DualAttNet: Synergistic fusion of image-level and fine-grained disease attention for multi-label lesion detection in chest X-rays.

Chest radiographs are the most commonly performed radiological examinations for lesion detection. Recent advances in deep learning have led to encouraging results in various thoracic disease detection tasks. Particularly, the architecture with feature pyramid network performs the ability to recognise targets with different sizes. However, such networks are difficult to focus on lesion regions in chest X-rays due to their high resemblance in vision. In this paper, we propose a dual attention supervised module for multi-label lesion detection in chest radiographs, named DualAttNet. It efficiently fuses global and local lesion classification information based on an image-level attention block and a fine-grained disease attention algorithm. A binary cross entropy loss function is used to calculate the difference between the attention map and ground truth at image level. The generated gradient flow is leveraged to refine pyramid representations and highlight lesion-related features. We evaluate the proposed model on VinDr-CXR, ChestX-ray8 and COVID-19 datasets. The experimental results show that DualAttNet surpasses baselines by 0.6% to 2.7% mAP and 1.4% to 4.7% AP50 with different detection architectures. The code for our work and more technical details can be found at https://github.com/xq141839/DualAttNet.

ENPP2 inhibitor improves proliferation in AOM/DSS-induced colorectal cancer mice via remodeling the gut barrier function and gut microbiota composition.

In our previous multicenter study, we delineated the inherent metabolic features of colorectal cancer (CRC). Therein, we identified a member of the ectonucleotide pyrophosphatase/ phosphodiesterase family (ENPP2) as a significant differential metabolite of CRC. In this study, the role of ENPP2 in CRC has been demonstrated using established in vitro and in vivo models including ENPP2 gene knockdown, and use of the ENPP2 inhibitor, GLPG1690. We found that CRC proliferation was decreased after either ENPP2 gene knockdown or use of ENPP2 inhibitors. We further evaluated the role of GLPG1690 in AOM/DSS-induced CRC mice via intestinal barrier function, macrophage polarization, inflammatory response and microbial homeostasis. Results of immunofluorescence staining and Western blotting showed that GLPG1690 can restore gut-barrier function by increasing the expression of tight junction proteins, claudin-1, occludin and ZO-1. M2 tumor-associated macrophage polarization and colonic inflammation were attenuated after treatment with GLPG1690 using the Azoxymethane/Dextran Sodium Sulfate (AOM/DSS) model. Moreover, 16 S rDNA pyrosequencing and metagenomic analysis showed that GLPG1690 could alleviate gut dysbiosis in mice. Furthermore, administration of GLPG1690 with antibiotics as well as fecal microbiota transplantation assays demonstrated a close link between the efficacy of GLPG1690 and the gut microbiota composition. Finally, results of metabolomic analysis implicated mainly the gut microbiota-derived metabolites of aromatic amino acids in CRC progression. These findings may provide novel insights into the development of small-molecule ENPP2 inhibitors for the treatment of CRC.

DCSAU-Net: A deeper and more compact split-attention U-Net for medical image segmentation.

Deep learning architecture with convolutional neural network achieves outstanding success in the field of computer vision. Where U-Net has made a great breakthrough in biomedical image segmentation and has been widely applied in a wide range of practical scenarios. However, the equal design of every downsampling layer in the encoder part and simply stacked convolutions do not allow U-Net to extract sufficient information of features from different depths. The increasing complexity of medical images brings new challenges to the existing methods. In this paper, we propose a deeper and more compact split-attention u-shape network, which efficiently utilises low-level and high-level semantic information based on two frameworks: primary feature conservation and compact split-attention block. We evaluate the proposed model on CVC-ClinicDB, 2018 Data Science Bowl, ISIC-2018, SegPC-2021 and BraTS-2021 datasets. As a result, our proposed model displays better performance than other state-of-the-art methods in terms of the mean intersection over union and dice coefficient. More significantly, the proposed model demonstrates excellent segmentation performance on challenging images. The code for our work and more technical details can be found at https://github.com/xq141839/DCSAU-Net.

Metabolomics Study Revealing the Potential Risk and Predictive Value of Fragmented QRS for Acute Myocardial Infarction.

