A Novel Strategy for Extracting Richer Semantic Information Based on Fault Detection in Power Transmission Lines.

With the development of the smart grid, the traditional defect detection methods in transmission lines are gradually shifted to the combination of robots or drones and deep learning technology to realize the automatic detection of defects, avoiding the risks and computational costs of manual detection. Lightweight embedded devices such as drones and robots belong to small devices with limited computational resources, while deep learning mostly relies on deep neural networks with huge computational resources. And semantic features of deep networks are richer, which are also critical for accurately classifying morphologically similar defects for detection, helping to identify differences and classify transmission line components. Therefore, we propose a method to obtain advanced semantic features even in shallow networks. Combined with transfer learning, we change the image features (e.g., position and edge connectivity) under self-supervised learning during pre-training. This allows the pre-trained model to learn potential semantic feature representations rather than relying on low-level features. The pre-trained model then directs a shallow network to extract rich semantic features for downstream tasks. In addition, we introduce a category semantic fusion module (CSFM) to enhance feature fusion by utilizing channel attention to capture global and local information lost during compression and extraction. This module helps to obtain more category semantic information. Our experiments on a self-created transmission line defect dataset show the superiority of modifying low-level image information during pre-training when adjusting the number of network layers and embedding of the CSFM. The strategy demonstrates generalization on the publicly available PASCAL VOC dataset. Finally, compared with state-of-the-art methods on the synthetic fog insulator dataset (SFID), the strategy achieves comparable performance with much smaller network depths.

Simultaneous Improvement of Ionic Conductivity and Mechanical Strength in Block Copolymer Electrolytes with Double Conductive Nanophases.

The most daunting challenge of solid polymer electrolytes (SPEs) is the development of materials with simultaneously high ionic conductivity and mechanical strength. Herein, SPEs of lithium bis-(trifluoromethanesulfonyl)imide (LiTFSI)-doped poly(propylene monothiocarbonate)-b-poly(ethylene oxide) (PPMTC-b-PEO) block copolymers (BCPs) with both blocks associating with Li+ ions are prepared. It is found that the PPMTC-b-PEO/LiTFSI electrolytes with double conductive phases exhibit much higher ionic conductivity (2 × 10-4 S cm-1 at r.t.) than the BCP electrolytes with a single conductive phase. Concurrently, the storage moduli of PPMTCn -b-PEO44 /LiTFSI electrolytes are ≈1-4 orders of magnitude higher than that of the neat PEO/LiTFSI electrolytes. Therefore, simultaneous improvement of ionic conductivity and mechanical properties is achieved by construction of a microphase-separated and disordered structure with double conductive phases.

Crystallization-driven one-dimensional self-assembly of polyethylene-b-poly(tert-butylacrylate) diblock copolymers in DMF: effects of crystallization temperature and the corona-forming block.

Crystallization-driven self-assembly of polyethylene-b-poly(tert-butylacrylate) (PE-b-PtBA) block copolymers (BCPs) in N,N-dimethyl formamide (DMF) was studied. It is found that all three PE-b-PtBA BCPs used in this work can self-assemble into one-dimensional crystalline cylindrical micelles. When the BCP solution is cooled to crystallization temperature (Tc) from 130 °C, the seed micelles may be produced via two competitive processes in the initial period: stepwise micellization/crystallization and simultaneous crystallization/micellization. Subsequently, the seed micelles can undergo growth driven by the epitaxial crystallization of the unimers. The lengths of both the seed micelles and the grown micelles are longer for the BCP with a longer PtBA block at a higher Tc. Quasi-living growth of the PE-b-PtBA crystalline cylindrical micelles is achieved at a higher Tc. A longer PtBA block evidently retards the attachment of unimers to the crystalline micelles, leading to a slower growth rate.

The role of epicardial fat radiomic profiles for atrial fibrillation identification and recurrence risk with coronary CT angiography.

OBJECTIVES: Fat radiomic profile (FRP) was a promising imaging biomarker for identifying increased cardiac risk. We hypothesize FRP can be extended to fat regions around pulmonary veins (PV), left atrium (LA), and left atrial appendage (LAA) to investigate their usefulness in identifying atrial fibrillation (AF) and the risk of AF recurrence. METHODS: We analysed 300 individuals and grouped patients according to the occurrence and types of AF. We used receiver operating characteristic and survival curves analyses to evaluate the value of imaging biomarkers, including fat attenuation index (FAI) and FRP, in distinguishing AF from sinus rhythm and predicting post-ablation recurrence. RESULTS: FRPs from AF-relevant fat regions showed significant performance in distinguishing AF and non-AF with higher AUC values than FAI (peri-PV: FRP = 0.961 vs FAI = 0.579, peri-LA: FRP = 0.923 vs FAI = 0.575, peri-LAA: FRP = 0.900 vs FAI = 0.665). FRPs from peri-PV, peri-LA, and peri-LAA were able to differentiate persistent and paroxysmal AF with AUC values of 0.804, 0.819, and 0.694. FRP from these regions improved AF recurrence prediction with an AUC of 0.929, 0.732, and 0.794. Patients with FRP cut-off values of ≥0.16, 0.38, and 0.26 had a 7.22-, 5.15-, and 4.25-fold higher risk of post-procedure recurrence, respectively. CONCLUSIONS: FRP demonstrated potential in identifying AF, distinguishing AF types, and predicting AF recurrence risk after ablation. FRP from peri-PV fat depot exhibited a strong correlation with AF. Therefore, evaluating epicardial fat using FRP was a promising approach to enhance AF clinical management. ADVANCES IN KNOWLEDGE: The role of epicardial adipose tissue (EAT) in AF had been confirmed, we focussed on the relationship between EAT around pulmonary arteries and LAA in AF which was still unknown. Meanwhile, we used the FRP to excavate more information of EAT in AF.

