Boost 3-D Object Detection via Point Clouds Segmentation and Fused 3-D GIoU-L1 Loss.

The 3-D object detection is crucial for many real-world applications, attracting many researchers' attention. Beyond 2-D object detection, 3-D object detection usually needs to extract appearance, depth, position, and orientation information from light detection and ranging (LiDAR) and camera sensors. However, due to more degrees of freedom and vertices, existing detection methods that directly transform from 2-D to 3-D still face several challenges, such as exploding increase of anchors' number and inefficient or hard-to-optimize objective. To this end, we present a fast segmentation method for 3-D point clouds to reduce anchors, which can largely decrease the computing cost. Moreover, taking advantage of 3-D generalized Intersection of Union (GIoU) and L1 losses, we propose a fused loss to facilitate the optimization of 3-D object detection. A series of experiments show that the proposed method has alleviated the abovementioned issues effectively.

ModuleNet: Knowledge-Inherited Neural Architecture Search.

Although neural the architecture search (NAS) can bring improvement to deep models, it always neglects precious knowledge of existing models. The computation and time costing property in NAS also means that we should not start from scratch to search, but make every attempt to reuse the existing knowledge. In this article, we discuss what kind of knowledge in a model can and should be used for a new architecture design. Then, we propose a new NAS algorithm, namely, ModuleNet, which can fully inherit knowledge from the existing convolutional neural networks. To make full use of the existing models, we decompose existing models into different modules, which also keep their weights, consisting of a knowledge base. Then, we sample and search for a new architecture according to the knowledge base. Unlike previous search algorithms, and benefiting from inherited knowledge, our method is able to directly search for architectures in the macrospace by the NSGA-II algorithm without tuning parameters in these modules. Experiments show that our strategy can efficiently evaluate the performance of a new architecture even without tuning weights in convolutional layers. With the help of knowledge we inherited, our search results can always achieve better performance on various datasets (CIFAR10, CIFAR100, and ImageNet) over original architectures.

Characterization of antimicrobial resistance and virulence genes of Pseudomonas aeruginosa isolated from mink in China, 2011-2020.

Pseudomonas aeruginosa strains are potential pathogens that cause respiratory diseases in minks, and caused serious economic loss to mink breeding industry. In this study, we identified antimicrobial resistance and virulence genes in 125 P. aeruginosa isolates from mink in China from 2011 to 2020. The results showed at least one mutation in the gyrA (Thr83Val or Asp87Gly) and parC (Ser87 Leu) genes as well as single mutations in 56 isolates. At least 4-fold reductions in the fluoroquinolone minimum inhibitory concentration values were found when tested in the presence of PAßN in 23 isolates, while 44 isolates were positive for the extended spectrum ß-lactamases and 15 antibiotic resistance genes were identified in this population with a prevalence between 1-32%, including qnrA, CTX-M-1G, ermB and C, cmlA, flor, catl, intl1, tetA, B, C, and D as well as sul1, 2, and 3 genes. Interestingly, one isolate carried ten resistance genes. Five virulence genes were detected, where exoS and algD were the most frequently detected (76.8%), which were followed by plcH (76%), lasB (73.6%), and pilB (31.2%). The isolates carrying the antibiotic resistance or virulence genes were genetically variable, suggesting a horizontal spread through the population. Hence, this study provides novel and important data on the resistance and pathogenicity of P. aeruginosa in farmed mink infections. These data provide important insights into the mechanism of fluoroquinolone resistance in P. aeruginosa, highlighting its usefulness in the treatment and control of P. aeruginosa infections in minks.

In vivo Pharmacokinetic/Pharmacodynamic (PK/PD) Profiles of Tulathromycin in an Experimental Intraperitoneal Haemophilus parasuis Infection Model in Neutropenic Guinea Pigs.

