Attention-based 3D CNN with Multi-layer Features for Alzheimer's Disease Diagnosis using Brain Images.

Structural MRI and PET imaging play an important role in the diagnosis of Alzheimer's disease (AD), showing the morphological changes and glucose metabolism changes in the brain respectively. The manifestations in the brain image of some cognitive impairment patients are relatively inconspicuous, for example, it still has difficulties in achieving accurate diagnosis through sMRI in clinical practice. With the emergence of deep learning, convolutional neural network (CNN) has become a valuable method in AD-aided diagnosis, but some CNN methods cannot effectively learn the features of brain image, making the diagnosis of AD still presents some challenges. In this work, we propose an end-to-end 3D CNN framework for AD diagnosis based on ResNet, which integrates multi-layer features obtained under the effect of the attention mechanism to better capture subtle differences in brain images. The attention maps showed our model can focus on key brain regions related to the disease diagnosis. Our method was verified in ablation experiments with two modality images on 792 subjects from the ADNI database, where AD diagnostic accuracies of 89.71% and 91.18% were achieved based on sMRI and PET respectively, and also outperformed some state-of-the-art methods.

Multi-Modal Learning for Predicting the Genotype of Glioma.

The isocitrate dehydrogenase (IDH) gene mutation is an essential biomarker for the diagnosis and prognosis of glioma. It is promising to better predict glioma genotype by integrating focal tumor image and geometric features with brain network features derived from MRI. Convolutional neural networks show reasonable performance in predicting IDH mutation, which, however, cannot learn from non-Euclidean data, e.g., geometric and network data. In this study, we propose a multi-modal learning framework using three separate encoders to extract features of focal tumor image, tumor geometrics and global brain networks. To mitigate the limited availability of diffusion MRI, we develop a self-supervised approach to generate brain networks from anatomical multi-sequence MRI. Moreover, to extract tumor-related features from the brain network, we design a hierarchical attention module for the brain network encoder. Further, we design a bi-level multi-modal contrastive loss to align the multi-modal features and tackle the domain gap at the focal tumor and global brain. Finally, we propose a weighted population graph to integrate the multi-modal features for genotype prediction. Experimental results on the testing set show that the proposed model outperforms the baseline deep learning models. The ablation experiments validate the performance of different components of the framework. The visualized interpretation corresponds to clinical knowledge with further validation. In conclusion, the proposed learning framework provides a novel approach for predicting the genotype of glioma.

ACLMHA and FML: A brain-inspired kinship verification framework.

As an extended research direction of face recognition, kinship verification based on the face image is an interesting yet challenging task, which aims to determine whether two individuals are kin-related based on their facial images. Face image-based kinship verification benefits many applications in real life, including: missing children search, family photo classification, kinship information mining, family privacy protection, etc. Studies presented thus far provide evidence that face kinship verification still offers many challenges. Hence in this paper, we propose a novel kinship verification architecture, the main contributions of which are as follows: To boost the deep model to capture various and abundant local features from different local face regions, we propose an attention center learning guided multi-head attention mechanism to supervise the learning of attention weights and make different attention heads notice the characteristics of different regions. To combat the misclassification caused by single feature center loss, we propose a family-level multi-center loss to ensure a more proper intra/inter-class distance measurement for kinship verification. To measure the potential similarity of features among relatives better, we propose to introduce the relation comparison module to measure the similarity among features at a deeper level. Extensive experiments are conducted on the widely used kinship verification dataset-Family in the Wild (FIW) dataset. Compared with other state-of-art (SOTA) methods, encouraging results are obtained, which verify the effectiveness of our proposed method.

Genetically Dissecting the Novel Powdery Mildew Resistance Gene in Wheat Breeding Line PBDH1607.