Patients with nonobstructive coronary artery disease (NOCAD) have high risk associated with acute myocardial infarction (AMI), and fragmented QRS (fQRS) has a predictive value of AMI after percutaneous coronary intervention (PCI). A cohort of 254 participants were recruited including 136 NOCAD and 118 AMI patients from Xi'an No. 1 Hospital. Comprehensive metabolomics was performed by UPLC-Q/TOF-MS with multivariate statistical analyses. Hazard ratios were measured to discriminate the prognostic in AMI after PCI between differential metabolites and fQRS. OPLS-DA separated metabolites from NOCAD and AMI in serum. A total of 23 differential metabolites were identified between NOCAD and AMI. In addition, four differential metabolites, namely, acetylglycine, threoninyl-glycine, glutarylglycine, and nonanoylcarnitine, were identified between fQRS and non-fQRS in AMI. The hazard ratios demonstrate that the metabolites were associated with the risk of cardiac death, recurrent angina, readmissions, and major adverse cardiovascular events, which may clarify the mechanism of fQRS as a predictor in the prognostic of AMI after PCI. This study identified novel differential metabolites to distinguish the difference from NOCAD to AMI and clarify the mechanism of fQRS in prognostic of AMI after PCI, which may provide novel insights into potential risks and prognostic of AMI.

Metabolomics Study on the Effects of Salvianolic Acid B and Borneol for Treating Cerebral Ischemia in Rats by Ultra-Performance Liquid Chromatography Quadrupole Time-of-Flight Mass Spectrometry.

Salviae miltiorrliza-borneol Jun-Shi coupled-herbs have been widely used for treatment of ischemia stroke. Salvianolic acid B was the most abundant and bioactive compound of Salviae miltiorrliza and used for prevention and treatment of cerebrovascular diseases. However, the scientific intension and compatible mechanism of Salvianolic acid B - borneol combination were still unknown. A metabolomics study approach based on ultra-performance liquid chromatography quadrupole time-of-flight mass spectrometry (UPLC-Q/TOF-MS) combined with a pathological study has been applied to study the metabolic disturbances of cerebral ischemia and evaluate the efficacies of Sal B and Sal B/borneol against cerebral ischemia in middle cerebral artery occlusion (MCAO) rats. The neuroprotection of Sal B and Sal B/borneol was reversed through the evaluation of neurological deficits, infarct volume, and neuronal apoptosis in MCAO model. The metabonomic analysis revealed that the MCAO-induced cerebral ischemia could be ameliorated by Sal B through improving the energy metabolism, lipids metabolism, inflammatory responses, and oxidant stress. Borneol could enhance the neuroprotective effects, was associated with the increased concentration of Sal B, and attenuate the function of sphingolipid metabolism pathway in cerebral ischemia rats. These findings perhaps clarify the mechanism of neuroprotective effects of treating ischemia stroke by Sal B or Sal B/borneol preliminarily through metabolomics and push the quality promotion and the composition of borneol/Sal B in secondary development of prescription.

A supervised texton based approach for automatic segmentation and measurement of the fetal head and femur in 2D ultrasound images.

This paper presents a supervised texton based approach for the accurate segmentation and measurement of ultrasound fetal head (BPD, OFD, HC) and femur (FL). The method consists of several steps. First, a non-linear diffusion technique is utilized to reduce the speckle noise. Then, based on the assumption that cross sectional intensity profiles of skull and femur can be approximated by Gaussian-like curves, a multi-scale and multi-orientation filter bank is designed to extract texton features specific to ultrasound fetal anatomic structure. The extracted texton cues, together with multi-scale local brightness, are then built into a unified framework for boundary detection of ultrasound fetal head and femur. Finally, for fetal head, a direct least square ellipse fitting method is used to construct a closed head contour, whilst, for fetal femur a closed contour is produced by connecting the detected femur boundaries. The presented method is demonstrated to be promising for clinical applications. Overall the evaluation results of fetal head segmentation and measurement from our method are comparable with the inter-observer difference of experts, with the best average precision of 96.85%, the maximum symmetric contour distance (MSD) of 1.46 mm, average symmetric contour distance (ASD) of 0.53 mm; while for fetal femur, the overall performance of our method is better than the inter-observer difference of experts, with the average precision of 84.37%, MSD of 2.72 mm and ASD of 0.31 mm.

