YZYujie Zhu

Yujie Zhu

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yujie Zhu has a total of 23 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Affective Explanations for Autonomous Vehicles: From Framework to Scenario-Based Design Guidelines

    by Shuting Jin, Yujie Zhu, Xingtong Chen, Meichen Liu, Stephen Jia Wang

  • Open Access

    User-Centered Insights into Emotional Experiences and AI-based Emotion Monitoring and Regulation in China: A cross-sectional survey (Preprint)

    by Le Fang, Yujie Zhu, Jiajuan Li, Cong Fang, Yingqing Xu, Stephen Jia Wang

    BACKGROUND Emotional challenges such as stress, anxiety, and depression are on the rise globally. This trend highlights an urgent need for more effective emotion monitoring and regulation (EMR) strategies. Traditional approaches, such as self-reports, often fall short due to the complexities of emotional assessment. At the intersection of artificial intelligence (AI) and human-computer interaction (HCI), emerging technologies offer new opportunities for EMR. Yet, much of the existing research has prioritized technical advancements and usability evaluations of specific systems, with limited attention to users’ lived experiences and attitudes toward EMR technologies. OBJECTIVE This study aimed to investigate emotional experiences, perceptions of AI-based EMR systems, and potential usage scenarios. METHODS A cross-sectional survey was conducted from September to December 2023, recruiting participants aged 18 years and older in China. RESULTS Out of 624 responses, 458 valid questionnaires were analyzed (female: male = 214:244; mean age 34.28 years, SD = 7.57). Emotional challenges were reported daily by 13 participants, weekly by 210, monthly by 170, annually by 62, and never by only 3. The top scenarios associated with negative emotional experiences included family (287), communication (226), friendships (208), work (193), and study (180). Regarding emotion regulation, sports were the most common strategy (274), followed by games (219) and eating (179). More than half (52.8%, 242) were aware of EMR technologies, and 50.2% (230) had prior experience with them. Key factors influencing acceptance included objective analysis (259), accuracy (237), and privacy/security (269). A comparison with counselling highlighted both benefits (objective analysis: 318; efficient availability: 264) and limitations (lack of empathy: 290; lack of body language: 297). Correlation analysis revealed moderate positive correlations among consideration factors (ρ = 0.4–0.7, p < 0.01) towards the use of EMR tools. CONCLUSIONS AI-based EMR systems have the potential to support emotional well-being, but user acceptance depends on trust, accuracy, privacy, and ethical considerations. The study provides actionable design guidelines, emphasizing personalization, transparency, and integration of human oversight (human-in-the-loop). Findings highlight the importance of user-centered approaches to ensure EMR technologies are effective, acceptable, and context-sensitive.

  • Open Access

    AI Doctor for ASD: Physician Perceptions and Adoption Challenges in Autism Clinical Practice

    by Yujie Zhu, Mingyuan Zhang, Cong Fang, Le Fang, Meichen Liu, Yonghao Long, Kun-Pyo Lee, Lie Zhang, Stephen Jia Wang

    The rapid increase in the number of individuals with Autism Spectrum Disorder (ASD) has drawn extensive attention from both the general public and researchers. Artificial Intelligence (AI) has been applied in the assessment, early diagnosis, and intervention of ASD to enhance the efficiency of clinicians and reduce tension in medical resources. However, the adoption of AI systems in clinical practice is relatively limited due to the challenge of complexity and diversity of ASD. Thus, involving insights into clinicians' perceptions and barriers toward the role of AI is crucial for enhancing clinicians-AI cooperation for autism. Through conducting the semi-structured interview with 18 physicians across tertiary and secondary hospitals in various regions, this study indicates the positive attitude toward collaborating with AI among physicians. Additionally, some concerns are also reported, such as the complexity of ASD, uncertainty of AI capabilities, and understandability of AI. The findings of this study highlight the significance of human-centered AI in satisfying different stakeholders' needs and discuss the potential implications of AI capabilities for adopting AI in future autism research.


  • Top Cited Papers
    Open Access

    Digital Behavior Change Intervention Designs for Habit Formation: Systematic Review

    by Yujie Zhu, Yonghao Long, Hailiang Wang, Kun Pyo Lee, Lie Zhang, Stephen Jia Wang