Powdery mildew is one of the most destructive diseases in wheat production. Identifying novel resistance genes and deploying them in new cultivars is the most effective approach to minimize wheat losses caused by powdery mildew. In this study, wheat breeding line PBDH1607 showed high resistance to powdery mildew at both the seedling and adult plant stages. Genetic analysis of the seedling data demonstrated that the resistance was controlled by a single dominant gene, tentatively designated PmPBDH. The ΔSNP index based on bulked segregant RNA sequencing indicated that PmPBDH was associated with an interval of about 30.8 Mb (713.5 to 744.3 Mb) on chromosome arm 4AL. Using newly developed markers, we mapped PmPBDH to a 3.2-cM interval covering 7.1 Mb (719,055,516 to 726,215,121 bp). This interval differed from those of Pm61 (717,963,176 to 719,260,469 bp), MlIW30 (732,769,506 to 732,790,522 bp), and MlNSF10 (729,275,816 to 731,365,462 bp) reported on the same chromosome arm. PmPBDH also differed from Pm61, MlIW30, and MlNSF10 by its response spectrum, origin, or inheritance mode, suggesting that PmPBDH should be a new Pm gene. In the candidate interval, five genes were found to be associated with PmPBDH via time course gene expression analysis, and thus they are candidate genes of PmPBDH. Six closely linked markers, including two kompetitive allele-specific PCR markers, were confirmed to be applicable for tracking PmPBDH in marker-assisted breeding.

Ethylene-induced NbMYB4L is involved in resistance against tobacco mosaic virus in Nicotiana benthamiana.

Several MYB transcription factors are known to play important roles in plant resistance to environmental stressors. However, the mechanism governing the involvement of MYBs in regulating tobacco mosaic virus (TMV) resistance in plants is still unclear. In this study, we found that not only is Nicotiana benthamiana MYB4-like involved in defence against TMV, but also that the ethylene pathway participates in MYB4L-mediated resistance. Transcription of NbMYB4L was up-regulated in N. benthamiana infected with TMV. Silencing of NbMYB4L led to intensified TMV replication, whereas overexpression of NbMYB4L induced significant resistance to TMV. Transcription of NbMYB4L was greater in 1-aminocyclopropanecarboxylic acid (ACC, ethylene precursor)-pretreated plants but lower when the ethylene signalling pathway was blocked during TMV infection. Gene expression analysis showed that the transcription of NbMYB4L was largely suppressed in ETHYLENE INSENSITIVE 3-like 1(EIL1)-silenced plants. The results of electrophoretic mobility shift assay and chromatin immunoprecipitation-quantitative PCR (ChIP-qPCR) experiments indicated that NbEIL1 could directly bind to two specific regions of the NbMYB4L promoter. Furthermore, a luciferase assay revealed that NbEIL1 significantly induced the reporter activity of the MYB4L promoter in N. benthamiana. These results point to NbEIL1 functioning as a positive regulator of NbMYB4L transcription in N. benthamiana against TMV. Collectively, our work reveals that EIL1 and MYB4L constitute a coherent feed-forward loop involved in the robust regulation of resistance to TMV in N. benthamiana.

Spatial adaptive and transformer fusion network (STFNet) for low-count PET blind denoising with MRI.

PURPOSE: Positron emission tomography (PET) has been widely used in various clinical applications. PET is a type of emission computed tomography and operates by positron annihilation radiation. With magnetic resonance imaging (MRI) providing anatomical information, joint PET/MRI reduces the radiation exposure risk of patients. Improved hardware and imaging algorithms have been proposed to further decrease the dose from radioactive tracers or the bed duration, but few methods focus on denoising low-count PET with MRI input. The existing methods are based on fixed conventional convolution and local attention, which do not sufficiently extract and fuse contextual and complementary information from multimodal input. There is still much room for improvement. Therefore, we propose a novel deep learning method for low-count PET/MRI denoising called the spatial-adaptive and transformer fusion network (STFNet), which consists of a Siamese encoder with a spatial-adaptive block (SA-block) and the transformer fusion encoder (TFE). METHODS: Our proposed STFNet consists of a Siamese encoder with an SA-block, TFE, and two branches of the decoder. First, in the encoder, we adapt the SA-block in the Siamese encoder. The SA-block comprises deformable convolution with fusion modulation (DCFM) and two convolutional operations, which can promote network extraction of more relative and long-range contextual features. Second, the pixel-to-pixel TFE helps the network establish a local and global relationship between high-level feature maps of PET and MRI. In the decoder part, we design two branches for PET denoising and MRI translation, and predictions are obtained by trainable weighted summation. This proposed algorithm is implemented to predict synthetic standard-dose neck PET images from low-count neck PET images and MRI. Additionally, this method is compared with the existing U-Net and residual U-Net methods with and without MRI input. RESULTS: To demonstrate the advantages of our method, we introduce configuration studies about TFE, ablation studies, and empirical comparative studies. Quantitative analyses are based on root mean square error (RSME), peak signal-to-noise ratio (PSNR), structural similarity (SSIM), and Pearson correlation coefficient (PCC). Additionally, qualitative results show the comparisons between our proposed method and other existing methods. All experimental results and visualizations show that our method achieves state-of-the-art performance in quantification and qualification. CONCLUSIONS: Based on our experiments, STFNet performs better than existing methods in measurement and visualization. However, our proposed method may still be suboptimal because we apply only the L1 loss to train our data set, and the data set includes corrupted PET with different low counts. In the future, we may exploit a generative adversarial network (GAN)-based paradigm in our STFNet to further improve the visual quality.

