Extracting information from unstructured clinical text is a fundamental and challenging task in medical informatics. Our study aims to construct a natural language processing (NLP) workflow to ...extract information from Chinese electronic dental records (EDRs) for clinical decision support systems (CDSSs). We extracted attributes, attribute values, and tooth positions based on an existing ontology from EDRs. A workflow integrating deep learning with keywords was constructed, in which vectors representing texts were unsupervised learned. Specifically, we implemented Sentence2vec to learn sentence vectors and Word2vec to learn word vectors. For attribute recognition, we calculated similarity values among sentence vectors and extracted attributes based on our selection strategy. For attribute value recognition, we expanded the keyword database by calculating similarity values among word vectors to select keywords. Performance of our workflow with the hybrid method was evaluated and compared with keyword-based method and deep learning method. In both attribute and value recognition, the hybrid method outperforms the other two methods in achieving high precision (0.94, 0.94), recall (0.74, 0.82), and F score (0.83, 0.88). Our NLP workflow can efficiently structure narrative text from EDRs, providing accurate input information and a solid foundation for further data-based CDSSs.
A one-step synthesis using the reversed-phase suspension polymerization method and ultraviolet light curing is proposed for preparing the Raman-encoded suspension array (SA). The encoded ...microcarriers are prepared by doping the Raman reporter molecules into an aqueous phase, and then dispersing the aqueous phase in an oil phase and curing by ultraviolet light irradiation. The multiplexed biomolecule detection and various concentration experiments confirm the qualitative and quantitative analysis capabilities of the Raman-encoded SA with a limit of detection of 52.68 pM. The narrow bandwidth of the Raman spectrum can achieve a large number of codes in the available spectral range and the independence between the encoding channel and the fluorescent label channel provides the encoding method with high accuracy. This preparation method is simple and easy to operate, low in cost, and high in efficiency. A large number of hydrogel-based encoding microbeads could be quickly obtained with good biocompatibility. Most importantly, concentrating plenty of Raman reporter molecules inside the microbeads increases the signal intensity and means the molecular assembly is not limited by the functional groups; thus, the types of materials available for Raman encoding method are expanded. Furthermore, the signal intensity–related encoding method is verified by doping different proportions of Raman reporter molecules with our proposed synthesis method, which further increases the detection throughput of Raman-encoded SA.
Graphical Abstract
Full text
Available for:
DOBA, EMUNI, FIS, FZAB, GEOZS, GIS, IJS, IMTLJ, IZUM, KILJ, KISLJ, MFDPS, NLZOH, NUK, OBVAL, OILJ, PILJ, PNG, SAZU, SBCE, SBJE, SBMB, SBNM, UILJ, UKNU, UL, UM, UPUK, VKSCE, ZAGLJ
Medical event detection in narrative clinical notes of electronic health records (EHRs) is a task designed for reading text and extracting information. Most of the previous work of medical event ...detection treats the task as extracting concepts at word granularity, which omits the overall structural information of the clinical notes. In this work, we treat each clinical note as a sequence of short sentences and propose an end-to-end deep neural network framework.
We redefined the task as a sequence labelling task at short sentence granularity, and proposed a novel tag system correspondingly. The dataset were derived from a third-level grade-A hospital, consisting of 2000 annotated clinical notes according to our proposed tag system. The proposed end-to-end deep neural network framework consists of a feature extractor and a sequence labeller, and we explored different implementations respectively. We additionally proposed a smoothed Viterbi decoder as sequence labeller without additional parameter training, which can be a good alternative to conditional random field (CRF) when computing resources are limited.
Our sequence labelling models were compared to four baselines which treat the task as text classification of short sentences. Experimental results showed that our approach significantly outperforms the baselines. The best result was obtained by using the convolutional neural networks (CNNs) feature extractor and the sequential CRF sequence labeller, achieving an accuracy of 92.6%. Our proposed smoothed Viterbi decoder achieved a comparable accuracy of 90.07% with reduced training parameters, and brought more balanced performance across all categories, which means better generalization ability.
