MRL-Net: Multi-Scale Representation Learning Network for COVID-19 Lung CT Image Segmentation.

Accuracy segmentation of COVID-19 lesions in lung CT images can aid patient screening and diagnosis. However, the blurred, inconsistent shape and location of the lesion area poses a great challenge to this vision task. To tackle this issue, we propose a multi-scale representation learning network (MRL-Net) that integrates CNN with Transformer via two bridge unit: Dual Multi-interaction Attention (DMA) and Dual Boundary Attention (DBA). First, to obtain multi-scale local detailed feature and global contextual information, we combine low-level geometric information and high-level semantic features extracted by CNN and Transformer, respectively. Secondly, for enhanced feature representation, DMA is proposed to fuse the local detailed feature of CNN and the global context information of Transformer. Finally, DBA makes our network focus on the boundary features of the lesion, further enhancing the representational learning. Amounts of experimental results show that MRL-Net is superior to current state-of-the-art methods and achieves better COVID-19 image segmentation performance.

COVID-19 diagnosis via chest X-ray image classification based on multiscale class residual attention.

Aiming at detecting COVID-19 effectively, a multiscale class residual attention (MCRA) network is proposed via chest X-ray (CXR) image classification. First, to overcome the data shortage and improve the robustness of our network, a pixel-level image mixing of local regions was introduced to achieve data augmentation and reduce noise. Secondly, multi-scale fusion strategy was adopted to extract global contextual information at different scales and enhance semantic representation. Last but not least, class residual attention was employed to generate spatial attention for each class, which can avoid inter-class interference and enhance related features to further improve the COVID-19 detection. Experimental results show that our network achieves superior diagnostic performance on COVIDx dataset, and its accuracy, PPV, sensitivity, specificity and F1-score are 97.71%, 96.76%, 96.56%, 98.96% and 96.64%, respectively; moreover, the heat maps can endow our deep model with somewhat interpretability.

A theoretical insight into excited-state decay and proton transfer of p-nitrophenylphenol in the gas phase and methanol solution.

The excited-state decay (ESD) and proton transfer (EPT) of p-nitrophenylphenol (NO2-Bp-OH), especially in the triplet states, were not characterized with high-level theoretical methods to date. Herein, the MS-CASPT2//CASSCF and QM(MS-CASPT2//CASSCF)/MM methods were employed to gain an atomic-level understanding of the ESD and EPT of NO2-Bp-OH in the gas phase and its hydrogen-bonded complex in methanol. Our calculation results revealed that the S1 and S2 states of NO2-Bp-OH are of 1ππ* and 1nπ* characters at the Franck-Condon (FC) point, which correspond to the ICT-EPT and intramolecular charge-transfer (ICT) states in spectroscopic experiments. The former state has a charge-transfer property that could facilitate the EPT reaction, while the latter one might be unfavorable for EPT. The vertical excitation energies of these states are almost degenerate at the FC region and the electronic configurations of 1ππ* and 1nπ* will exchange from the S1 FC region to the S1 minimum, which means that the 1nπ* state will participate in ESD once NO2-Bp-OH departs from the S1 FC region. Besides, we found that three triplets lie below the first bright state and will play very important roles in intersystem crossing processes. In terms of several pivotal surface crossings and relevant linearly interpolated internal coordinate (LIIC) paths, three feasible but competing ESD channels that could effectively lead the system to the ground state or the lowest triplet state were put forward. Once arrived at the T1 state, the system has enough time and internal energy to undergo the EPT reaction. The methanol solvent has a certain effect on the relative energies and spin-orbit couplings, but does not qualitatively change the ESD processes of NO2-Bp-OH. By contrast, the solvent effects will remarkably stabilize the proton-transferred product by the hydrogen bond networks and assist to form the triplet anion. Our present work would pave the road to properly understand the mechanistic photochemistry of similar hydroxyaromatic compounds.

COVID-19 CT image segmentation based on improved Res2Net.

PURPOSE: Corona virus disease 2019 (COVID-19) is threatening the health of the global people and bringing great losses to our economy and society. However, computed tomography (CT) image segmentation can make clinicians quickly identify the COVID-19-infected regions. Accurate segmentation infection area of COVID-19 can contribute screen confirmed cases. METHODS: We designed a segmentation network for COVID-19-infected regions in CT images. To begin with, multilayered features were extracted by the backbone network of Res2Net. Subsequently, edge features of the infected regions in the low-level feature f2 were extracted by the edge attention module. Second, we carefully designed the structure of the attention position module (APM) to extract high-level feature f5 and detect infected regions. Finally, we proposed a context exploration module consisting of two parallel explore blocks, which can remove some false positives and false negatives to reach more accurate segmentation results. RESULTS: Experimental results show that, on the public COVID-19 dataset, the Dice, sensitivity, specificity, S α ${S}_\alpha $ , E ∅ m e a n $E_\emptyset ^{mean}$ , and mean absolute error (MAE) of our method are 0.755, 0.751, 0.959, 0.795, 0.919, and 0.060, respectively. Compared with the latest COVID-19 segmentation model Inf-Net, the Dice similarity coefficient of our model has increased by 7.3%; the sensitivity (Sen) has increased by 5.9%. On contrary, the MAE has dropped by 2.2%. CONCLUSIONS: Our method performs well on COVID-19 CT image segmentation. We also find that our method is so portable that can be suitable for various current popular networks. In a word, our method can help screen people infected with COVID-19 effectively and save the labor power of clinicians and radiologists.

