Progressive attention integration-based multi-scale efficient network for medical imaging analysis with application to COVID-19 diagnosis.

In this paper, a novel deep learning-based medical imaging analysis framework is developed, which aims to deal with the insufficient feature learning caused by the imperfect property of imaging data. Named as multi-scale efficient network (MEN), the proposed method integrates different attention mechanisms to realize sufficient extraction of both detailed features and semantic information in a progressive learning manner. In particular, a fused-attention block is designed to extract fine-grained details from the input, where the squeeze-excitation (SE) attention mechanism is applied to make the model focus on potential lesion areas. A multi-scale low information loss (MSLIL)-attention block is proposed to compensate for potential global information loss and enhance the semantic correlations among features, where the efficient channel attention (ECA) mechanism is adopted. The proposed MEN is comprehensively evaluated on two COVID-19 diagnostic tasks, and the results show that as compared with some other advanced deep learning models, the proposed method is competitive in accurate COVID-19 recognition, which yields the best accuracy of 98.68% and 98.85%, respectively, and exhibits satisfactory generalization ability as well.

Self-assembled three-dimensional periodic micro-nano structures in bulk quartz crystal induced by femtosecond laser pulses.

In-volume, self-assembled, three-dimensional, periodic micro-nano structures are induced in quartz crystal by tightly focused, 500-kHz femtosecond laser pulses. With suitable pulse energy, three different types of periodic structures can be observed in modified regions using scanning electron microscopy. The first one with period (ΛE) of ~400 nm in the direction of the laser polarization, i.e. nanograting, shows indicative features similar to that in fused silica. The second one with period (ΛS) in the scan direction and the third one with period (Λk) in the laser propagation direction are both equally spaced by ~1 µm, which is close to the laser wavelength. Moreover, the structure with period (Λk) covers almost the whole cross-section of modified regions, which is distinctive to that observed in fused silica. Through the comparison of the structures induced by 1-kHz pluses and those by 500-kHz pluses, we deduce that the heat accumulation effect may have a positive influence on the formation of nanogratings in quartz crystal.

Enhanced up-conversion luminescence in transparent glass-ceramic containing KEr3F10:Er3+ nanocrystals and its application in temperature detection.

Transparent Er3+ doped glass-ceramics containing KEr3F10:Er3+ nanocrystals were prepared via the traditional melt-quenching technique. The micro-structures, optical properties and up-conversion luminescence behaviors of the Er3+ doped glass-ceramics were systemically studied using X-ray diffraction, absorption and up-conversion luminescence spectra. Under the excitation of a laser at 980 nm, the intensity of red emission of the glass-ceramics increases more than 70 times after heat treatment compared with that of the precursor glass ceramic. Moreover, the fluorescence intensity ratio of the thermally coupled energy levels (2H11/2 and 4S3/2) shows a good linear relationship with temperature and maintains relatively high sensitivities of 0.398% per K at 303-573 K, which indicates that the glass ceramics have promising applications in temperature detection.

Correction: Enhanced up-conversion luminescence in transparent glass-ceramic containing KEr3F10:Er3+ nanocrystals and its application in temperature detection.

[This corrects the article DOI: 10.1039/C9RA01945F.].

Development and parameter identification of a visco-hyperelastic model for the periodontal ligament.

The present study developed and implemented a new visco-hyperelastic model that is capable of predicting the time-dependent biomechanical behavior of the periodontal ligament. The constitutive model has been implemented into the finite element package ABAQUS by means of a user-defined material subroutine (UMAT). The stress response is decomposed into two constitutive parts in parallel which are a hyperelastic and a time-dependent viscoelastic stress response. In order to identify the model parameters, the indentation equation based on V-W hyperelastic model and the indentation creep model are developed. Then the parameters are determined by fitting them to the corresponding nanoindentation experimental data of the PDL. The nanoindentation experiment was simulated by finite element analysis to validate the visco-hyperelastic model. The simulated results are in good agreement with the experimental data, which demonstrates that the visco-hyperelastic model developed is able to accurately predict the time-dependent mechanical behavior of the PDL.

Mechanical responses of the periodontal ligament based on an exponential hyperelastic model: a combined experimental and finite element method.

The V-W exponential hyperelastic model is adopted to describe the instantaneous elastic response of the periodontal ligament (PDL). The general theoretical framework of constitutive modeling is described based on nonlinear continuum mechanics, and the elasticity tensor used to develop UMAT subroutine is formulated. Nanoindentation experiment is performed to characterize mechanical properties of an adult pig PDL specimen. Then the experiment is simulated by using the finite element (FE) analysis. Meanwhile, the optimized material parameters are identified by the inverse FE method. The good agreement between the simulated results and experimental data demonstrates that the V-W model is capable of describing the mechanical behavior of the PDL. Therefore, the model and its implementation into FE code are validated. By using the model, we simulate the tooth movement under orthodontic loading to predict the mechanical responses of the PDL. The results show that local concentrations of stress and strain in the PDL are found.