Fragmented QRS complexes have predictive value of imperfect ST-segment resolution in patients with STEMI after primary percutaneous coronary intervention.

OBJECTIVES: To investigate the clinical characteristics of patients with the fragmented QRS complexes (fQRS) and the predictive value of fQRS in patients undergoing primary percutaneous coronary intervention (p-PCI). METHODS: The study enrolled 227 consecutive patients with ST-elevation myocardial infarction who underwent p-PCI. Baseline clinical characteristics, the percentage of ST-segment resolution, and parameters of electrocardiography and coronary angiography were investigated. The relationship between fQRS on pre-PCI and post-PCI electrocardiogram and the percentage of ST-segment resolution after PCI were studied. RESULTS: Patients with fQRS have higher troponin I, creatine kinase-MB levels, prolonged QRS duration, higher Gensini score, lower percentage of total ST-segment resolution, and left ventricular ejection fraction compared with patients without fQRS. Gensini score (odds ratio [OR], 1.013; 95% confidence interval [CI], 1.002-1.024; P < .006) and percentage of total ST-segment resolution (OR, 0.384; 95% CI, 0.186-0.795; P = .01) were independently associated with the presence of fQRS. A multivariate logistic regression analysis selected presence of fQRS pre-PCI (OR, 2.908; 95% CI, 1.095-7.723; P = .032) and the number of leads with fQRS before PCI (OR, 1.582; 95% CI, 1.250-2.002; P < .001) as independent predictors of imperfect ST-segment resolution. CONCLUSIONS: The presence of fQRS is a predictor in ST-elevation myocardial infarction patients undergoing p-PCI. The occurrence of fQRS is beneficial to identify the patients with severe coronary lesion, left ventricular contraction dysfunction, and larger areas of ischemic injury.

[Association between fragmented QRS complexes and imperfect ST-segment resolution in patients with ST-elevation myocardial infarction after primary percutaneous coronary intervention].

OBJECTIVE: To explore the relationship between fragmented QRS complexes (fQRS) and imperfect ST-segment resolution in ST elevation myocardial infarction (STEMI) patients undergoing primary percutaneous coronary intervention (p-PCI). METHODS: This study included 227 consecutive patients with STEMI who underwent p-PCI. They were divided into two groups: ECG with fQRS (n = 142) and without fQRS (n = 85). Baseline clinical characteristics,Gensini score, coronary angiography features and the rate of ST-segment resolution were compared between the two groups. RESULTS: (1) Patients with fQRS of ECG had higher cTnI, CK, CK-MB levels and Gensini score, prolonged QRS interval, lower rate of ST-segment resolution and left ventricular ejection fraction (LVEF) than in patients without fQRS (all P < 0.01 or P < 0.05). (2) Pearson correlation analysis showed that the rate of ST-segment resolution (r = -0.207, P = 0.002),Gensini score (r = 0.191, P = 0.004), LVEF(r = -0.188, P = 0.006), cTnI(r = 0.172, P = 0.010), and the TIMI grade post p-PCI (r = -0.148, P = 0.028) were significantly related with the presence of fQRS. (3) Multivariate logistic regression analysis demonstrated that presence of fQRS at pre-PCI (OR = 2.908, 95%CI:1.095-7.723, P = 0.032) , the number of leads with fQRS before PCI (OR = 1.582, 95%CI:1.250-2.002, P < 0.001), and increased QRS interval (OR = 0.955, 95%CI: 0.924-0.988, P = 0.008) were independent predictors of imperfect ST-segment resolution. CONCLUSIONS: fQRS is related to imperfect ST-segment resolution in STEMI patients undergoing p-PCI.fQRS may be a useful parameter to identify the patients with severe coronary lesion, larger areas of ischemic injury and myocardial infarction as well as severe left ventricular contracted dysfunction.