The Infection Control Route in the Operating Room Effectively Reduces the Wound Infection of Patients.

Surgical care is one of the significant aspects of global healthcare, with approximately 234 million operations being conducted annually. Surgical treatment has a substantial risk of complications and death. This study was conducted to explore the application effect of the infection control route in the operating room on the wound infection prevention care of patients. The clinical data of 136 patients receiving surgical treatment from October 2018 to October 2019 were retrospectively analyzed. The participants were assigned via random draw at a ratio of 1 : 1 to receive either routine care management (control group) or the infection control route (research group). The surgical wound infections of patients in the two groups were compared. The research group had higher scores in surgical materials management and disinfectant management than the control group (P < 0.01). In the research group, the total number of colonies within 5 minutes before surgery, 25 minutes after the start of surgery, and after surgery were all smaller than those in the control group (P < 0.01). There were no significant differences in the grade B healing rate between the two groups (P > 0.05), and the research group had a significantly higher healing rate in grade A than the control group, but its grade C healing rate and wound infection rate were significantly lower than those in the control group (P < 0.05). In the research group, the length of hospital stay, the time to get out of bed, the antibiotic use duration, and the stitch removal time was significantly shorter than those in the control group (P < 0.0001). The research group received a higher clinical nursing satisfaction than the control group (P < 0.05). The infection control route in the operating room for infection prevention care effectively reduces the wound infection rate of patients and accelerates their postoperative recovery.

Visual Recognition of Traffic Signs in Natural Scenes Based on Improved RetinaNet.

Aiming at recognizing small proportion, blurred and complex traffic sign in natural scenes, a traffic sign detection method based on RetinaNet-NeXt is proposed. First, to ensure the quality of dataset, the data were cleaned and enhanced to denoise. Secondly, a novel backbone network ResNeXt was employed to improve the detection accuracy and effection of RetinaNet. Finally, transfer learning and group normalization were adopted to accelerate our network training. Experimental results show that the precision, recall and mAP of our method, compared with the original RetinaNet, are improved by 9.08%, 9.09% and 7.32%, respectively. Our method can be effectively applied to traffic sign detection.

Mechanistic photophysics and photochemistry of unnatural bases and sunscreen molecules: insights from electronic structure calculations.

Photophysics and photochemistry are basic subjects in the study of light-matter interactions and are ubiquitous in diverse fields such as biology, energy, materials, and environment. A full understanding of mechanistic photophysics and photochemistry underpins many recent advances and applications. This contribution first provides a short discussion on the theoretical calculation methods we have used in relevant studies, then we introduce our latest progress on the mechanistic photophysics and photochemistry of two classes of molecular systems, namely unnatural bases and sunscreens. For unnatural bases, we disclose the intrinsic driving forces for the ultrafast population to reactive triplet states, impacts of the position and degree of chalcogen substitutions, and the effects of complex environments. For sunscreen molecules, we reveal the photoprotection mechanisms that dissipate excess photon energy to the surroundings by ultrafast internal conversion to the ground state. Finally, relevant theoretical challenges and outlooks are discussed.

Mechanistic Photophysics of Tellurium-Substituted Uracils: Insights from Multistate Complete-Active-Space Second-Order Perturbation Calculations.

The photophysical mechanisms of tellurium-substituted uracils were studied at the multistate complete-active-space second-order perturbation level with a particular focus on how the position and number of tellurium substitutions affect their nonadiabatic relaxation processes. Electronic structure analysis reveals that the lowest several excited states are closely concerned with the n and π orbitals at the Te7-C2 [Te8-C4] moiety of 2-tellurouracil (2TeU) [4TeU and 24TeU]. Both planar and twisted minima were optimized for 2TeU, whereas only planar ones were obtained for 4TeU and 24TeU, except for a twisted T1 minimum of 4TeU. Based on intersection structures and linearly interpolated internal coordinate paths, we proposed several feasible excited-state deactivation paths. It is found that the relaxation channels for 2TeU are more complicated than those of 4TeU and 24TeU. The electronic population transfer to the T1 state for 2TeU is easier than that for 4TeU and 24TeU in consideration of the barrier heights from the S2 Franck-Condon point to the S2/S1 or S2/T2 intersections. In addition, the recovery of the ground state from the T1 state for 2TeU will be more efficient than that for the other two systems as well.

MS-CASPT2 studies on the mechanistic photophysics of tellurium-substituted guanine and cytosine.