Characterization of the Powdery Mildew Resistance Gene in Wheat Breeding Line KN0816 and Its Evaluation in Marker-Assisted Selection.

Wheat powdery mildew, caused by Blumeria graminis (DC.) Speer f. sp. tritici is a destructive disease seriously threatening yield and quality of common wheat (Triticum aestivum L., 2n=6x=42, AABBDD). Characterization of resistance genes against powdery mildew is useful in parental selection and for developing disease-resistant cultivars. Chinese wheat breeding line KN0816 has superior agronomic performance and resistance to powdery mildew at all growth stages. Genetic analysis using populations of KN0816 crossed with different susceptible parents indicated that a single dominant gene, tentatively designated PmKN0816, conferred seedling resistance to different B. graminis f. sp. tritici isolates. Using a bulked segregant analysis, PmKN0816 was mapped to the Pm6 interval on chromosome arm 2BL using polymorphic markers linked to the cataloged genes Pm6, Pm52, and Pm64, and flanked by the markers CISSR02g-6 and CIT02g-2, both with genetic distances of 0.7 cM. Analysis of closely linked molecular markers indicated that the marker alleles of PmKN0816 differed from those of other powdery mildew resistance genes on 2BL, including Pm6, Pm33, Pm51, Pm64, and PmQ. Based on the genetic and physical locations and response pattern to different B. graminis f. sp. tritici isolates, PmKN0816 is most likely a new powdery mildew resistance gene and possesses effective resistance to all the 14 tested B. graminis f. sp. tritici isolates. In view of the elite agronomic performance of KN0816 combined with the resistance, PmKN0816 is expected to become a valuable resistance gene in wheat breeding. To transfer PmKN0816 to different genetic backgrounds using marker-assisted selection (MAS), closely linked markers of PmKN0816 were evaluated, and four of them (CIT02g-2, CISSR02g-6, CIT02g-10, and CIT02g-17) were confirmed to be applicable for MAS in different genetic backgrounds.

Diagnostic Kompetitive Allele-Specific PCR Markers of Wheat Broad-Spectrum Powdery Mildew Resistance Genes Pm21, PmV, and Pm12 Developed for High-Throughput Marker-Assisted Selection.

Wheat powdery mildew is a devastating disease that seriously threatens yield worldwide. Utilization of host resistance is considered an effective strategy to minimize powdery mildew damage. Pm21, PmV, and Pm12 confer broad-spectrum resistance to wheat powdery mildew in China, of which Pm21 and PmV are allelic genes derived from the 6VS chromosome of Dasypyrum villosum, and Pm12 is derived from the 6SS chromosome of Aegilops speltoides and most likely orthologous to the former two genes. To accurately and efficiently transfer and pyramid these genes using marker-assisted selection (MAS), distinctive single-nucleotide polymorphisms (SNPs) among the exon sequences of Pm21, PmV, and Pm12 and their homologous sequences in the common wheat genome were identified and then used for developing diagnostic Kompetitive Allele-Specific PCR (KASP) markers. The markers were validated in different genotypes including transgenic vectors, transgenic lines, translocation lines, resistance stocks with documented Pm genes, and in multiple susceptible cultivars without Pm genes. As a result, we initially developed a KASP marker that can simultaneously diagnose Pm21, Pm12, and PmV. Subsequently, we obtained a highly diagnostic KASP marker for each of the three genes that could distinguish among the three genes and also accurately distinguish them from other resistant stocks with documented Pm genes and from multiple susceptible genotypes. Compared with previously reported markers, the highly diagnostic KASP markers developed in this study have the advantages of low cost, easy assay, accuracy, and potentially high throughput for MAS.