Tulathromycin is a semi-synthetic macrolide antimicrobial that has an important role in veterinary medicine for respiratory disease. The objective of the study was to develop a pharmacokinetic/pharmacodynamic (PK/PD) model to examine the efficacy and determine an optimal dosage of tulathromycin intramuscular (IM) treatment against Haemophilus parasuis infection induced after intraperitoneal inoculation in neutropenic guinea pigs. The PKs of tulathromycin in serum and lung tissue after intramuscular administration at doses of 1, 10, and 20 mg/kg in H. parasuis-infected neutropenic guinea pigs were evaluated by liquid chromatography-tandem mass spectrometry (LC-MS/MS). The tulathromycin minimum inhibitory concentration (MIC) against H. parasuis was ~16 times lower in guinea pig serum (0.03 µg/mL) than in cation-adjusted Mueller-Hinton broth (CAMHB) (0.5 µg/mL). The ratio of the 168-h area under the concentration-time curve (AUC) to MIC (AUC168h/MIC) positively correlated with the in vivo antibacterial effectiveness of tulathromycin (R 2 = 0.9878 in serum and R 2 = 0.9911 in lung tissue). The computed doses to achieve a reduction of 2-log10 CFU/lung from the ratios of AUC72h/MIC were 5.7 mg/kg for serum and 2.5 mg/kg for lung tissue, which lower than the values of 13.2 mg/kg for serum and 8.9 mg/kg for lung tissue with AUC168h/MIC. In addition, using as objective a 2-log10 reduction and an AUC0-72h as the value of the PK/PD index could be more realistic. The results of this study could provide a solid foundation for the application of PK/PD models in research on macrolide antibiotics used to treat respiratory diseases.

Does Serum Uric Acid Status Influence the Association Between Left Atrium Diameter and Atrial Fibrillation in Hypertension Patients?

Objective: Both serum uric acid (SUA) levels and left atrium diameter (LAD) associate with AF. However, the influence of SUA status for the associated risk of AF related to LAD in hypertension patients is currently unknown. Methods: We retrospectively analyzed a hospital-based sample of 9,618 hypertension patients. Standard electrocardiograms were performed on all patients and were interpreted by expert electro-physiologists. Results: Overall 1,028 (10.69%) patients had AF out of 9,618 patients. In men >65 years of age, the prevalence of AF in the1st, 2nd, and 3rd tertiles of SUA among those grouped in the third tertile of LAD were 9, 12.3, and 21.7%, respectively. In the hyperuricemia group, the OR (95% CI) of AF for the highest tertile of LAD in men ≤ 65 years of age was 3.150 (1.756, 5.651; P < 0.001). Similarly, the hyperuricemic men in the 3rd LAD tertile had a higher likelihood of AF than those belonging to the 1st tertile. The ORs and (95% CIs) were 3.150 (1.756, 5.651; P < 0.001) and 5.522 (2.932, 10.400; P ≤ 0.001) for patients ≤ 65 and >65 years of age. An increase in SUA values was significantly associated with an increased likelihood of AF among women at the top tertiles of LAD, with the OR (95% CI) = 4.593 (1.857, 11.358; P = 0.001). Also, men> 65 years of age with large LAD, present at the third tertile of SUA, had a higher likelihood of AF, with the OR (95% CI) = 2.427 (1.039, 5.667; P < 0.05). Conclusion: SUA levels and LAD are associated with AF in patients with hypertension and the risk of AF associated with LAD increases among those with hyperuricemia.

Cancer patients with potential eligibility for vascular endothelial growth factor antagonists use have an increased risk for cardiovascular diseases comorbidities.

BACKGROUND: Recent studies have reported the prevalence of cardiovascular diseases (CVDs) among cancer patients following the use of the vascular endothelial growth factor (VEGF) signaling inhibitors. However, data for patients with a history of cancer before active cancer treatment are lacking. This study aims to investigate the distribution of CVD-related comorbidities before cancer treatment in potential VEGF antagonists candidates. METHODS: A total of 22 500 newly diagnosed cancer patients registered from 1 January 2011 to 31 December 2017 were included. Cancer patients with colorectal cancer (CRC), renal cell carcinoma (RCC), thyroid cancer, hepatocellular carcinoma (HCC), and lung cancer were selected. RESULTS: Hypertension (HTN), coronary heart diseases, atrial fibrillation, and heart failure were top CVD comorbidities among studied cancers. HTN was the most prevalent CVD (26.0%). The prevalence of HTN in RCC, CRC (33.5 and 29.4% respectively) was significantly higher than that in HCC, lung cancer, and thyroid cancer patients (25.1, 24.5, and 23.1%, respectively). Among cancer patients with HTN, the majority of cancer patients fall in grade III (75.7%) and very high cardiovascular risk level (85.4%). Out of the 5847 HTN patients, 26% were not in antihypertensive use, and 34.2% failed to achieve the target blood pressure. CONCLUSION: Cancer patients carry a high burden of CVD-related comorbidities before the application of VEGF antagonists. HTN is the most prevalent comorbid condition, and cancer patients with HTN constitute substantial cardiovascular risks and a higher co-prevalence of other CVDs.