Sulfur-substituted nucleobases are highly promising photosensitizers that are widely used in photodynamic therapy, and there are numerous studies exploring their unique photophysical behaviors. However, relevant photophysical investigations on selenium and tellurium substitutions are still rare. Herein, the high-level multistate complete-active-space second-order perturbation (MS-CASPT2) method was performed for the first time to explore the excited-state relaxation processes of tellurium-substituted guanine (TeG) and cytosine (TeC). Based on the electronic state properties in the Franck-Condon (FC) region, we found that the lowest five (S0, S1, S2, T1, and T2) and six (S0, S1, S2, T1, T2 and T3) states will participate in the nonadiabatic transition processes of TeG and TeC systems, respectively. In these electronic states, two kinds of minimum and intersection structures (i.e., planar and twisted structures) were obtained for both TeG and TeC systems. The linearly interpolated internal coordinate (LIIC) paths and spin-orbit coupling (SOC) constants revealed several possible planar and twisted excited-state decay channels, which could lead the systems to the lowest reactive triplet state of T1. Small energy barriers in the T1 state will trap the TeG and TeC systems for a while before they finally populate to the ground state. Although tellurium substitution would further redshift the absorption wavelength and enhance the intersystem crossing (ISC) rate to the T1 state compared with sulfur and selenium substitutions, the rapid ISC process of T1 → S0 may make it a less effective photosensitizer to sensitize the molecular oxygen. We believe our present work will provide important mechanistic insights into the photophysics of tellurium-substituted nucleobases.

Nonadiabatic dynamics simulation of photoinduced ring-opening reaction of 2(5H)-thiophenone with internal conversion and intersystem crossing.

In the present work, the quantum trajectory mean-field approach, which is able to overcome the overcoherence problem, was generalized to simulate internal conversion and intersystem crossing processes simultaneously. The photoinduced ring-opening and subsequent rearrangement reactions of isolated 2(5H)-thiophenone were studied based on geometry optimizations on critical structures and nonadiabatic dynamics simulations using this method. Upon 267 nm irradiation, the molecule is initially populated in the 1ππ* state. After a sudden rupture of one C-S bond within 100 fs in this state, the lowest two singlet excited states and the lowest two triplet excited states become quasi-degenerated, and then the intersystem crossing processes between singlet and triplet states accompanied by rearrangement reactions can be observed several times. Compared with our previous nonadiabatic simulations in the absence of intersystem crossing (ChemPhotoChem, 2019, 3, 897-906), some new nonadiabatic relaxation pathways involving triplet states and different ring-opening products were identified. The present work provides new mechanistic insights into the photoinduced ring-opening of thio-substituted heterocyclic molecules and reveals the importance of nonadiabatic dynamics simulation that is able to deal with multiple electronic states with different spin multiplicities.

[Determination of homocysteine in plasma by precolumn derivatization-high performance liquid chromatography with fluorescence detection].

A precolumn derivatization-high performance liquid chromatographic method for the determination of homocysteine (Hcy) in plasma was established. Tris (2-carboxyethyl) phosphine hydrochloride (TCEP) and N-(1-pyrenyl) maleimide (NPM) were used as the reduced reagent and derivatization reagent, respectively. The separation was carried out on an Agilent Hypersil C-18 column (250 mm x 4.0 mm, 5 microm) in gradient elution mode. The mobile phase consisted of A (15 mmol/L sodium acetate solution), B (acetonitrile) and C (300 mL water containing 1 mL acetic acid and 1 mL phosphoric acid). The eluate was monitored by the fluorescence detector at an excitation wavelength of 330 nm and an emission wavelength of 380 nm. The mean recovery of Hcy was (102.08 +/- 4.94)%. The linear range was from 0.500 micromol/L to 100 micromol/L, with a detection limit of 0.016 micromol/L. The intra-day and inter-day relative standard deviations (RSDs) for Hcy were less than 5%. Seven plasma samples of patients with hypertension and seven plasma samples of healthy controls were tested, and the results demonstrated that the Hcy in the plasma from the hypertension group was significantly different from that of the control group (p < 0.05). The developed method is simple, fast, accurate, and suitable for clinical measurement.

Simultaneous determination of tryptophan, kynurenine and 5-hydroxytryptamine by HPLC: Application in uremic patients undergoing hemodialysis.

OBJECTIVES: To develop a reliable HPLC method for the simultaneous determination of plasma tryptophan, kynurenine and 5-hydroxytryptamine to analyze tryptophan metabolism. DESIGN AND METHODS: Separation was carried out on a C8 column with the mobile phase composed of acetate buffer (pH 4.5) and acetonitrile using theophylline as internal standard. The eluates were monitored by ultraviolet detection with programmed wavelength. RESULTS: Analysis was achieved in less than 8.0min. The limits of quantification were 3.97µmol/L, 4.36nmol/L and 0.421µmol/L for tryptophan, 5-hydroxytryptamine and kynurenine, respectively. Reproducibility and recovery were satisfactory. Twenty healthy adults and 20 uremic patients undergoing hemodialysis were analyzed using the present method. Tryptophan metabolism was found to be disturbed in uremic patients and was improved obviously after hemodialysis. CONCLUSIONS: The developed HPLC method is simple, reliable and suitable for monitoring tryptophan metabolism in uremic patients undergoing hemodialysis.