between August and October, and the peak of leptospirosis incidence varied in provinces. A higher number of leptospirosis cases was seen among men than among women, with a male to female ratio of 2.3:1, and the median age of leptospirosis cases was 50 years (interquartile range, 23 years), with the highest proportion of leptospirosis cases reported at ages of 51 to 60 years (23.21%). Among all reported leptospirosis cases, 53.28% were confirmed cases, and the proportion of confirmed cases increased from 35.05% in 2010 to 61.66% in 2022. In addition, there were 67.22% of leptospirosis cases (2 937 cases) reported by comprehensive hospitals, 20.44% (893 cases) by disease control and prevention institutions, 7.23% (316 cases) by grassroots healthcare institutions and 5.10% (223 cases) by other healthcare and medical institutions, and the mortality of reported leptospirosis cases was 1.07% in China from 2010 to 2022, with a higher mortality seen among men than among women (1.39% vs. 0.36%; χ2 = 9.52, P = 0.002). CONCLUSIONS: The incidence of leptospirosis remained at a low level in China from 2010 to 2022, and southern China was still the main endemic area for leptospirosis. The epidemiological characteristics of leptospirosis cases varied in endemic provinces, and leptospirosis cases had been continued to be reported in Shaanxi and Henan provinces, which should be paid much attention to. Intensified surveillance of leptospirosis, improved diagnosis and treatment capability of leptospirosis cases and leptospirosis control with adaptations to local circumstance are recommended.

Prediction of programmed death-1 expression status in non-small cell lung cancer based on intratumoural and peritumoral computed tomography (CT) radiomics nomogram.

AIMS: This study aimed to predict the expression of programmed death-1 (PD-1) in non-small cell lung cancer (NSCLC) using intratumoral and peritumoral computed tomography (CT) radiomics nomogram. MATERIALS AND METHODS: Two hundred patients pathologically diagnosed with NSCLC from two hospitals were retrospectively analyzed. Of these, 159 NSCLC patients from our hospital were randomly divided into a training cohort (n=96) and an internal validation cohort (n=63) at a ratio of 6:4, while 41 NSCLC patients from another medical institution served as the external validation cohort. The radiomic features of the gross tumor volume (GTV) and peritumoral volume (PTV) were extracted from the CT images. Optimal radiomics features were selected using least absolute shrinkage and selection operator regression analysis. Finally, a CT radiomics nomogram of clinically independent predictors combined with the best rad-score was constructed. RESULTS: Compared with the 'GTV' and 'PTV' radiomics models, the combined 'GTV + PTV' radiomics model showed better predictive performance, and its area under the curve (AUC) values in the training, internal validation, and external validation cohorts were 0.90 (95% confidence interval [CI]: 0.83-0.97), 0.85 (95% CI: 0.74-0.96) and 0.78 (95% CI: 0.63-0.92). The nomogram constructed by the rad-score of the 'GTV + PTV' radiomics model combined with clinical independent predictors (prealbumin and monocyte) had the best performance, with AUC values in each cohort being 0.92 (95% CI: 0.85-0.98), 0.88 (95% CI: 0.78-0.97), and 0.80 (95% CI: 0.66-0.94), respectively. CONCLUSION: The intratumoral and peritumoral CT radiomics nomogram may facilitate individualized prediction of PD-1 expression status in patients with NSCLC.

Retraction Note: BNIP1 inhibits cell proliferation, migration and invasion, and promotes apoptosis by mTOR in cervical cancer cells.

The article "BNIP1 inhibits cell proliferation, migration and invasion, and promotes apoptosis by mTOR in cervical cancer cells", by F.-H. Li, L. Xiang, L. Ran, S. Zhou, Z. Huang, M. Chen, W.-F. Yu, published in Eur Rev Med Pharmacol Sci 2019; 23 (4): 1397-1407-DOI: 10.26355/eurrev_201902_17096-PMID: 30840260 has been retracted by the Editor in Chief for the following reasons. Following some concerns raised on PubPeer regarding a possible overlap in Figure 2A, the Editor in Chief has started an investigation to assess the validity of the results as well as possible figure manipulation. The journal investigation revealed a duplication in Figure 2A between BNIP1 panels, migration and invasion, respectively and in Control and invasion panels. Consequently, the Editor in Chief mistrusts the results presented and has decided to withdraw the article. The authors have been informed about the journal's investigation but remained unresponsive. https://www.europeanreview.org/article/17096 This article has been retracted. The Publisher apologizes for any inconvenience this may cause.

