Automatic quantitative stroke severity assessment based on Chinese clinical named entity recognition with domain-adaptive pre-trained large language model.

BACKGROUND: Stroke is a prevalent disease with a significant global impact. Effective assessment of stroke severity is vital for an accurate diagnosis, appropriate treatment, and optimal clinical outcomes. The National Institutes of Health Stroke Scale (NIHSS) is a widely used scale for quantitatively assessing stroke severity. However, the current manual scoring of NIHSS is labor-intensive, time-consuming, and sometimes unreliable. Applying artificial intelligence (AI) techniques to automate the quantitative assessment of stroke on vast amounts of electronic health records (EHRs) has attracted much interest. OBJECTIVE: This study aims to develop an automatic, quantitative stroke severity assessment framework through automating the entire NIHSS scoring process on Chinese clinical EHRs. METHODS: Our approach consists of two major parts: Chinese clinical named entity recognition (CNER) with a domain-adaptive pre-trained large language model (LLM) and automated NIHSS scoring. To build a high-performing CNER model, we first construct a stroke-specific, densely annotated dataset "Chinese Stroke Clinical Records" (CSCR) from EHRs provided by our partner hospital, based on a stroke ontology that defines semantically related entities for stroke assessment. We then pre-train a Chinese clinical LLM coined "CliRoberta" through domain-adaptive transfer learning and construct a deep learning-based CNER model that can accurately extract entities directly from Chinese EHRs. Finally, an automated, end-to-end NIHSS scoring pipeline is proposed by mapping the extracted entities to relevant NIHSS items and values, to quantitatively assess the stroke severity. RESULTS: Results obtained on a benchmark dataset CCKS2019 and our newly created CSCR dataset demonstrate the superior performance of our domain-adaptive pre-trained LLM and the CNER model, compared with the existing benchmark LLMs and CNER models. The high F1 score of 0.990 ensures the reliability of our model in accurately extracting the entities for the subsequent automatic NIHSS scoring. Subsequently, our automated, end-to-end NIHSS scoring approach achieved excellent inter-rater agreement (0.823) and intraclass consistency (0.986) with the ground truth and significantly reduced the processing time from minutes to a few seconds. CONCLUSION: Our proposed automatic and quantitative framework for assessing stroke severity demonstrates exceptional performance and reliability through directly scoring the NIHSS from diagnostic notes in Chinese clinical EHRs. Moreover, this study also contributes a new clinical dataset, a pre-trained clinical LLM, and an effective deep learning-based CNER model. The deployment of these advanced algorithms can improve the accuracy and efficiency of clinical assessment, and help improve the quality, affordability and productivity of healthcare services.

Digital economy development boosts urban resilience-evidence from China.

Focusing on the impact of the digital economy on urban resilience is beneficial to the sustainable development of cities. This paper empirically examines the impact of digital economic development on urban resilience and its mechanisms by measuring urban resilience and the level of urban digital economy with the entropy-weighted TOPSIS method using the data of 252 Chinese cities from 2011 to 2020. The findings show that digital economic development effectively promotes urban resilience at the 1% significance level, and this conclusion remains valid after a series of endogeneity and robustness tests. The channel mechanism suggests that the development of the digital economy can improve urban resilience by optimizing urban distributional effects and promoting the upgrading of urban industrial structures. This paper discusses the nonlinear relationship between the two using the MMQR model and the threshold model. The results show that urban resilience development level is in a higher quartile of cities, and digital economy development has a greater impact on urban resilience improvement. Meanwhile, there are two threshold values for the nonlinear impact of the digital economy on urban resilience, which are 0.026 and 0.082, respectively. Further, the spatial effect between the two is also verified. From the perspective of heterogeneity analysis, the digital economy development of high-class cities, key city clusters, and cities in eastern and western regions has a greater effect on urban resilience. This study can provide ideas and inspiration for countries to enhance urban resilience and promote sustainable urban development through the development of the digital economy.

Analysis of the Impact of Public Services on Residents' Health: A Spatial Econometric Analysis of Chinese Provinces.

