JLJinjie Liu

Jinjie Liu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Jinjie Liu has a total of 96 co-authors. Here are the top co-authors by number of shared publications:

  • Sibo Liu

    Department of Intensive Care Unit, Central Hospital of Dalian University of Technology

    6 shared publications
  • Hang Li

    Department of Geriatrics, Affiliated Dalian Friendship Hospital of Dalian Medical University

    4 shared publications
  • Ximing Nie

    Department of Neurology, Beijing Tiantan Hospital, Capital Medical University, Beijing, China

    4 shared publications
  • Dean Gu

    Department of Neurology, Qingdao Central Hospital, University of Health and Rehabilitation Sciences

    3 shared publications
  • Lina Zheng

    Department of Neurology, Beijing Tiantan Hospital, Capital Medical University, Beijing, China

    3 shared publications
  • Long Wang

    School of Computer and Communication Engineering, University of Science and Technology Beijing

    3 shared publications

Latest Publications

  • Open Access

    An explainable streaming early identification model for early neurological deterioration based on coordinated fusion of ECG waveforms and vital signs

    by Yuyan Zhang, Shihan Yao, Bo Wen, Jinjie Liu

    Endovascular treatment (EVT) has emerged as a cornerstone in the clinical management of stroke. However, postoperative Early Neurological Deterioration (END) persists as a formidable challenge, frequently correlating with long-term disability and poor prognosis. Consequently, timely and accurate prediction of END is imperative for guiding early clinical intervention and improving patient outcomes. This study introduces DSF-Net, a cost-sensitive, dual-stream multimodal framework engineered for the continuous postoperative monitoring of stroke patients. The architecture leverages a dual-stream design: the first stream employs a 1D convolutional neural network (1D-CNN) to extract latent representations from high-frequency physiological waveforms, while the second stream utilizes a multilayer perceptron (MLP) to encode structured clinical indicators. These fused representations are subsequently processed by a Transformer encoder to capture complex temporal dependencies within granular monitoring windows. To mitigate the challenges posed by extreme class imbalance (i.e., the scarcity of END-positive instances), we integrate cost-sensitive learning with a dynamic threshold optimization strategy prioritized for F1-score maximization. Experimental results demonstrate that DSF-Net achieves superior predictive performance on postoperative monitoring datasets, yielding an Area Under the Curve (AUC) of 0.9996 and an F1-score of 0.9841. Notably, with the optimized threshold, the model enhances the END recall rate to 99%, a significant improvement over the 74% achieved by a conventional LSTM baseline. Furthermore, interpretability analysis using Integrated Gradients (IG) reveals that the model can identify subtle morphological variations in waveforms preceding deterioration, thereby facilitating transparent and actionable early warnings. These findings suggest that DSF-Net provides a robust and interpretable solution for intelligent monitoring in post-stroke care.

  • Top Cited Papers
    Open Access

    Trends and cross-country inequalities in the global burden of Alzheimer’s disease and other dementias among adults aged 65+ years, 1990–2021: a population based study with projections into 2050

    by Sibo Liu, Guozhi Wu, Xiao Ma, Kai Xu, Xiaoyan Lan, Long Wang, Hang Li, Dean Gu, Mengxing Wang, Jinjie Liu

    Background: Alzheimer’s disease and other dementias (ADOD) pose a significant global health challenge, with projected annual increases. The growing elderly population exacerbates burdens, underscoring the need for interventions. Methods: Using data from the Global Burden of Disease Study 2021, we analysed global ADOD trends from 1990 to 2021 among adults aged 65+ and projected to 2050. Average annual percentage change (AAPC) in the age-standardised prevalence, mortality and disability-adjusted life years (DALYs) rates of ADOD were calculated to quantify the temporal trends. Results: Globally, the number of adults aged 65 years or older living with ADOD increased from 18.1 (95% uncertainty interval [UI] 14.4–22.5) million in 1990 to 49.1 (38.7–61.3) million in 2021 (AAPC 0.09%). While mortality rates remained relatively stable (AAPC 0%), DALYs increased by 176% (AAPC 0.02%). Females consistently exhibited a higher age-standardised prevalence (7,603 [95% UI 6,023–9,469] cases per 100,000 population) vs. 5,744 [4,486–7,205]) and mortality (304 [78 to 782] vs. 225 [54–626]) rate compared to males, with notable regional variations. From 1990 to 2021, the age-standardised prevalence of ADOD increased only in East Asia and High-income Asia Pacific (AAPC 0.79% and 0.16%). High fasting plasma glucose, high body-mass index, and smoking were identified as primary risk factors. Projections indicate a 50.1% increase in age-standardised prevalence by 2050, with an estimated 191 (52–330) million cases, and a predicted death toll of 6.8 (2.4–11.3) million. Conclusion: The global number of people living with dementia nearly tripled from 1990 to 2021, mainly due to increases in population ageing and growth. The findings emphasize the need for comprehensive strategies to address ADOD, including prevention, early diagnosis, and effective management, with a focus on gender and regional disparities.


  • Open Access

    Phenotypes of white matter hyperintensities, mechanisms and aetiological differentiation: future targets pre-dementia—a systematic review

    by Dean Gu, Sibo Liu, Yuying Zhang, Zilin Zhao, Hang Li, Yanzhao Li, Xiaoyan Lan, Long Wang, Jinjie Liu, Chenhua Ji

    White matter hyperintensities (WMHs), a major neuroimaging biomarker of cerebral small vessel disease (CSVD), are increasingly recognized as predictors of dementia risk. The subregional distribution and diverse imaging phenotypes of WMHs, including spatial distribution patterns, morphological characteristics, signal intensity, and concomitant lesions, are correlated with distinct mechanisms and aetiologies. This systematic review synthesizes evidence from 31 studies identified through comprehensive searches of PubMed, Embase, Web of Science and the Cochrane Library, following PRISMA guidelines. A key finding is that frontal/deep confluent WMHs with lacunes in the deep white matter are associated with mixed dementia characterized by CSVD and amyloid-β (Aβ) deposition. In contrast, posterior periventricular WMHs coexisting with hippocampal and temporal lobe atrophy and co-localized Aβ deposition serve as early biomarkers of Alzheimer’s disease. Cerebral amyloid angiopathy related WMHs are characterized by juxtacortical multispot patterns and lobar microbleeds, resulting from vascular Aβ deposition and glymphatic system dysfunction. These findings highlight the etiological heterogeneity of WMHs, where spatial distribution patterns and associated imaging features provide critical insights into underlying pathological mechanisms. The identification of these phenotype-specific correlations offers a framework for improving early diagnosis and developing targeted interventions for dementia subtypes.


  • Top Cited Papers
    Open Access

    Global, regional, and national trends in ischaemic stroke burden and risk factors among adults aged 20 + years (1990–2021): a systematic analysis of data from the Global Burden of Disease study 2021 with projections into 2050

    by Sibo Liu, Yanzhao Li, Xiaoyan Lan, Long Wang, Hang Li, Dean Gu, Mengxing Wang, Jinjie Liu

    Background: The objective was to provide standardized, comprehensive, and updated estimates of the global, regional, and national burdens of and risk factors for ischaemic stroke (IS) in adults aged 20 years and older.Methods: This was a population-based study (Global Burden of Disease, Injuries and Risk Factors Study 2021). Adults aged 20 years and older from 204 countries and territories and 811 subnational locations from 1990 to 2021 were included. The primary outcomes were IS-related age-standardized prevalence, mortality, disability-adjusted life years (DALYs), average annual percentage change (AAPC), and risk factors associated with DALYs.Results: From 1990 to 2021, the global age-standardized prevalence of IS decreased from 1,309 (95% UI 1,151 to 1,481) to 1,266 (95% UI 1,120 to 1,423) cases per 100,000 population, with an average annual decrease of −0.12%. However, the prevalence increased notably in the middle sociodemographic index (SDI) regions and East Asia but remained stable in Southeast Asia. The total number of IS cases still increased significantly from 33.2 million to 68.4 million. During the same period, the overall age-standardized mortality for IS decreased from 116 to 70 cases per 100,000 population, with an average annual change of −1.60%. Similarly, the overall age-standardized DALYs for IS decreased by 35%, with an average annual change of −1.36%. The decrease in both the age-standardized mortality and DALY rates was correlated with the SDI, with the most rapid decrease occurring in high-SDI regions. Conversely, in subregions of southern sub-Saharan Africa, an increase was observed. Males consistently faced a greater burden of IS across all subgroups. High systolic blood pressure and high low-density lipoprotein cholesterol (LDL-C) levels consistently represented the most significant risk factors contributing to DALYs from 1990 to 2021.Conclusion: Globally, the total IS caseload has increased. Targeted strategies, such as metabolic risk control in high-SDI regions, low-cost interventions in middle-/low-SDI regions, and improved neuroimaging infrastructure in sub-Saharan Africa, are needed. Future research should focus on subtype-specific burdens, the legacy effects of COVID-19, and intervention cost effectiveness to guide policymakers in developing efficient strategies to combat the global burden of ischaemic stroke.


  • Cerebral Edema Progression and Outcomes in Large Infarct Patients Undergoing Endovascular Thrombectomy

    by Ximing Nie, Jinjie Liu, Bernard Yan, Felix C. Ng, Sibo Liu, Yongle Wang, Mengxing Wang, Lina Zheng, Zan Wang, Yuying Wang, et al.

    Objective: The goal was to analyze the progression of cerebral edema post‐endovascular thrombectomy (EVT) in large infarcts and its association with functional outcomes.Methods: A secondary analysis of the Endovascular Therapy in Acute Anterior Circulation Large Vessel Occlusive Patients with a Large Infarct Core trial was conducted in patients with large ischemic cores randomized to receive either EVT or medical management (MM) alone. Patients who had follow‐up imaging within 7 days post‐randomization were involved. The primary outcome was midline shift (MLS). Mediation analysis was performed with EVT as the independent variable, MLS as the mediator, and modified Rankin scale scores at 90 days served as the endpoint. An exploratory analysis was conducted on the progression of net water uptake (ΔNWU).Results: Of 434 patients, median age was 66.0 years (standard deviation [SD], 9.9), with 61.3% (266) being males. EVT was associated with an early increase in MLS at 24 (±12) hours after randomization (mean [SD], 3.0 [4.2] vs 2.4 [3.6]mm; p = 0.03) compared with the MM group, partially mediating poorer functional outcomes post‐EVT (mediation proportion, −25%; 95% CI, −46.54 to −4.10), but did not negate the overall efficacy of thrombectomy. The progression of NWU remained slower after EVT throughout 7 days, inconsistent with MLS.Interpretation: In patients with large infarct cores, EVT was associated with an early increased mass effect compared with MM, potentially mediating poorer functional outcomes. Despite the evident overall benefits from thrombectomy, accurate prediction and effective anti‐edema interventions for the early mass effect post‐EVT may further improve outcomes. The complex relationship between NWU and cerebral edema progression warrants further investigation. ANN NEUROL 2025;98:258–269


  • Open Access

    Delirium in the ICU: how much do we know? A narrative review

    by Si Bo Liu, Hong Yu Wu, Mei Li Duan, Rong Li Yang, Chen Hua Ji, Jin Jie Liu, Hongtao Zhao

    Delirium in critical ill patients is a complex and common neurological syndrome in the intensive care unit (ICU) that is caused by a range of structural or functional abnormalities. ICU Delirium is associated with reduced compliance, prolonged hospital stays, greater use or delayed withdrawal of sedatives, higher rates and durations of mechanical ventilation, and higher rates of mortality. The aetiology and pathogenesis of ICU delirium are unclear, and the lack of better prediction, prevention, and treatment measures leads to a non-standardized control of delirium. By searching the relevant literature, we aim in this narrative review to describe progress in the pathogenesis, predictive biomarkers, diagnosis, and treatment of ICU delirium.


  • Association of blood pressure and outcomes differs upon cerebral perfusion post-thrombectomy in patients with acute ischemic stroke

    by Jinjie Liu, Ximing Nie, Zhe Zhang, Wanying Duan, Xin Liu, Hongyi Yan, Lina Zheng, Changgeng Fang, Jiaping Chen, Yuyi Wang, et al.

    Background The relationship between post-endovascular thrombectomy (EVT) blood pressure (BP) and outcomes in patients with acute ischemic stroke (AIS) remains contentious. We aimed to explore whether this association differs with different cerebral perfusion statuses post-EVT. Methods In a multicenter observational study of patients with AIS with large vessel occlusion who underwent EVT, we enrolled those who accepted CT perfusion (CTP) imaging within 24 hours post-EVT. We recorded post-EVT systolic (SBP) and diastolic BP. Patients were stratified into favorable perfusion and unfavorable perfusion groups based on the hypoperfusion intensity ratio (HIR) on CTP. The primary outcome was good functional outcome (90-day modified Rankin Scale score of ≤3). Secondary outcomes included early neurological deterioration, infarct size growth, and symptomatic intracranial hemorrhage. Results Of the 415 patients studied (mean age 62 years, 75% male), 233 (56%) achieved good functional outcomes. Logistic regression showed that post-EVT HIR and 24-hour mean SBP were significantly associated with functional outcomes. Among the 326 (79%) patients with favorable perfusion, SBP Conclusions In patients with favorable perfusion post-EVT, SBP Clinical trial registration ChiCTR1900022154.


  • Open Access

    Machine learning-based prediction of mild cognitive impairment among individuals with normal cognitive function

    by Xia Wei Zhu, Si Bo Liu, Chen Hua Ji, Jin Jie Liu, Chao Huang

    Background: Previous studies mainly focused on risk factors in patients with mild cognitive impairment (MCI) or dementia. The aim of the study was to provide basis for preventing MCI in cognitive normal populations.Methods: The data came from a longitudinal retrospective study involving individuals with brain magnetic resonance imaging scans, clinical visits, and cognitive assessment with interval of more than 3 years. Multiple machine-learning technologies, including random forest, support vector machine, logistic regression, eXtreme Gradient Boosting, and naïve Bayes, were used to establish a prediction model of a future risk of MCI through a combination of clinical and image variables.Results: Among these machine learning models; eXtreme Gradient Boosting (XGB) was the best classification model. The classification accuracy of clinical variables was 65.90%, of image variables was 79.54%, of a combination of clinical and image variables was 94.32%. The best result of the combination was an accuracy of 94.32%, a precision of 96.21%, and a recall of 93.08%. XGB with a combination of clinical and image variables had a potential prospect for the risk prediction of MCI. From clinical perspective, the degree of white matter hyperintensity (WMH), especially in the frontal lobe, and the control of systolic blood pressure (SBP) were the most important risk factor for the development of MCI.Conclusion: The best MCI classification results came from the XGB model with a combination of both clinical and imaging variables. The degree of WMH in the frontal lobe and SBP control were the most important variables in predicting MCI.


  • Top Cited Papers
    Open Access

    Chinese Stroke Association guidelines for clinical management of ischaemic cerebrovascular diseases: executive summary and 2023 update

    by Liping Liu, Zixiao Li, Hongyu Zhou, Wanying Duan, Xiaochuan Huo, Weihai Xu, Shujuan Li, Ximing Nie, Huihui Liu, Jinjie Liu, et al.

    Background: China is one of the countries with the highest burden of stroke. Implementing multidimensional management guidelines will help clinicians practise evidence-based care, improve patient outcomes and alleviate societal burdens. This update of the 2019 edition will provide the latest comprehensive recommendations for the diagnosis and treatment of ischaemic cerebrovascular diseases. Methods: We conducted a comprehensive search on MEDLINE (via PubMed) up to 31 August 2023. The writing team established the recommendations through multiple rounds of online and offline discussions. Each recommendation was graded using the evidence grading algorithm developed by the Chinese Stroke Association (CSA). The draft was reviewed and finalised by the CSA Stroke Guidelines Writing Committee. Results: This update included revisions of 15 existing recommendations and 136 new recommendations in the following areas of stroke care: emergency assessment and diagnosis of ischaemic cerebrovascular disease, acute-phase reperfusion therapy, evaluation of underlying mechanisms, antithrombotic therapy, prevention and treatment of complications, and risk factor management. Conclusions: This guideline updated the recommendations for the clinical management of ischaemic cerebrovascular disease from 2019.