Delving Into Crispness: Guided Label Refinement for Crisp Edge Detection.

Learning-based edge detection usually suffers from predicting thick edges. Through extensive quantitative study with a new edge crispness measure, we find that noisy human-labeled edges are the main cause of thick predictions. Based on this observation, we advocate that more attention should be paid on label quality than on model design to achieve crisp edge detection. To this end, we propose an effective Canny-guided refinement of human-labeled edges whose result can be used to train crisp edge detectors. Essentially, it seeks for a subset of over-detected Canny edges that best align human labels. We show that several existing edge detectors can be turned into a crisp edge detector through training on our refined edge maps. Experiments demonstrate that deep models trained with refined edges achieve significant performance boost of crispness from 17.4% to 30.6%. With the PiDiNet backbone, our method improves ODS and OIS by 12.2% and 12.6% on the Multicue dataset, respectively, without relying on non-maximal suppression. We further conduct experiments and show the superiority of our crisp edge detection for optical flow estimation and image segmentation.

Evaluation of the Diagnostic Performance of mNGS in Detecting Intra-Abdominal Infections of the Emergency Department Patients.

Purpose: Intra-abdominal infections (IAI) are gradually becoming common in the emergency department, though the incidence is low and the prognosis is fair, as the symptoms are similar to other intra-abdominal diseases, rapid and accurate diagnosis of the causative agents is essential for clinical management. This study aimed to evaluate the diagnostic performance of metagenomic next-generation sequencing (mNGS) in detecting IAI in the emergency department. Patients and Methods: This was a retrospective, single-centered study including patients admitted to the emergency department from January 1st, 2021 to August 31st, 2022 with diagnosis of IAI. The comparison between mNGS and microbial culture using paracentesis fluid samples was performed to evaluate the diagnostic performance of mNGS for IAI. Meanwhile, paracentesis fluid and peripheral blood mNGS were compared to explore the sample specificity. Further, the microbial community structure of the patients with pyogenic liver abscesses (PLA) was analyzed. Results: Thirty-four IAI patients including 23 with pyogenic liver abscesses (PLA), 3 with parapancreatic abscesses, and 8 with other IAI were included in this study. Compared with the conventional microbial culture of paracentesis fluid, mNGS using paracentesis fluid detected more positive cases of IAI (93.75% vs 81.25%), and identified more species of pathogens, especially in obligate anaerobes and viral pathogens. Peripheral blood mNGS presented a relatively high consistency with the paracentesis fluid mNGS (91% mutual positive). The microbial community structure of PLA patients with diabetes is less diverse than that of those without diabetes. Patients with diabetes are at high risk of PLA caused by Klebsiella pneumonia. Conclusion: mNGS has advantages in detecting IAI in the emergency department, and peripheral blood mNGS can be a non-invasive choice for early diagnosis.

MIP-T3 Expression Associated with Defects of Ciliogenesis in Airway of COPD Patients.

Introduction. Some studies have found that cilia were shorter in COPD smokers than in nonsmokers or healthy smokers. However, the structural abnormalities of cilia and the cause of such abnormalities in COPD patients still remain unknown. Tumor necrosis factor alpha receptor 3 interacting protein 1 (MIP-T3) may play an important role in the progress of ciliary protein transporting. Objectives: This study aimed at exploring the dominated structural abnormalities of cilia and the involvement of MIP-T3 in the pathogenesis of cilia of COPD patients. Methods: Patients who accepted pulmonary lobectomy were divided into 3 groups: the chronic obstructive pulmonary disease (COPD) smoker group, the healthy smoker group, and the nonsmoker group, according to smoking history and pulmonary function. The ultrastructure of cilia and the percentage of abnormal cilia were analyzed using a transmission electron microscope. Real-time PCR, immunohistochemical staining, and western blotting in bronchial epithelium were used to determine MIP-T3 mRNA and protein expression. The relationship between the percentage of abnormal cilia and lung function and MIP-T3 protein expression was analyzed. Results: Patients in the COPD smoker group had increased percentage of abnormal cilia comparing to both the healthy smoker group and the nonsmoker group (both P values <0.05). MIP-T3 expression was significantly declined in the COPD smoker group (P values <0.05). MIP-T3 expression was significantly declined in the COPD smoker group (P values <0.05). MIP-T3 expression was significantly declined in the COPD smoker group (P values <0.05). MIP-T3 expression was significantly declined in the COPD smoker group (. Conclusions: Our results suggested that the abnormal ciliary ultrastructure, which was common in COPD patients, might be due to MIP-T3 downregulation.