Objectives: The aim of this study was to explore the mechanism between public services and residents' health, focusing on the role of spatial geographical factors. Methods: Leveraging a comprehensive panel dataset encompassing 30 mainland Chinese provinces from 2007 to 2019, this study engineered a spatial Durbin model furnished with dual fixed effects through the application of the Lagrange multiplier, Hausman, and likelihood ratio tests. The primary objective was to delve into the repercussions of varying public service levels on residents' health outcomes. Results: The empirical findings reveal a palpable spatial autocorrelation between residents' health outcomes and the public services levels dispensed across Chinese provinces. Intriguingly, an elevation in the public service level in a given province not only ameliorates its residents' health outcomes but also triggers a spatial spillover effect, thereby positively influencing residents' health in neighboring provinces. The rigorous endogeneity and robustness checks affirm the reliability of the principal outcomes. Conclusion: Due to the increase in social uncertainty, all regions should break free of the administrative monopoly, enhance regional integration and development, and improve residents' health status by clustering public service supply.

Translation and cross-cultural adaption of the Chinese version of the Vanderbilt Head and Neck Symptom Survey version 2.0: a tool for oral symptom assessment in head and neck cancer patients.

BACKGROUND: Patients with head and neck cancer (HNC) who are receiving radiotherapy commonly face detrimental complications, including oral issues. However, oral symptoms are not well understood given the lack of available specific assessment instruments. The Vanderbilt Head and Neck Symptom Survey version (VHNSS) 2.0 is an instrument specifically developed to identify oral symptoms in HNC patients receiving radiotherapy in the United States. OBJECTIVE: To perform the translation and cross-cultural adaptation of the original English version of VHNSS 2.0 into a Chinese version (Mainland China). METHODS: The translation and cultural adaptation process involved translation by independent translators, construction of a consensus version, back translation into the original English version, analysis by the expert committee and a pretest. The pretest was administered to 90 patients with HNC to assess the feasibility and practicality of the tool. RESULTS: The final Chinese version approved by the expert committee was well understood by all participants in the study. The instrument had satisfactory content validity, with indexes of 0.83 for semantic and idiomatic equivalence, 0.90 for cultural equivalence, and 0.91 for conceptual equivalence. Furthermore, this version had good internal consistency, with Cronbach's alpha coefficients ranging from 0.74 to 0.95. CONCLUSION: The Chinese version of VHNSS 2.0 was translated and cross-culturally adapted for use in China. This translation is a feasible instrument to assess oral health-related quality of life in HNC patients undergoing radiotherapy and will be useful for symptom management by clinicians and researchers in China.

Comparison and screening of different risk assessment models for deep vein thrombosis in patients with solid tumors.

To increase the detection rate of deep vein thrombosis (DVT) and to compare the predictive value of four different risk assessment scales (Caprini, Autar, Pauda, and Khorana scales) for DVT in patients with solid tumors by the receiver operating curve (ROC). A total of 361 patients with all kinds of malignant solid tumors, who accepted anti-tumor therapy in the cancer center between March 3, 2015 to April 13, 2018, were assigned to a group of 230 cases diagnosed with DVT and a control group of 131 cases without DVT. Data were recorded and summarized, and the predictive value of the above four risk assessment scales for DVT in solid tumor patients was compared based on the area under the ROC curve (AUC). The AUC values determined for the Caprini, Autar, Pauda, and Khorana scales were (0.631 ± 0.030), (0.686 ± 0.028), (0.654 ± 0.029), and (0.599 ± 0.032), respectively; maximum sensitivity, specificity, and Youden index were 80.9% for Khorana, 86.3% for Caprini, and 29.6% for Autar scale, respectively. We found no statistically significant differences in the AUC values between Autar and Caprini, Autar and Khorana, as well as Khorana and Pauda (p > 0.05). However, the AUC differences between Autar and Pauda, Caprini and Khorana, as well as Caprini and Pauda were statistically significant (p < 0.05). All four risk assessment models showed some value in the risk prediction of DVT in patients with solid tumors, but every model also exhibited its own restrictions; maximum sensitivity, specificity, and Youden index were 80.9% for Khorana, 86.3% for Caprini, and 29.6% for Autar scale, respectively. We confirmed that the detection rate can be improved by modifying the BMI cut-off value of the scale or by combining appropriate scales.

Clinical assessment of cerebral hemodynamics in Moyamoya disease via multiple inversion time arterial spin labeling and dynamic susceptibility contrast-magnetic resonance imaging: A comparative study.