    Background: With the development of emerging technologies, digital behavior change interventions (DBCIs) help to maintain regular physical activity in daily life. Objective: To comprehensively understand the design implementations of habit formation techniques in current DBCIs, a systematic review was conducted to investigate the implementations of behavior change techniques, types of habit formation techniques, and design strategies in current DBCIs. Methods: The process of this review followed the PRISMA (Preferred Reporting Item for Systematic Reviews and Meta-Analyses) guidelines. A total of 4 databases were systematically searched from 2012 to 2022, which included Web of Science, Scopus, ACM Digital Library, and PubMed. The inclusion criteria encompassed studies that used digital tools for physical activity, examined behavior change intervention techniques, and were written in English. Results: A total of 41 identified research articles were included in this review. The results show that the most applied behavior change techniques were the self-monitoring of behavior, goal setting, and prompts and cues. Moreover, habit formation techniques were identified and developed based on intentions, cues, and positive reinforcement. Commonly used methods included automatic monitoring, descriptive feedback, general guidelines, self-set goals, time-based cues, and virtual rewards. Conclusions: A total of 32 commonly design strategies of habit formation techniques were summarized and mapped to the proposed conceptual framework, which was categorized into target-mediated (generalization and personalization) and technology-mediated interactions (explicitness and implicitness). Most of the existing studies use the explicit interaction, aligning with the personalized habit formation techniques in the design strategies of DBCIs. However, implicit interaction design strategies are lacking in the reviewed studies. The proposed conceptual framework and potential solutions can serve as guidelines for designing strategies aimed at habit formation within DBCIs.


  • Open Access

    Developing cue-behavior association for habit formation: A qualitative study to explore the role of avatar in hypertension

    by Yujie Zhu, Yonghao Long, Lai Wei, Yaqi Zhang, Zhengtao Ma, Kun-Pyo Lee, Lie Zhang, Stephen J. Wang

    Background: Electronic health (eHealth) has been widely adopted in chronic disease management. Prior studies focused on time-based reminders as a cue to facilitate behavior change intentions, ignoring the development of automatic cue-behavior associations via other cue types.Objective: Hence, this study utilized avatar appearance as a visual-based cue to help establish the automatic association between appearance transformation and health behavior to form habits without intention.Methods: To better understand users’ attitudes and experiences toward applying changes in avatar appearance to develop cue-behavior associations for hypertensive patients. Fifteen participants were recruited in a 14-day experiment. After excluding one participant who dropped out of the experiment, others were randomly assigned to two groups. One group consisted of a visual-based cue (a virtual plant) and basic behavior change techniques (BCTs). The other group only included basic BCTs. Attitudes and experience outcomes were collected by interview, and qualitative data were analyzed using thematic analysis.Results: 57% of participants had been diagnosed with hypertension for more than five years, and more than 50% of participants have experience using mobile apps or wearables. 66% of participants did physical activity more than three times every week. The result shows that tailored time-based reminders, blood pressure monitoring, and daily dietary intake were the most attractive features. Additionally, hypertensive participants have positive attitudes toward avatar appearance as a visual-based cue to develop cue-behavior association, which enhances self-management motivation.Conclusion: This study proposes a visual-based cue design for habit formation and conducts a qualitative method to explore hypertensive patients’ perceptions. The findings offer insights from user's perspectives into hypertensive patients’ attitudes toward visual-based cues and perception of the connection between avatar appearance and health behavior for self-management. Subsequent discussions present eHealth design guidelines of habit formation from intention, automatic cue-behavior association, and self-management perspectives.


  • Open Access

    Digital Behavior Change Intervention Designs for Habit Formation: Systematic Review (Preprint)

    by Yujie Zhu, Yonghao Long, Hailiang Wang, Kun Pyo Lee, Lie Zhang, Stephen Jia Wang

    BACKGROUND With the development of emerging technologies, digital behavior change interventions (DBCIs) help to maintain regular physical activity in daily life. OBJECTIVE To comprehensively understand the design implementations of habit formation techniques in current DBCIs, a systematic review was conducted to investigate the implementations of behavior change techniques, types of habit formation techniques, and design strategies in current DBCIs. METHODS The process of this review followed the PRISMA (Preferred Reporting Item for Systematic Reviews and Meta-Analyses) guidelines. A total of 4 databases were systematically searched from 2012 to 2022, which included Web of Science, Scopus, ACM Digital Library, and PubMed. The inclusion criteria encompassed studies that used digital tools for physical activity, examined behavior change intervention techniques, and were written in English. RESULTS A total of 41 identified research articles were included in this review. The results show that the most applied behavior change techniques were the self-monitoring of behavior, goal setting, and prompts and cues. Moreover, habit formation techniques were identified and developed based on intentions, cues, and positive reinforcement. Commonly used methods included automatic monitoring, descriptive feedback, general guidelines, self-set goals, time-based cues, and virtual rewards. CONCLUSIONS A total of 32 commonly design strategies of habit formation techniques were summarized and mapped to the proposed conceptual framework, which was categorized into target-mediated (generalization and personalization) and technology-mediated interactions (explicitness and implicitness). Most of the existing studies use the explicit interaction, aligning with the personalized habit formation techniques in the design strategies of DBCIs. However, implicit interaction design strategies are lacking in the reviewed studies. The proposed conceptual framework and potential solutions can serve as guidelines for designing strategies aimed at habit formation within DBCIs.