Association of serum 25-hydroxyvitamin D with cardiovascular mortality and kidney outcome in patients with early stages of CKD.

PURPOSE: While serum 25-hydroxyvitamin D (25[OH]D) deficiency is prevalent in chronic kidney disease (CKD), the effects of 25(OH)D deficiency on cardiovascular mortality and kidney outcomes in patients with early-stage CKD remain incompletely understood. METHODS: This multicenter retrospective cohort study included adult patients with stages 1-3 CKD from 19 medical centers across China between January 2000 and May 2021. The primary outcome was cardiovascular mortality. The secondary study outcome included CKD progression (defined as a sustained > 40% eGFR decrease from baseline or progress to end-stage kidney disease), and annual percentage change of eGFR. RESULTS: Of 9229 adults with stages 1-3 CKD, 27.0% and 38.9% had severe (< 10 ng/mL) and moderate (10 to < 20 ng/mL) serum 25(OH)D deficiency, respectively. Compared with patients having 25(OH)D ≥ 20 ng/mL, a significantly higher risk of cardiovascular mortality (hazard ratio [HR] 1.90, 95% CI 1.37-2.63), CKD progression (HR 2.20, 95% CI 1.68-2.88), and a steeper annual decline in eGFR (estimate - 7.87%; 95% CI - 10.24% to - 5.51% per year) was found in those with serum 25(OH)D < 10 ng/mL. Similar results were obtained in subgroups and by sensitivity analyses. CONCLUSIONS: 25(OH)D deficiency is associated with increased risks of cardiovascular mortality and CKD progression in patients with early-stage CKD. Studies are needed to determine whether early intervention for 25(OH)D deficiency could improve the prognosis of patients with early-stage CKD.

Which images to label for few-shot medical image analysis?

The success of deep learning methodologies hinges upon the availability of meticulously labeled extensive datasets. However, when dealing with medical images, the annotation process for such abundant training data often necessitates the involvement of experienced radiologists, thereby consuming their limited time resources. In order to alleviate this burden, few-shot learning approaches have been developed, which manage to achieve competitive performance levels with only several labeled images. Nevertheless, a crucial yet previously overlooked problem in few-shot learning is about the selection of template images for annotation before learning, which affects the final performance. In this study, we propose a novel TEmplate Choosing Policy (TECP) that aims to identify and select "the most worthy" images for annotation, particularly within the context of multiple few-shot medical tasks, including landmark detection, anatomy detection, and anatomy segmentation. TECP is composed of four integral components: (1) Self-supervised training, which entails training a pre-existing deep model to extract salient features from radiological images; (2) Alternative proposals for localizing informative regions within the images; and (3) Representative Score Estimation, which involves the evaluation and identification of the most representative samples or templates. (4) Ranking, which rank all candidates and select one with highest representative score. The efficacy of the TECP approach is demonstrated through a series of comprehensive experiments conducted on multiple public datasets. Across all three medical tasks, the utilization of TECP yields noticeable improvements in model performance.

[Exploring the causality between intestinal flora and hyperplastic scars of human based on two-sample Mendelian randomization analysis].