[Sensing of Cu²âº Based on Fenton Reaction and Unmodified Gold Nanoparticles].

Heavy metal pollution has received great attentions in recent years. The traditional methods for heavy metal detection rely on the expensive laboratory instruments and need time-consuming preparation steps; therefore, it is urgent to develop quick and highly sensitive new technologies for heavy metal detection. The colorimetric method based on the gold nanoparticles (AuNPs) features with simple operation, high sensitivity and low cost, therefore, enabling it widely concerned and used in the environmental monitoring, food safety and chemical and biological sensing fields. This work developed a simple, rapid and highly sensitive strategy based on the Fenton reaction and unmodified AuNPs for the detection of Cu²âº in water samples. The hydroxyl radical ( · OH) generated by the Fenton reaction between the Cu²âº and sodium ascorbate (SA) oxidized the single stranded DNA (ssDNA) attached on the AuNPs surface into variable sequence fragments. The cleavage of ssDNA induced the aggregation of AuNPs in a certain salt solution, therefore, resulting in the changes on the absorbance of solution. The assay conditions were optimized to be pH value of 7.9, 11 mg · L⁻¹ ssDNA, 8 mmol · L⁻¹ SA and 70 mmol · L⁻¹ NaCl. Results showed that the absorbance ratio values at the wavelengths of 700 and 525 nm (A700/A525) were linearly correlated with the Cu²âº concentrations. The linear detection range was 0.1-10.0 µmol · L⁻¹ with a detection limit of 24 nmol · L⁻¹ (3σ). Spiked recoveries ranged from 87%-120% in three sorts of water, including drinking water, tap water and lake water, which confirmed that the potentials of the proposed assay for Cu²âº detection in reality.

A hydrophobic hole transporting oligothiophene for planar perovskite solar cells with improved stability.

An oligothiophene derivative named DR3TBDTT with high hydrophobicity was synthesized and functioned as the hole transporting material without an ion additive. 8.8% of power conversion efficiency was obtained for CH3NH3PbI3-xClx based planar solar cells with improved stability, compared to devices using Li-TFSI doped spiro-MeOTAD.

Vertical phase separation in bulk heterojunction solar cells formed by in situ polymerization of fulleride.

Vertical phase separation of the donor and the acceptor in organic bulk heterojunction solar cells is crucial to improve the exciton dissociation and charge transport efficiencies. This is because whilst the exciton diffusion length is limited, the organic film must be thick enough to absorb sufficient light. However, it is still a challenge to control the phase separation of a binary blend in a bulk heterojunction device architecture. Here we report the realization of vertical phase separation induced by in situ photo-polymerization of the acrylate-based fulleride. The power conversion efficiency of the devices with vertical phase separation increased by 20%. By optimising the device architecture, the power conversion efficiency of the single junction device reached 8.47%. We believe that in situ photo-polymerization of acrylate-based fulleride is a universal and controllable way to realise vertical phase separation in organic blends.

Gold nanorod-covered kanamycin-loaded hollow SiO2 (HSKAu(rod)) nanocapsules for drug delivery and photothermal therapy on bacteria.

A hybrid bactericidal material, gold nanorod-covered kanamycin-loaded hollow SiO(2) (HSKAu(rod)) nanocapsules, is constructed. The hybrid material combines the features of a chemical drug with photothermal physical sterilization which decreases the dosage of broad-spectrum antibiotic and the physical damage of biological systems. Hollow SiO(2) nanocapsules are used as carriers for drug delivery. The nanocapsules load a model drug, kanamycin, and are covered with gold nanorods to avoid drug leakage and realize photothermal treatment. The sterilizing effect on the bacterial strain is investigated by incubating E. coli BL21 with the hybrid nanocapsules and irradiating under near-infrared light (NIR) for 20 min. A bactericidal effect, i.e., a sterilizing rate of 53.47%, is achieved for the HSKAu(rod) nanocapsules under NIR irradiation, with respect to a net sum sterilizing rate of 34.49% for the individual components of the HSKAu(rod) nanocapsules, e.g., carrier nanocapsules, chemical sterilization of kanamycin and physical sterilization due to the gold nanorods under NIR irradiation. It is demonstrated that the combination of chemical drug and physical sterilization results in an obvious synergistic effect and makes the sterilization more effective. This novel hybrid has great potential as an adjuvant therapeutic alternative material for sterilization or even for the control of disease.