BACKGROUND AND PURPOSE: For Moyamoya disease (MMD) patients, accurate hemodynamic assessment is critical for treatment selection and efficacy assessment. This study aims to investigate the clinical value of mTI-ASL in assessing the cerebral hemodynamics of MMD patients before and after revascularization, relative to DSC-MRI. MATERIALS AND METHODS: Forty-one MMD patients underwent mTI-ASL and DSC-MRI during blood perfusion. Quantitative parameters for the bilateral supply vessels of middle and anterior cerebral arteries, including DSC-TTP, DSC-CBF, ASL-BAT, and ASL-CBF were measured. The correlations between DSC-ΔTTP (TTPhemisphere - TTPbrainstem) and ASL-ΔBAT (BAThemisphere - BATbrainstem) and between DSC-CBF and ASL-CBF were determined. The consistency between the two techniques in assessing the cerebral ischemic state before and after revascularization was analyzed. RESULTS: DSC-ΔTTP and ASL-ΔBAT (r=0.36, P<0.001) and DSC-CBF and ASL-CBF (r=0.32, P<0.001) exhibited significant correlation on 824 regions of interest (ROIs) and similar numbers of ischemic areas on 902 ROIs (κ=0.82, P<0.001). The ischemic scores were 3.17±3.02 and 2.98±2.81 by DSC-MRI and ASL-MRI, respectively (ICC=0.92). For 15 surgically treated patients, the scores for blood perfusion improvement on the operated side were 3.13±1.68 and 3.27±1.33 with DSC-TTP and ASL-BAT, respectively (ICC=0.94). CONCLUSION: Compared to DSC-MRI, mTI-ASL can assess the cerebral hemodynamics in MMD and evaluate ischemic state before revascularization and ischemia reduction after revascularization effectively. And mTI-ASL is more advantageous because it does not require contrast agents.

Next-Generation Molecular Testing of Newborn Dried Blood Spots for Cystic Fibrosis.

Newborn screening for cystic fibrosis enables early detection and management of this debilitating genetic disease. Implementing comprehensive CFTR analysis using Sanger sequencing as a component of confirmatory testing of all screen-positive newborns has remained impractical due to relatively lengthy turnaround times and high cost. Here, we describe CFseq, a highly sensitive, specific, rapid (<3 days), and cost-effective assay for comprehensive CFTR gene analysis from dried blood spots, the common newborn screening specimen. The unique design of CFseq integrates optimized dried blood spot sample processing, a novel multiplex amplification method from as little as 1 ng of genomic DNA, and multiplex next-generation sequencing of 96 samples in a single run to detect all relevant CFTR mutation types. Sequence data analysis utilizes publicly available software supplemented by an expert-curated compendium of >2000 CFTR variants. Validation studies across 190 dried blood spots demonstrated 100% sensitivity and a positive predictive value of 100% for single-nucleotide variants and insertions and deletions and complete concordance across the polymorphic poly-TG and consecutive poly-T tracts. Additionally, we accurately detected both a known exon 2,3 deletion and a previously undetected exon 22,23 deletion. CFseq is thus able to replace all existing CFTR molecular assays with a single robust, definitive assay at significant cost and time savings and could be adapted to high-throughput screening of other inherited conditions.

A 35-day gavage safety assessment of ginger in rats.

Ginger (Zingiber officinale Roscoe, Zingiberacae) is one of the most commonly used spices around the world and a traditional medicinal plant that has been widely used in Chinese, Ayurvedic and Unani-Tibb medicines for several thousand years. However, there was still lack of systemic safety evaluation. We conducted a 35-day toxicity study on ginger in rats. Both male and female rats were daily treated with ginger powder at the dosages of 500, 1000 and 2000 mg/kg body weight by a gavage method for 35 days. The results demonstrated that this chronic administration of ginger was not associated with any mortalities and abnormalities in general conditions, behavior, growth, and food and water consumption. Except for dose-related decrease in serum lactate dehydrogenase activity in males, ginger treatment induced similar hematological and blood biochemical parameters to those of controlled animals. In general, ginger treatment caused no overt organ abnormality. Only at a very high dose (2000 mg/kg), ginger led to slightly reduced absolute and relative weights of testes (by 14.4% and 11.5%, respectively). This study provides a new understanding of the toxicological properties of ginger.