Objective: To investigate the causality between intestinal flora and hypertrophic scars (HS) of human. Methods: This study was a study based on two-sample Mendelian randomization (TSMR) analysis. The data on intestinal flora (n=18 473) and HS (n=208 248) of human were obtained from the genome-wide association study database. Genetically variable genes at five levels (phylum, class, order, family, and genus) of known intestinal flora, i.e., single nucleotide polymorphisms (SNPs), were extracted as instrumental variables for linkage disequilibrium (LD) analysis. Human genotype-phenotype association analysis was performed using PhenoScanner V2 database to exclude SNPs unrelated to HS in intestinal flora and analyze whether the selected SNPs were weak instrumental variables. The causal relationship between intestinal flora SNPs and HS was analyzed through four methods of TSMR analysis, namely inverse variance weighted (IVW), MR-Egger regression, weighted median, and weighted mode. Scatter plots of significant results from the four aforementioned analysis methods were plotted to analyze the correlation between intestinal flora SNPs and HS. Both IVW test and MR-Egger regression test were used to assess the heterogeneity of intestinal flora SNPs, MR-Egger regression test and MR-PRESSO outlier test were used to assess the horizontal multiplicity of intestinal flora SNPs, and leave-one-out sensitivity analysis was used to determine whether HS was caused by a single SNP in the intestinal flora. Reverse TSMR analyses were performed for HS SNPs and genus Intestinimonas or genus Ruminococcus2, respectively, to detect whether there was reverse causality between them. Results: A total of 196 known intestinal flora, belonging to 9 phyla, 16 classes, 20 orders, 32 families, and 119 genera, were obtained, and multiple SNPs were obtained from each flora as instrumental variables. LD analysis showed that the SNPs of the intestinal flora were consistent with the hypothesis that genetic variation was strongly associated with exposure factors, except for rs1000888, rs12566247, and rs994794. Human genotype-phenotype association analysis showed that none of the selected SNPs after LD analysis was excluded and there were no weak instrumental variables. IVW, MR-Egger regression, weighted median, and weighted mode of TSMR analysis showed that both genus Intestinimonas and genus Ruminococcus2 were causally associated with HS. Among them, forest plots of IVW and MR-Egger regression analyses also showed that 16 SNPs (the same SNPs number of this genus below) of genus Intestinimonas and 15 SNPs (the same SNPs number of this genus below) of genus Ruminococcus2 were protective factors for HS. Further, IVW analysis showed that genus Intestinimonas SNPs (with odds ratio of 0.62, 95% confidence interval of 0.41-0.93, P<0.05) and genus Ruminococcus2 SNPs (with odds ratio of 0.62, 95% confidence interval of 0.40-0.97, P<0.05) were negatively correlated with the risk of HS. Scatter plots showed that SNPs of genus Intestinimonas and genus Ruminococcus2 were protective factors of HS. Both IVW test and MR-Egger regression test showed that SNPs of genus Intestinimonas (with Q values of 5.73 and 5.76, respectively, P>0.05) and genus Ruminococcus2 (with Q values of 13.67 and 15.61, respectively, P>0.05) were not heterogeneous. MR-Egger regression test showed that the SNPs of genus Intestinimonas and genus Ruminococcus2 had no horizontal multiplicity (with intercepts of 0.01 and 0.06, respectively, P>0.05); MR-PRESSO outlier test showed that the SNPs of genus Intestinimonas and genus Ruminococcus2 had no horizontal multiplicity (P>0.05). Leave-one-out sensitivity analysis showed that no single intestinal flora SNP drove the occurrence of HS. Reverse TSMR analysis showed no reverse causality between HS SNPs and genus Intestinimonas or genus Ruminococcus2 (with odds ratios of 1.01 and 0.99, respectively, 95% confidence intervals of 0.97-1.06 and 0.96-1.04, respectively, P>0.05). Conclusions: There is a causal relationship between intestinal flora and HS of human, in which genus Intestinimonas and genus Ruminococcus2 have a certain effect on inhibiting HS.

Learning to Sketch: A Neural Approach to Item Frequency Estimation in Streaming Data.

Recently, there has been a trend of designing neural data structures to go beyond handcrafted data structures by leveraging patterns of data distributions for better accuracy and adaptivity. Sketches are widely used data structures in real-time web analysis, network monitoring, and self-driving to estimate item frequencies of data streams within limited space. However, existing sketches have not fully exploited the patterns of the data stream distributions, making it challenging to tightly couple them with neural networks that excel at memorizing pattern information. Starting from the premise, we envision a pure neural data structure as a base sketch, which we term the meta-sketch, to reinvent the base structure of conventional sketches. The meta-sketch learns basic sketching abilities from meta-tasks constituted with synthetic datasets following Zipf distributions in the pre-training phase and can be quickly adapted to real (skewed) distributions in the adaption phase. The meta-sketch not only surpasses its competitors in sketching conventional data streams but also holds good potential in supporting more complex streaming data, such as multimedia and graph stream scenarios. Extensive experiments demonstrate the superiority of the meta-sketch and offer insights into its working mechanism.