Fenton's reagent-tuned DNA-templated fluorescent silver nanoclusters as a versatile fluorescence probe and logic device.

A label-free strategy based on the Fenton reaction with DNA-templated silver nanoclusters (DNA-Ag NCs) as a probe is demonstrated for the sequential detection of Cu(2+), ascorbic acid (AA) and H(2)O(2). Cu(2+) causes a structural change of the DNA template in DNA-Ag NCs to resist the environmental quenching and emit stronger fluorescence. The addition of AA in the presence of Cu(2+) results in a further fluorescence increase of the DNA-Ag NCs. Interestingly, an even higher fluorescence enhancement is recorded by introducing Cu(2+) into the DNA-Ag NCs-AA probing system. The fluorescence turn-on probe offers detection limits of 3 nM for Cu(2+) and 7 nM for AA. Thereafter, the addition of H(2)O(2) generates hydroxyl radicals from the Fenton reaction, which induces cleavage of the DNA template, leading to fluorescence quenching of the DNA-Ag NCs. This facilitates H(2)O(2) detection. Moreover, based on the DNA-templated fluorescent silver nanoclusters and Fenton reaction, a multiple logic gate system, including AND and a three-input logic gate, is constructed, with Cu(2+), AA and H(2)O(2) as inputs, and the fluorescence intensity of the DNA-Ag NCs probe as output.

A deep-blue emitter with electron transporting property to improve charge balance for organic light-emitting device.

Highly efficient deep-blue organic light-emitting devices (OLEDs) have been fabricated using 2,7-di(2,2':6',2″-terpyridin-4-yl)-9,9-dioctyl-9H-fluorene (DTPF) as the emitter, which has a wide energy gap, high emission quantum yield (Φf = 0.88), and high electron transporting property to improve the charge balance. A high efficiency of 2.55 cd/A and 2.67 lm/W are obtained in OLED. The device also exhibits a low turn-on voltage of 3.0 V and Commission Internationale de l'Éclairage (CIE) coordinates of (0.16, 0.09).

Label-free colorimetric sensing of ascorbic acid based on Fenton reaction with unmodified gold nanoparticle probes and multiple molecular logic gates.

A label-free strategy based on Fenton reaction with unmodified gold nanoparticles (AuNPs) as probe is demonstrated for ascorbic acid (AA) sensing. AuNPs is stable in the presence of single stranded DNA (ssDNA) which prevents salt-induced aggregation of AuNPs in solution. The hydroxyl free radicals generated by Fenton reaction lead to ssDNA cleavage into different sequence fragments which induce aggregation of AuNPs to produce a red-to-blue color change. As an efficient biological antioxidant, AA could effectively scavenge free radicals to avoid the cleavage of ssDNA, so that it prevents color change of the AuNPs solution. Thus, the color change of AuNPs in the presence and absence of AA provides a new approach for the detection of AA. The absorbance ratio at two wavelengths, A(670)/A(520), decreases linearly with AA content within 1-15 µM, giving rise to a detection limit of 0.3 µM and a RSD of 2.8% (10 µM). The color display of AuNPs solution makes it feasible for the estimation of AA content by naked eye visualization. Moreover, based on Fenton reaction and unmodified gold nanoparticles, a multiple logic gate system includes two logic operations, i.e., INHIBIT and NOR, has been designed with small molecules (AA, l-cysteine, glutathione) as inputs and the colorimetric changes of AuNPs solution as outputs.

The inhibition of fluorescence resonance energy transfer between quantum dots for glucose assay.

