Search

Background: Early-stage diagnosis and treatment can improve survival rates of liver cancer patients. Dynamic contrast-enhanced MRI provides the most comprehensive information for differential diagnosis of liver tumors. However, MRI diagnosis is affected by subjective experience, so deep learning may supply a new diagnostic strategy. We used convolutional neural networks (CNNs) to develop a deep learning system (DLS) to classify liver tumors based on enhanced MR images, unenhanced MR images, and clinical data including text and laboratory test results.Methods: Using data from 1,210 patients with liver tumors (N = 31,608 images), we trained CNNs to get seven-way classifiers, binary classifiers, and three-way malignancy-classifiers (Model A-Model G). Models were validated in an external independent extended cohort of 201 patients (N = 6,816 images). The area under receiver operating characteristic (ROC) curve (AUC) were compared across different models. We also compared the sensitivity and specificity of models with the performance of three experienced radiologists.Results: Deep learning achieves a performance on par with three experienced radiologists on classifying liver tumors in seven categories. Using only unenhanced images, CNN performs well in distinguishing malignant from benign liver tumors (AUC, 0.946; 95% CI 0.914–0.979 vs. 0.951; 0.919–0.982, P = 0.664). New CNN combining unenhanced images with clinical data greatly improved the performance of classifying malignancies as hepatocellular carcinoma (AUC, 0.985; 95% CI 0.960–1.000), metastatic tumors (0.998; 0.989–1.000), and other primary malignancies (0.963; 0.896–1.000), and the agreement with pathology was 91.9%.These models mined diagnostic information in unenhanced images and clinical data by deep-neural-network, which were different to previous methods that utilized enhanced images. The sensitivity and specificity of almost every category in these models reached the same high level compared to three experienced radiologists.Conclusion: Trained with data in various acquisition conditions, DLS that integrated these models could be used as an accurate and time-saving assisted-diagnostic strategy for liver tumors in clinical settings, even in the absence of contrast agents. DLS therefore has the potential to avoid contrast-related side effects and reduce economic costs associated with current standard MRI inspection practices for liver tumor patients.

Choroidal neovascularization (CNV) generally occurs at the late stage of senile macular degeneration (AMD),and accurate segmentation of CNV in optical coherence tomography (SD-OCT) is of great significance for the diagnosis and treatment of AMD.This paper proposes a CNV multi-task segmentation network that combines time series model and attention mechanism.The continuous SD-OCT image is input into the segmentation network,and the multi-scale information of the picture is extracted in the encoder part.In order to better extract the local features of the picture,the attention gate is added in the skip connection part.In order to solve the problem of discontinuous scanning segmentation,after the segmentation network is pooled,the timing constraint network is passed to generate the continuity constraint of adjacent frames and gradient constraints are added to the loss function to better preserve the lesion boundary.The spatial pyramid is used to fuse the two parts of the network feature map to produce segmentation loss,which improves the final segmentation accuracy.Based on patient independence,effective cross-validation is performed on 200 eyes of 12 patients.The Dice coefficient reaches 76.3% and the overlap reaches 60.7%.CNV can be reliably segmented in SD-OCT images.

We here report the multicolor electrochemiluminescence (ECL) of binary microcrystals prepared from a blue‐emissive iridium complex 1 and an orange‐emissive ruthenium complex 2. These materials display a plate‐like morphology with high crystallinity, as demonstrated by microscopic and powder X‐ray diffraction analyses. Under light excitation, these microcrystals exhibit gradient emission color changes as a result of the efficient energy transfer between two complexes. When modified on glass carbon electrodes, these microcrystals exhibit tunable ECLs with varied emission colors including sky‐blue, white, orange, and red, depending on the doping ratio of complex 2 and the applied potential. Furthermore, organic amines with different molecular sizes are used as the co‐reactant to examine their influences on the ECL efficiency of the porous microcrystals of 1. The analysis on the luminance and RGB values of ECL suggests the existence of energy transfer in the generation of multicolor ECLs in these binary crystals. [ABSTRACT FROM AUTHOR]

Organic optical heterostructures have emerged as promising candidates for applications in organic photonics. In this study, we demonstrate a facile strategy to integrate different emission and polarization properties into the sub-blocks of luminescent organic heterostructures, by a mask-assisted vapor treatment method on the microcrystals of a pyridine-functionalized tetraphenylethylene chromophore. [ABSTRACT FROM AUTHOR]