A deep learning-based fully automatic and clinical-ready framework for regional myocardial segmentation and myocardial ischemia evaluation.

Myocardial ischemia diagnosis with CT perfusion imaging (CTP) is important in coronary artery disease management. Traditional analysis procedure is time-consuming and error-prone due to the semi-manual and operator-dependent nature. To improve the diagnostic performance, a deep learning-based, fully automatic, and clinical-ready framework was developed. Two collaborating deep learning networks including a 3D U-Net for left ventricle segmentation and a CNN for anatomical landmarks detection were trained on 276 subjects. With our processing framework, the 17-segment left ventricular model was automatically generated conformed to the clinical standard. Myocardial blood flow computed by commercial software was extracted within each segment and visualized against the bull's eye plot. The performance was validated on another 45 subjects. Coronary angiography and invasive fractional flow reserve measurements were also performed in these patients to serve as the gold standard for myocardial ischemia diagnosis. As a result, the diagnostic accuracy for our method was 81.08%, much higher than that for commercially available CTP analysis software (56.75%), and our method demonstrated a higher consistency (Kappa coefficient 0.759 vs. 0.585). Besides, the average processing time of our method was much lower (30 ± 10.5 s/subject vs. over 30 min/subject). In conclusion, the proposed deep learning-based framework could be a promising tool for assisting CTP analysis.

One-stop patient-specific myocardial blood flow quantification technique based on allometric scaling law.

Establishing a patient-specific and non-invasive technique to derive blood flow as well as coronary structural information from one single cardiac CT imaging modality. 336 patients with chest pain or ST segment depression on electrocardiogram were retrospectively enrolled. All patients underwent adenosine-stressed dynamic CT myocardial perfusion imaging (CT-MPI) and coronary computed tomography angiography (CCTA) in sequence. Relationship between myocardial mass (M) and blood flow (Q), defined as log(Q) = b · log(M) + log(Q0), was explored based on the general allometric scaling law. We used 267 patients to obtain the regression results and found strong linear relationship between M (gram) and Q (mL/min) (b = 0.786, log(Q0) = 0.546, r = 0.704; p < 0.001). We Also found this correlation was applicable for patients with either normal or abnormal myocardial perfusion (p < 0.001). Datasets from the other 69 patients were used to validate this M-Q correlation and found the patient-specific blood flow could be accurately estimated from CCTA compared to that measured from CT-MPI (146.480 ± 39.607 vs 137.967 ± 36.227, r = 0.816, and 146.480 ± 39.607 vs 137.967 ± 36.227, r = 0.817, for the left ventricle region and LAD-subtended region, respectively, all unit in mL/min). In conclusion, we established a technique to provide general and patient-specific myocardial mass-blood flow correlation obeyed to allometric scaling law. Blood flow information could be directly derived from structural information acquired from CCTA.

Self-Assembled Amphiphilic Block Copolymers/CdTe Nanocrystals for Efficient Aqueous-Processed Hybrid Solar Cells.

Due to their low cost and high efficiency, polymer/nanocrystal hybrid solar cells (HSCs) have attracted much attention in recent years. In this work, water-soluble hybrid materials consisting of amphiphilic block copolymers (ABCPs) and cadmium telluride nanocrystals (CdTe NCs) were used as the active layer to fabricate the HSCs via aqueous processing. The ABCPs composed of poly(3-hexylthiophene) (P3HT) and poly(acrylic acid) (PAA) self-assembled into ordered nanostructured micelles which then transformed to nanowires by comicellization with P3HT additives. Furthermore, after annealing, the hybrid materials formed an interpenetrating network which resulted in a maximum power conversion efficiency of 4.8% in the HSCs. The properties of the hybrid materials and the film morphology were studied and correlated to the device performance. The results illustrate how the inclusion of ABCPs for directed assembly and homo-P3HT for charge transport and light absorption improves device performance. The aqueous-processed HSCs based on the ABCPs and NCs offer an effective method for the fabrication of efficient solar cells.