Evaluated on our annotated dataset, the comparison results demonstrated the effectiveness of our approach for medical event detection in Chinese clinical notes of EHRs. The best feature extractor is the CNNs feature extractor, and the best sequence labeller is the sequential CRF decoder. And it was empirically verified that our proposed smoothed Viterbi decoder could bring better generalization ability while achieving comparable performance to the sequential CRF decoder.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Abnormal intraneuronal accumulation of soluble and insoluble α-synuclein (α-Syn) is one of the main pathological hallmarks of synucleinopathies, such as Parkinson's disease (PD). It has been well ...documented that the reversible liquid-liquid phase separation of α-Syn can modulate synaptic vesicle condensates at the presynaptic terminals. However, α-Syn can also form liquid-like droplets that may convert into amyloid-enriched hydrogels or fibrillar polymorphs under stressful conditions. To advance our understanding on the mechanisms underlying α-Syn phase transition, we employed a series of unbiased proteomic analyses and found that actin and actin regulators are part of the α-Syn interactome. We focused on Neural Wiskott-Aldrich syndrome protein (N-WASP) because of its association with a rare early-onset familial form of PD. In cultured cells, we demonstrate that N-WASP undergoes phase separation and can be recruited to synapsin 1 liquid-like droplets, whereas it is excluded from α-Syn/synapsin 1 condensates. Consistently, we provide evidence that wsp-1/WASL loss of function alters the number and dynamics of α-Syn inclusions in the nematode Caenorhabditis elegans. Together, our findings indicate that N-WASP expression may create permissive conditions that promote α-Syn condensates and their potentially deleterious conversion into toxic species.
Self-reference detection is necessary and important to a biosensor. The linear weak measurement system based on total internal reflection has attracted widespread attention due to its high stability, ...label-free detection, and easy integration. In this paper, we propose a differential detection method based on the linear total internal reflection weak measurement system. We introduce the half-wave plate (HWP) to convert the H light and the V light to each other, thereby obtaining the difference in phase change of the optical path before and after the HWP. Experiments show that the system can not only achieve differential detection, but also has high stability. The linear differential weak measurement system proposed in this paper not only provides a new differential measurement method for real-time biosensors, but also enriches the types of weak measurement sensors.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
A continuous haze event was recorded on November 14th∼17th, 2020 over Shijiazhuang. Two flights of King-air 350 meteorological research aircraft were performed on November 14th and 16th for the ...retrieval and observations of meteorological elements, aerosols, and black carbon. In this study, we combined airborne data with air pollution data (PM
2.5
), ground meteorological data, and ERA-5 reanalysis data to describe the vertical distribution of aerosols (namely 0.1–3.0 μm) and black carbon. We further explicated the formation of this haze event. PM
2.5
pollution dominated this haze event, and the highest concentration of PM
2.5
was 209 μg/m
3
. The intensity and height of thermal layers highly linked with the vertical transport of pollution. The highest number concentration of aerosols and black carbon was found below the thermal layers on both airborne sounding days. On the 14th, both BC and aerosol concentrations showed unimodal distribution, and the highest concentrations of BC and aerosols were 12683 ng/m
3
and 6965.125#/L at 250 m within layer Ⅰ. The intensity of the thermal layer near-ground was weaker on the 16th that the number concentrations of BC and aerosols also remained at high levels in layer Ⅱ. Backward trajectories of air mass indicated the long-range transport of pollution contributed to the high level of pollution on the 16th. Vapor conditions were more favorable for aerosols growth through moisture absorption. The maximum concentration of 943.58#/L was recorded at particles with a diameter of 0.4 μm on the 16th, while 749.26#/L was reached at 0.14 μm on the 14th. The corresponding height was consistent with the height of maximum concentration in the vertical distribution.