Photonic heterostructures with codable properties have shown great values as versatile information carriers at the micro‐ and nanoscale. These heterostructures are typically prepared by a step‐by‐step growth or post‐functionalization method to achieve varied emission colors with different building blocks. In order to realize high‐throughput and multivariate information loading, we report here a strategy to integrate polarization signals into photonic heterojunctions. A U‐shaped di‐Pt(II) complex has been assembled into highly polarized yellow‐phosphorescent crystalline microrods (Y‐rod) by strong intermolecular Pt⋅⋅⋅Pt interaction. Upon end‐initiated desorption of the incorporated CH2Cl2 solvents, the Y‐rod is transformed in a domino fashion into tri‐block polarized photonic heterojunctions (PPHs) with alternate red‐yellow‐red emissions or red‐phosphorescent microrods (R‐rods). The red emissions of these structures are also highly polarized; however, their polarization directions are just orthogonal to those of the yellow phosphorescence of the Y‐rod. With the aid of a patterned mask, the R‐rod can be further programmed into multi‐block PPHs with precisely controlled block sizes by side‐allowed adsorption of CH2Cl2 vapor. X‐ray diffraction analysis and theoretical calculations suggest that the solvent‐regulated modulation of the crystal packing and excited‐state property is critical for the construction of these PPHs. [ABSTRACT FROM AUTHOR]

Anchoring groups play an important role in the assembly of functional molecules onto semiconductor substrates for dye‐sensitized photocatalysis. In addition to directly influencing the adsorption stability, anchoring groups significantly impact the efficiency of electron or hole injection into the substrate, which is a crucial step for photocatalytic reactions. Most reported dye‐sensitized photocatalytic systems, including semiconductor nanoparticles and photoelectrochemical cells, are prepared using the carboxylic or phosphonic acid anchors. The systems prepared with these conventional anchors usually suffer from low adsorption stability in aqueous media. Alternatively, pyridine has emerged as an anchoring group of photosensitizers and molecular catalysts in the construction of photocatalytic systems. The resulting semiconductor nanoparticles and photoelectrodes show superior adsorption stability in aqueous media, providing a simple and efficient way to sensitize the metal or metal oxide substrate. This review focuses on the recent advances of such a pyridine anchoring strategy in the assembly of photosensitizers and molecular catalysts for photocatalytic applications, mainly including photocatalytic water oxidation, reduction, and splitting. The further exploration and understanding of the pyridine‐anchoring method are believed to boost the development of advanced dye‐sensitized photocatalytic devices and technologies. [ABSTRACT FROM AUTHOR]

Functionalization of Si‐bound methyl group provides an efficient access to diverse organosilanes. However, the asymmetric construction of silicon‐stereogenic architectures by functionalization of Si‐bound methyl group has not yet been described despite recent significant progress in producing chiral silicon. Herein, we disclosed the enantioselective silylmethyl functionalization involving the aryl to alkyl 1,5‐palladium migration to access diverse naphthalenes possessing an enantioenriched stereogenic silicon center, which are inaccessible before. It is worthy to note that the realization of asymmetric induction at the step of metal migration itself remains challenging. Our study constitutes the first enantioselective aryl to alkyl 1,5‐palladium migration reaction. The key to the success is the discovery and fine‐tuning of the different substituents of α,α,α,α‐tetraaryl‐1,3‐dioxolane‐4,5‐dimethanol (TADDOL)‐based phosphoramidites, which ensure the enantioselectivity and desired reactivity. [ABSTRACT FROM AUTHOR]

Single‐molecule white‐light and near‐infrared circularly polarized luminescent materials are effective complements to the demonstrated enhancements of visible emission in the high contrast 3D images or organic light‐emitting diodes. In this contribution, two binaphthalene‐containing luminogenic enantiomers (BIT or BITM) with aggregation‐induced emission are synthesized which exhibit circularly polarized luminescence in solution with dissymmetry factors of ±5.0 × 10−3 and ±4.8 × 10−3, respectively. Additionally, the dual‐fluorescence spectra of single‐molecule white emission (CIE: 0.33, 0.34) in toluene solution are observed with (R)/(S)‐BIT in low polarity solvent originating from locally excited and twisted intramolecular charge‐transfer emission. While (R)/(S)‐BITM only shows single NIR luminescence with peaks at 720 nm in powder state. The nondoped OLEDs exhibit high maximum external quantum efficiencies of up to 4.2% with small efficiency roll‐off and NIR electroluminescence peak at 709 nm. [ABSTRACT FROM AUTHOR]