Diagnosis and detection of pneumonia using weak-label based on X-ray images: a multi-center study.

PURPOSE: Development and assessment the deep learning weakly supervised algorithm for the classification and detection pneumonia via X-ray. METHODS: This retrospective study analyzed two publicly available dataset that contain X-ray images of pneumonia cases and normal cases. The first dataset from Guangzhou Women and Children's Medical Center. It contains a total of 5,856 X-ray images, which are divided into training, validation, and test sets with 8:1:1 ratio for algorithm training and testing. The deep learning algorithm ResNet34 was employed to build diagnostic model. And the second public dataset were collated by researchers from Qatar University and the University of Dhaka along with collaborators from Pakistan and Malaysia and some medical doctors. A total of 1,300 images of COVID-19 positive cases, 1,300 normal images and 1,300 images of viral pneumonia for external validation. Class activation map (CAM) were used to location the pneumonia lesions. RESULTS: The ResNet34 model for pneumonia detection achieved an AUC of 0.9949 [0.9910-0.9981] (with an accuracy of 98.29% a sensitivity of 99.29% and a specificity of 95.57%) in the test dataset. And for external validation dataset, the model obtained an AUC of 0.9835[0.9806-0.9864] (with an accuracy of 94.62%, a sensitivity of 92.35% and a specificity of 99.15%). Moreover, the CAM can accurately locate the pneumonia area. CONCLUSION: The deep learning algorithm can accurately detect pneumonia and locate the pneumonia area based on weak supervision information, which can provide potential value for helping radiologists to improve their accuracy of detection pneumonia patients through X-ray images.

Fully biodegradable polylactide foams with ultrahigh expansion ratio and heat resistance for green packaging.

Long chain branching (LCB) structures are efficiently introduced into polylactide (PLA) by employing sustainable soybean oil (SO) under the initiation of trace amount of cyclic peroxide, which displays robust foamability and heat resistance. It is discovered that with the introduction of 0.6 wt% SO, the expansion ratio and Vicat softening temperature of LCB PLA are sharply raised to 75.2-fold and 155.8 °C, respectively, which is about 17.9 and 2.6 times those of linear PLA. This is because that the amounts of LCB structures are significantly increased in LCB PLA by the addition of SO with low reactivity of internal CC bonds, which can avoid the oligomerization reaction, resulting in more dramatically improved melting strength and crystallization performance of LCB PLA. Moreover, the hydrolytic degradation of LCB PLA is largely expedited as compared to linear PLA, owing to the more rapid water permeation caused by the loose packing of LCB structures. Finally, the PLA foam tray with light weight and good heat resistance is successfully developed by using LCB PLA with 0.6 wt% SO through extrusion foaming with supercritical carbon oxide and thermoforming techniques. Hence, this research offers a green route to produce eco-friendly light-weight and high-heat-resistance LCB-PLA foam with full biodegradability, which is an ideal alternative to the non-degradable oil-based plastics in the field of disposable packaging products.

Construction of Multiplex Muscle Network for Precision Pinch Force Control.

Muscle synergy is a fundamental mechanism of motor control. Despite a number of studies focusing on muscle synergy during power grip and pinch at high-level force, relatively less is known about the functional interactions between muscles within low-level force production during precision pinch. Traditional analytical tools such as nonnegative matrix factorization or principal component analysis have limitations in processing nonlinear dynamic electromyographic signals and have confined sensitivity particularly for the low-level force production. In this study, we developed a novel method - multiplex muscle networks, to investigate the dynamical coordination of muscle activities at low-level force production during precision pinch. The multiplex muscle network was constructed based on multiplex limited penetrable horizontal visibility graph (MLPHVG). Seven forearm and hand muscles, including brachioradialis (BR), flexor carpi ulnaris (FCU), flexor carpi radialis (FCR), flexor digitorum superficialis (FDS), extensor digitorum communis (EDC), abductor pollicis brevis (APB) and first dorsal interosseous (FDI), were examined using surface electromyography (sEMG). Eight healthy subjects were instructed to perform a visuomotor force tracking task by producing higher (10% MVC) and lower (1% MVC) precision pinch. Interlayer mutual information I, average edge overlap ω weighted clustering coefficient CW, weighted characteristic path length LW were selected as network metrics. We assessed the undirected weighted network generated from multiplex muscle network after taking the I between paired muscle network layers as edge. There are significant differences between higher and lower force level with higher I, ω, CW and lower LW at higher force level. Advanced efficiency of information processing in the regional and global perspective indicated dynamical alterations when human faces the higher force tracking task. It suggested that ω may be an important characteristic to classify different force control states with the average classification accuracy of 82.21%. These findings reveal related alterations of functional interactions between muscles involved in precision pinch. The novel method for constructing multiplex muscle network may provide insights into muscle synergies during precision pinch force control.

Synthesis of poly(ionic liquid) for trifunctional epoxy resin with simultaneously enhancing the toughness, thermal and dielectric performances.

Poly(ionic liquid) (PIL), integrating the characteristics of both polymers and ionic liquid, is synthesized and employed to modify diglycidyl-4,5-epoxy-cyclohexane-1,2-dicarboxylate (TDE-85). With the addition of PIL, the fracture toughness, and thermal and dielectric performances of TDE-85 were discovered to be simultaneously improved, meanwhile the tensile modulus and strength is increased. Upon an optimal loading of 3 wt% PIL, the critical stress intensity factor (K IC), tensile modulus and strength are raised by 92.9%, 13.3% and 10.7%, respectively. Multi-toughening mechanisms due to spherical domains of PIL formed in TDE-85 during curing are responsible for the improved toughness. Moreover, the dielectric and thermal properties of TDE-85 are also enhanced by adding PIL. With the optimal addition of 5 wt% PIL, the dielectric constant of the composites is enhanced by 62.5%, the glass transition temperature is increased by 16.58 °C and the residual weight of carbon is increased by 59%.

Directed Connectivity in Large-scale Brain Networks for Precision Grip Force Control.

Precision grip requires accurate but dynamic control of the magnitude and direction of fingertip forces. It is still not well known whether directed information flow across activated cortical regions could change with different force outputs during precision grip. This study aimed to investigate the directed connectivity in large-scale brain networks for precision grip force control. Eight healthy volunteers participated in the experiment. Totally 32 channel electroencephalography (EEG) signals were recorded during a precision grip task that requires accurate force control following a dynamically changed visual target. The force target changed between 10% and 1% Maximal Voluntary Contraction (MVC). Horizontal visibility graph transfer entropy (HVG-TE) was used to measure the directed connectivity between pairs of EEG channels. In addition, the relative HVG-TE (rHVG-TE) was applied to evaluate the activation of each EEG channel. Results showed relatively higher rHVG-TE values in the posterior brain region but lower rHVG-TE values in the anterior brain region within the whole spectrum (4-70Hz), indicating a posterior-to-anterior information flow. This study revealed that the activation of posterior region is higher than the anterior region. There is a directed functional connectivity of cortex from posterior to anterior region for precision grip force control. This study shed light on how to quantify activation and large-scale connectivity of cortical regions, and reveal the in-depth central mechanisms for peripheral fine motor control.