Ubiquitin carboxyl terminal hydrolase L1 (UCH-L1) is one of the deubiquitinating enzymes in the ubiquitin-proteasome system. It has been shown that UCH-L1 could markedly decrease neointima formation ...through suppressing vascular smooth muscle cell (VSMC) proliferation in the balloon-injured rat carotid. However, whether UCH-L1 plays roles in VSMC migration remains to be determined. In this study, the primary VSMCs were isolated from aortic media of rats and TNF-α to was used to induce VSMC migration. Using a modified Boyden chamber and wound healing assay, it was found that TNF-α can dose and time-dependently induce VSMC migration with a maximal effect at 10 ng/mL. Moreover, UCH-L1 expression increased gradually with the prolonged induction time at 10 ng/mL of TNF-α. UCH-L1 content in VSMC was then modulated by recombinant adenoviruses expressing UCH-L1 or RNA interference to evaluate its roles in cell migration. The results showed that over-expression of UCH-L1 attenuated VSMC migration, while knockdown of it enhanced cell migration significantly no matter whether TNF-α treatment or not. Finally, the effect of UCH-L1 on NF-κB activation was demonstrated by NF-κB nuclear translocation and DNA binding activity, and the levels of IL-6 and IL-8 in cell culture media were examined by ELISA. It was showed that UCH-L1 over-expression inhibited NF-κB activation and decrease IL-6 and IL-8 levels, while knockdown of it enhanced NF-κB activation and increase IL-6 and IL-8 levels during TNF-α treatment. These data suggest that UCH-L1 can inhibit TNF-α-induced VSMCs migration, and this kind of effect may partially due to its suppression role in NF-κB activation.
We propose a self-referential fast detection scheme for a frequency domain weak measurement system for the detection of enantiomeric impurities in chiral molecules. In a transmissive weak measurement ...system, the optical rotation (OR) is used to modify the pre-selected polarization state and the post-selection polarization state. We obtained the sum and difference of the optical rotations produced by the sample and the standard by rotating the quarter wave plate in the system. Then, we estimate the ratio of chiral molecules to enantiomeric impurities using the ratio of the central wavelength shifts caused by the addition and subtraction states described above. In this paper, our system has an optical resolution of 1.88 × 10
°. At the same time, we completed the detection of the ratio of the two substances in the mixture of L-proline and D-proline in different proportions, which proved that our system can quickly detect the content of enantiomeric impurities in chiral molecules.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
A deep learning network called "residual neural network" (ResNet) was used to decode Raman spectra-encoded suspension arrays (SAs). With narrow bandwidths and stable signals, Raman spectra have ideal ...encoding properties. The different Raman reporter molecules assembled micro-quartz pieces (MQPs) were grafted with various biomolecule probes, which enabled simultaneous detection of numerous target analytes in a single sample. Multiple types of mixed MQPs were measured by Raman spectroscopy and then decoded by ResNet to acquire the type information of analytes. The good classification performance of ResNet was verified by a t-distributed stochastic neighbor embedding (t-SNE) diagram. Compared with other machine learning models, these experiments showed that ResNet was obviously superior in terms of classification stability and training convergence to different datasets. This method simplified the decoding process and the classification accuracy reached 100%.
A deep learning network called "residual neural network" (ResNet) was used to decode Raman spectra-encoded suspension arrays (SAs).
We propose a method for sensing the non-specific orientation magnetic field based on a simple frequency domain weak measurement system. The magnetic fields with different directions and magnitudes ...induce the phase differences. It experimentally and theoretically demonstrates the high sensitivity of the system to phase differences, which confirms the potential of detecting magnetic fields. We obtain a sensitivity of 1.19 nm/mT and a resolution of 0.70 × 10−3 mT. Further, the axial and radial magnetic fields can be calculated simultaneously by using the characteristics of the central wavelength and intensity of the spectrum. The method has a simple set-up and offers high sensitivity, real-time response to the magnetic field, suggesting a new way of enriching the detection of the vector magnetic field.
•A non-specific orientation magnetic field sensor based on frequency domain weak measurement method is proposed.•The magnetic field sensitivity of the sensor is 1.19 nm/mT, and the resolution of 0.70×10-3 mT.•The proposed magnetic field sensor measures the axial and radial magnetic fields simultaneously.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP