MHMingxin He

Mingxin He

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Mingxin He has a total of 18 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Effectiveness of Conversational Agents for Depression and Anxiety: Systematic Review and Network Dose-Response Meta-Analysis of Agent Types, Dose, and Human Support

    by Pengyu Yang, Ping Lei Chui, Mingxin He, Yao Wu, Pengfei Cheng, Minxiang Li, Haojie Sun, Lurong Cai, Hua Zhang

    Background: Conversational agents (CAs) are increasingly used to deliver scalable interventions for depression and anxiety, but existing evidence has often treated these tools as a single category. The comparative effectiveness of different agent types, intervention doses, and levels of human support remains unclear. Objective: This review systematically evaluated the comparative effectiveness of various CA modalities and levels of human support, as well as potential dose-response relationships across different CA types. Methods: A comprehensive search of PubMed, Embase, Web of Science, Scopus, Ovid, CINAHL, and Psychology and Behavioral Sciences Collection (PBSC) databases was conducted from inception to December 2025. Eligible studies were randomized controlled trials (RCTs) evaluating CA-based interventions. We conducted a random-effects network meta-analysis (NMA) using postintervention outcomes and assessed inconsistency and risk of bias (RoB). Evidence certainty was assessed using the CINeMA (confidence in network meta-analysis) framework. Where data permitted, model-based network dose-response meta-analysis (MBNMA) was performed to examine dose-response associations. Results: A total of 41 studies (n=7577) were included. NMA showed that AI-enhanced intensive human support (AEIHS) appeared to produce greater symptom improvement for depression and anxiety compared with pure human digital guidance (PHDG; Hedges g=–4.28, 95% CI –5.56 to –2.99) and open-ended multimodal-enhanced chatbot (OEMM; Hedges g=–4.49, 95% CI –6.31 to –2.66), whereas most other comparisons were not statistically significant. MBNMA indicated that goal-oriented conversational agent (GOCA) and highly structured modular chatbot (HSMC) appeared to exhibit nonlinear effects, with preliminary evidence suggesting potentially enhanced response at low-to-moderate doses, though these findings require further investigation. However, publication bias was detected for depression (Egger P=.02), and the overall certainty of evidence was very low, primarily due to RoB, indirectness, and imprecision. Conclusions: While hybrid CA models with enhanced human involvement appeared to show potential advantages over fully automated systems, the evidence base remains limited. The very low certainty of evidence constrains confidence in comparative rankings and dose recommendations. Future high-quality trials are needed to establish optimal CA architectures and intervention doses for scalable mental health care. Trial Registration: PROSPERO CRD420251145207; https://www.crd.york.ac.uk/PROSPERO/view/CRD420251145207

  • Open Access

    Effectiveness of Conversational Agents for Depression and Anxiety: Systematic Review and Network Dose-Response Meta-Analysis of Agent Types, Dose, and Human Support (Preprint)

    by Pengyu Yang, Ping Lei Chui, Mingxin He, Yao Wu, Pengfei Cheng, Minxiang Li, Haojie Sun, Lurong Cai, Hua Zhang

    BACKGROUND Conversational agents (CAs) are increasingly used to deliver scalable interventions for depression and anxiety, but existing evidence has often treated these tools as a single category. The comparative effectiveness of different agent types, intervention doses, and levels of human support remains unclear. OBJECTIVE This review systematically evaluated the comparative effectiveness of various CA modalities and levels of human support, as well as potential dose-response relationships across different CA types. METHODS A comprehensive search of PubMed, Embase, Web of Science, Scopus, Ovid, CINAHL, and Psychology and Behavioral Sciences Collection (PBSC) databases was conducted from inception to December 2025. Eligible studies were randomized controlled trials (RCTs) evaluating CA-based interventions. We conducted a random-effects network meta-analysis (NMA) using postintervention outcomes and assessed inconsistency and risk of bias (RoB). Evidence certainty was assessed using the CINeMA (confidence in network meta-analysis) framework. Where data permitted, model-based network dose-response meta-analysis (MBNMA) was performed to examine dose-response associations. RESULTS A total of 41 studies (n=7577) were included. NMA showed that AI-enhanced intensive human support (AEIHS) appeared to produce greater symptom improvement for depression and anxiety compared with pure human digital guidance (PHDG; Hedges g=–4.28, 95% CI –5.56 to –2.99) and open-ended multimodal-enhanced chatbot (OEMM; Hedges g=–4.49, 95% CI –6.31 to –2.66), whereas most other comparisons were not statistically significant. MBNMA indicated that goal-oriented conversational agent (GOCA) and highly structured modular chatbot (HSMC) appeared to exhibit nonlinear effects, with preliminary evidence suggesting potentially enhanced response at low-to-moderate doses, though these findings require further investigation. However, publication bias was detected for depression (Egger P=.02), and the overall certainty of evidence was very low, primarily due to RoB, indirectness, and imprecision. CONCLUSIONS While hybrid CA models with enhanced human involvement appeared to show potential advantages over fully automated systems, the evidence base remains limited. The very low certainty of evidence constrains confidence in comparative rankings and dose recommendations. Future high-quality trials are needed to establish optimal CA architectures and intervention doses for scalable mental health care. CLINICALTRIAL PROSPERO CRD420251145207; https://www.crd.york.ac.uk/PROSPERO/view/CRD420251145207

  • Open Access

    Construction and Efficiency Estimation of a Nursing Human Resource Evaluation System in Integrated Medical‐Nursing Elderly Care Institutions Using Data Envelopment Analysis

    by Mingxin He, Pengfei Cheng, Yan Liu, Ziwei Zhang, Yuhan Yang, Ka Yan Ho, Yanyan Li

    Background: Efficient allocation of nursing human resources (NHR) is critical for optimizing care quality in integrated medical‐nursing elderly care institutions. However, standardized tools for assessing NHR efficiency remain underdeveloped. Objective: This study aimed to develop and validate a data envelopment analysis (DEA)–based evaluation system for nursing human resource efficiency in integrated elderly care institutions, with empirical application in clinical settings. Methods: Employing a cross‐sectional design, the research evaluated NHR efficiency in integrated medical‐nursing elderly care institutions. A preliminary indicator system was developed through comprehensive literature review and field investigations, followed by two rounds of Delphi consultations with 17 multidisciplinary experts specializing in nursing management, elderly care administration, integrated medical‐nursing operations, health economics, and public health. Expert reliability was determined by assessing response rates, authority levels (qualifications and experience), consensus (agreement rates), and coordination (Kendall’s W test). Based on this established indicator system, the DEA model was employed to evaluate the operational efficiency of 12 integrated medical‐nursing elderly care institutions in Hainan Province, China. Results: The constructed evaluation system featured a three‐level hierarchical structure totaling 68 indicators (9 first‐level, 19 second‐level, and 40 third‐level indicators). Two rounds of expert consultation demonstrated strong participation (response rates: 100% and 94.1%, respectively) and high reliability (authority coefficients: 0.88 and 0.92). Expert consensus was confirmed through statistical analysis (Kendall’s W : 0.471 and 0.348; average coefficient of variation: 0.16 and 0.12; all p < 0.001). Subsequent DEA implementation across 12 institutions revealed 5 were fully efficient (OE = TE = SE = 1.000), while the remaining 7 showed varying inefficiency patterns: 4 exhibited pure technical efficiency (TE = 1.000) with scale inefficiency (SE < 1.000), and 3 demonstrated both technical and scale inefficiency (TE < 1.000, SE < 1.000). Conclusion: The nursing human resource efficiency evaluation system developed in this study demonstrated strong validity through a rigorous Delphi expert consultation process, showing high expert engagement and authoritative consensus. The comprehensive three‐level indicator system exhibits well‐organized structure with strong specialty‐specific relevance for integrated medical‐nursing care settings. DEA application confirmed the system’s effectiveness in objectively evaluating nursing efficiency, supporting its practical utility for healthcare management in elderly care institutions.

  • Open Access

    Effectiveness of a virtual reality-based sensory stimulation intervention in preventing delirium in intensive care units: a randomised-controlled trial protocol

    by Surui Liang, Yong Liu, Taoxue Wen, Dan Luo, Mingxin He, Jinfei Tian

    Introduction: Delirium is a common acute cognitive impairment characterised by confusion, disorientation and attention deficits, particularly prevalent in intensive care unit (ICU) settings. Given its significant impact on patients, caregivers and healthcare resources, preventing delirium in patients in the ICU is of paramount importance. This is the first randomised-controlled trial designed to evaluate the effects of a virtual reality-based sensory stimulation intervention on preventing delirium in ICU patients.Methods and analysis: We employed a paired randomisation method to match eligible participants based on a validated delirium risk scoring model for patients in the ICU. The study will commence in September 2024 and conclude in June 2026. A consecutive sample of 198 patients in the ICU admitted to the study setting will be recruited. Eligible participants will be randomly allocated to receive either virtual reality-based sensory stimulation in addition to usual care or usual care alone. The virtual reality-based sensory stimulation intervention will last for up to 14 days, with all interventions administered by a research team. We define delirium-free days over a 14-day period as the primary outcome. The secondary outcomes will include delirium incidence, duration and severity; patients’ psychological well-being (post-traumatic stress disorder, sleep quality and ICU memory); patients’ clinical outcomes and other outcomes (quality of life, independence and cognitive function). Data will be collected at baseline, post-intervention and 6 months post-intervention. Two independent t-tests or Wilcoxon-Mann-Whitney tests will be used for continuous variables, while χ2or Fisher’s exact tests will be employed for categorical variables. The analysis will adhere to both the intention-to-treat and per-protocol principles. Additionally, mixed-effects models and subgroup analysis will be planned.Ethics and dissemination: This protocol was approved by the Research Ethics Committee of Shenzhen Hospital of Southern Medical University (NYSZYYEC20230068). All participants or their family caregivers will provide written informed consent. Results will be disseminated through scientific publications, and presentations at local and international conferences.Trial registration number: ClinicalTrials.govNCT06153472. Trial registration date: 22 November 2023.