Jinxia Zhang

Jinxia Zhang


Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
5 August 2024 in BMC Medical Informatics and Decision Making
Verified
Journal Article
Open access
PDF file
HCHongyu Chen
Hongyu Chen
LDLi Dan
Li Dan
YLYonghe Lu
Yonghe Lu
MCMinghong Chen
Minghong Chen
JZJinxia Zhang
Jinxia Zhang
Performing data augmentation in medical named entity recognition (NER) is crucial due to the unique challenges posed by this field. Medical data is characterized by high acquisition costs, specialized terminology, imbalanced distributions, and limited training resources. These factors make achieving high performance in medical NER particularly difficult. Data augmentation methods help to mitigate these issues by generating additional training samples, thus balancing data distribution, enriching the training dataset, and improving model generalization. This paper proposes two data augmentation methods—Contextual Random Replacement based on Word2Vec Augmentation (CRR) and Targeted Entity Random Replacement Augmentation (TER)—aimed at addressing the scarcity and imbalance of data in the medical domain. When combined with a deep learning-based Chinese NER model, these methods can significantly enhance performance and recognition accuracy under limited resources. Experimental results demonstrate that both augmentation methods effectively improve the recognition capability of medical named entities. Specifically, the BERT-BiLSTM-CRF model achieved the highest F1 score of 83.587%, representing a 1.49% increase over the baseline model. This validates the importance and effectiveness of data augmentation in medical NER.

18 November 2025 in Frontiers in Cardiovascular Medicine
Verified
Journal Article
Open access
PDF file
YZYinuo Zhu
Yinuo Zhu
LWLei Wang
Lei Wang
YLYan Liu
Yan Liu
GZGuoying Zheng
Guoying Zheng
MWMing Wu
Ming Wu
JZJinxia Zhang
Jinxia Zhang
Objective: The purpose of this research was to examine the risk factors associated with in-hospital mortality in patients with acute coronary syndrome (ACS) concomitant sepsis, and to develop and verify a nomogram model for predicting mortality risk. Methods: This multicenter retrospective analysis examined clinical data from patients with ACS concomitant sepsis who were hospitalized in the intensive care units of tertiary hospitals in Southern China between January 2013 and December 2023. In-hospital mortality functioned as the principal outcome. Univariate and multivariate logistic regression analysis, together with LASSO regression, were used to ascertain independent risk factors for the outcome. The evaluation of model performance was conducted by receiver operating characteristic (ROC) curves, area under the curve (AUC), and calibration plots. Results: This study comprised a total of 200 patients. During hospitalization, 122 people (61.0%) succumbed. Multivariate logistic regression analysis indicated that the diagnosis of ST-segment elevation myocardial infarction (STEMI) at admission (OR = 2.081, 95% CI: 1.120–3.866, P = 0.0206), an elevated initial neutrophil count (OR = 1.05, 95% CI: 1.000–1.102, P = 0.0495), and a history of coronary artery disease (OR = 2.953, 95% CI: 1.173–7.436, P = 0.0215) were independent risk factors for in-hospital mortality. The nomogram model that includes these parameters exhibited an AUC of 0.641 (95% CI: 0.564–0.718), with a sensitivity of 0.656 and a specificity of 0.603. Calibration curves demonstrated strong concordance between expected and observed results (Hosmer-Lemeshow test P > 0.05). Conclusion: Patients with ACS concomitant sepsis experience heightened in-hospital mortality, which is substantially correlated with a diagnosis of STEMI at admission, increased initial neutrophil count, and pre-existing coronary artery disease. While the discriminative capacity (AUC = 0.641) of this three-factor nomogram necessitates additional enhancement, its commendable calibration provides a first instrument for early risk categorization, illustrating practical applicability for swift evaluation. Extensive investigations are necessary to improve model efficacy.

30 March 2024 in International Journal of Pattern Recognition and Artificial Intelligence
Verified
Journal Article
Closed access
YLYonghe Lu
Yonghe Lu
RZRuijie Zhao
Ruijie Zhao
XWXiuxian Wen
Xiuxian Wen
XTXinyu Tong
Xinyu Tong
DXDingcheng Xiang
Dingcheng Xiang
JZJinxia Zhang
Jinxia Zhang

13 November 2024 in Cardiology
Verified
Journal Article
Open access
YGYan Guo
Yan Guo
YLYanzhuo Li
Yanzhuo Li
SLSi Li
Si Li
JMJun Ma
Jun Ma
JLJun Liu
Jun Liu
JZJinxia Zhang
Jinxia Zhang
Introduction: The aim of the study was to investigate the differences in safety and efficacy between high-frequency stimulation (HFS) and anatomically guided endocardial catheter ablation (AA) of the ganglionated plexi (GPs) for treating vasovagal syncope (VVS) in individuals engaged in high-intensity physical training. Methods: Forty-five patients (age 22.5 ± 4.4 years) undergoing high-intensity physical training were included from January 2020 to January 2023 at our hospital. Patients underwent GP ablation for recurrent syncope. Comprehensive evaluations, including head MRI, cardiac ultrasound, electrocardiogram (ECG), ambulatory ECG (Holter), ambulatory blood pressure monitoring, plate motion tests, and head-up tilt tests (HUT), were conducted to exclude other systemic disorders causing syncope. HFS- and AA-guided GP ablation were performed on 10 and 35 patients, respectively, all of whom tested positive for HUT. Differences between the two groups were compared regarding ablation sites, ablation time, safety, and effectiveness. Results: The ablation time was significantly shorter in the AA group compared to the HFS group (p < 0.001). The number of GPs selected for ablation using the AA method was reduced (p < 0.001). All patients in the HFS group experienced palpitations and discomfort, whereas only 31.43% of patients in the AA group reported these symptoms (p = 0.001). Fentanyl analgesia was administered in both groups, and no significant complications arose from the ablation. The longest follow-up duration was 52 months, while the shortest was 15 months. One case of pre-syncope occurred in the HFS group 8 months post-ablation, and one case of pre-syncope and 2 cases of syncope occurred in the AA group at 1 and 3 months post-ablation, respectively. There were no statistically significant differences in heart rate variability and cardiac deceleration capacity (DC) between the two groups after ablation (p > 0.05). Two cases in the AA group still exhibited type II second-degree atrioventricular block during sleep. Both groups of patients were able to complete high-intensity physical training and showed significant symptom improvement post-ablation. Conclusion: Young individuals with VVS engaged in high-intensity physical training can benefit from GP ablation using both HFS and AA methods. The AA method requires relatively simple equipment, shorter procedure time, and results in less discomfort during the ablation.

12 August 2024 in International Journal of Pattern Recognition and Artificial Intelligence
Verified
Journal Article
Closed access
YLYonghe Lu
Yonghe Lu
RZRuijie Zhao
Ruijie Zhao
XWXiuxian Wen
Xiuxian Wen
XTXinyu Tong
Xinyu Tong
DXDingcheng Xiang
Dingcheng Xiang
JZJinxia Zhang
Jinxia Zhang

9 March 2026 in Journal of the Formosan Medical Association
Verified
Journal Article
Open access
XWXiu Wen
Xiu Wen
QLQi Liu
Qi Liu
YDYu Deng
Yu Deng
JZJinxia Zhang
Jinxia Zhang