Fluorescence resonance energy transfer (FRET) between two quantum dots of different sizes causes fluorescence quenching. Hereby a binding site pre-blocking approach is proposed to avoid this effect. Pre-binding of glucose on the donor occupies the binding sites and thus blocks resonance energy transfer between the two quantum dots, protecting the fluorescence from being quenched. A glucose assay is developed based on this approach. The glucose content is correlated with the fluorescence difference in the absence and in the presence of glucose. In practice, Green QDs-Con A conjugates are used as donors and Red QDs-NH(2)-glu conjugates as acceptors to form FRET system. The inhibition of fluorescence quenching is then measured in the presence of glucose. A linear calibration graph is achieved within 0.1-2.0 mmolL(-1), along with a detection limit of 0.03 mmolL(-1) and a RSD of 2.1% (1.0 mmolL(-1)). 91-105% of glucose in serum and urine samples is recovered. It is worth mentioning that the present glucose assay approach also generates a fluorescence chromatic difference imaging, and the color display clearly identifies the glucose contents by visual detection with a distinguishing ability of ca. 0.5 mmolL(-1). The present approach can potentially be used for the clinical determination of glucose in biological samples which can be further developed into a glucose sensor.

Electro-optical study of chiral nematic liquid crystal/chiral ionic liquid composites with electrically controllable selective reflection characteristics.

A chiral nematic liquid crystal (N*-LC)/chiral ionic liquid (CIL) composite with unique electro-optical characteristics was prepared and filled into a planar treated cell. When an electric field was applied to the cell, the anions and the cations of CIL moved towards the anode and the cathode of the power supply, respectively, thus forming a density gradient of the chiral groups, which resulted in wideband reflection. By adjusting the intensity of the electric field, the reflection bandwidth can be controlled accurately and reversibly. Moreover, the electric field-induced states can be memorized after the applied electric field is turned off. The reflective properties of the composite are investigated in the visible and near-infrared region, respectively. Additionally, the changes of the reflection bandwidths with the intensity and the applied time of the electric field were also investigated. From scanning electron microscopy (SEM) investigations, the mechanism of the electrically controllable reflection was demonstrated. Potential applications of the composite are related to reflective, color electronic paper (E-paper) and smart reflective windows for the solar light management.

[DSC and temperature-dependent infrared spectroscopic study of hydrogen-bonded liquid crystal complexes].

The hydrogen-bonded liquid crystalline complexes based on 4', 4-bipyridine and 4-(trans-4-propylcyclohexyl) benzoic acid and trans-4-(trans-4-propylcyclohexyl)cyclohexyl carboxylic acid assigned as PCBA-BPy and PCCA-BPy were prepared and measured by polarized optical microscopy (POM), differential scanning calorimeter (DSC) and temperature-dependent FTIR It was found that PCBA-BPy and PCCA-BPy exhibited both smectic and nematic phase while all of their predecessors showed no smectic phase. The temperature-dependent FTIR studies revealed that the hydrogen bonding in complex PCBA-BPy was very different from that in PCCA-BPy. The wave number of C=O band had an obvious change at the crystal 1-crystal 2 transition but almost didn't change at smectic-nematic and nematic-isotropic transition; while in PCCA-BPy, it showed no sudden changes but shift to 1 709 cm(-1) gradually with the increase in temperature. The results from temperature-dependent FTIR studies also revealed that when the temperature was higher than the clearing point of the complexes, both of the complexes decomposed partially.

[Protective effects of puerarin against myocardial injury in patients with hypertension during perioperational period].

OBJECTIVE: To investigate the protective effects of puerarin against myocardial injury in patients with hypertension during perioperational period. METHODS: Thirty-four patients with hypertension underwent general anesthesia were randomly divided into the control group and the puerarin group, 500mg puerarin was given to the puerarin group 1hr before anesthesia induction by venoclysis and to the control group, normal saline was given instead. The concentration of serum cardiac troponin I (cTnI) and isoenzyme of creatine kinase containing M and B subunits (CK-MB) were measured before anesthesia induction and 2 hrs after operation respectively. RESULTS: The serum concentration of cTnI and CK-MB were insignificantly different in the two groups before induction, the two indexes increased in different degrees (P <0.01) 2 hrs after operation in both groups, but the increments in the puerarin group were significantly lower than those in the control group (P< 0.01). CONCLUSION: Stress of operation and anesthesia could induce myocardial injury in patients with hypertension, which can be prevented by puerarin medicated during perioperational period.