Selected work

Representative Papers

BMC4TimeSec: Verification Of Timed Security Protocols

Feb 19, 2026

This work addresses the challenge of verifying security protocols with timing constraints in complex temporal environments by proposing an end-to-end formal verification method based on Timed Interpretation Systems (TIS/TIIS). The approach uniquely integrates the TIS/TIIS framework with knowledge automata to jointly model the temporal behaviors and epistemic evolution of multiple agents, including adversaries. Automated verification is achieved through SMT solving and bounded model checking. The contributions include a reproducible open-source verification toolkit, accompanying source code, and demonstration videos, which have been successfully applied to verify the security of several representative timed protocols, significantly advancing the formal analysis capabilities for timing-sensitive security protocols.

0 citationsRead paper

Women have it Worse: an ICT Workplace Digital Transformation Stress Gender Gap

Oct 18, 2025

This study reveals that digital transformation (DT) exacerbates gender inequality in employee well-being within ICT workplaces, as women experience significantly higher digital transformation stress (DTS) due to entrenched technological stereotypes. Method: Addressing the lack of objective behavioral data in prior research, we introduce a novel mixed-methods approach integrating natural language processing (NLP)-based sentiment analysis of IT helpdesk ticket texts with standardized psychological self-report scales—validated across dual data sources: real-world service desk logs and employee surveys. Contribution/Results: Results robustly confirm a gender gap in DTS that persists after controlling for job role, tenure, and other covariates. We advance the “gendered technical support demand” perspective, arguing that organizational DT governance must explicitly incorporate targeted technological adaptation support for women. This provides empirically grounded insights and actionable intervention pathways for fostering inclusive digital work environments.

0 citationsRead paper

IoT and Older Adults: Towards Multimodal EMG and AI-Based Interaction with Smart Home

Jul 25, 2025

This study addresses the limited accessibility of standard interfaces (e.g., voice, touch) for older adults and people with disabilities in smart home environments. We propose a non-invasive, multimodal biosignal interaction paradigm integrating surface electromyography (EMG), electrooculography (EOG), and speech signals. A lightweight AI model enables real-time intent recognition, while participatory design iteratively refines human–machine collaboration logic, implemented in the Sagacity prototype system. Our key contribution is the first integration of low-cost EMG/EOG sensing with edge AI to deliver a contactless, low-cognitive-load control solution tailored to vulnerable users. Experimental evaluation in realistic domestic settings confirms feasibility and robustness, identifies critical technical bottlenecks, and elicits authentic user requirements. The work provides a reproducible design framework and empirical evidence for accessible intelligent interaction.

0 citationsRead paper

Immersive Technologies in Training and Healthcare: From Space Missions to Psychophysiological Research

Jun 30, 2025

This study addresses suboptimal human cognitive performance and health management efficacy in high-stakes, highly regulated domains—including space missions, clinical psychological assessment, and medical education—by proposing a cross-domain integrated extended reality (XR) framework. Methodologically, it pioneers the systematic integration of immersive simulation, multimodal physiological–psychological measurement, and real-time health monitoring to establish closed-loop interactive systems for astronaut autonomous health diagnosis, medical student surgical training, and patient rehabilitation interventions. Key contributions include enhanced ecological validity in psychological experimentation, improved accuracy in dynamic spaceflight health assessment, and increased situational fidelity in medical education and patient adherence in rehabilitation. Empirical results demonstrate significant improvements in astronauts’ self-diagnostic accuracy, medical students’ procedural proficiency, and patients’ rehabilitation engagement—validating XR’s effectiveness and broad applicability in high-pressure, high-precision operational contexts.

0 citationsRead paper

Towards AI-based Depression and Social Anxiety Screening Through Eye Tracking

Mar 22, 2025International journal of marketing, communication and new media

This study addresses the need for early, non-invasive screening of depression and social anxiety disorder. Methodologically, it introduces a novel AI-assisted assessment paradigm that converts individual gaze scanpaths into grayscale trajectory images—encoding spatiotemporal attention dynamics—and feeds them into a lightweight ResNet architecture to discriminate affective disorder subtypes via visual attention pattern analysis. Key contributions include: (1) the first formulation of scanpath image representation for eye-tracking data, and (2) an end-to-end deep learning framework for multi-class affective disorder classification. Empirical evaluation yields 48% accuracy in three-way classification (depression vs. social anxiety vs. healthy controls) and 62% in binary classification (clinical disorder vs. healthy), significantly validating the discriminative power of oculomotor features. The approach offers a scalable, ecologically valid, and low-cost solution for preliminary mental health screening.

0 citationsRead paper

Recent publications

Latest Papers

BMC4TimeSec: Verification Of Timed Security Protocols

Feb 19, 2026

This work addresses the challenge of verifying security protocols with timing constraints in complex temporal environments by proposing an end-to-end formal verification method based on Timed Interpretation Systems (TIS/TIIS). The approach uniquely integrates the TIS/TIIS framework with knowledge automata to jointly model the temporal behaviors and epistemic evolution of multiple agents, including adversaries. Automated verification is achieved through SMT solving and bounded model checking. The contributions include a reproducible open-source verification toolkit, accompanying source code, and demonstration videos, which have been successfully applied to verify the security of several representative timed protocols, significantly advancing the formal analysis capabilities for timing-sensitive security protocols.

0 citationsRead paper

Women have it Worse: an ICT Workplace Digital Transformation Stress Gender Gap

Oct 18, 2025

This study reveals that digital transformation (DT) exacerbates gender inequality in employee well-being within ICT workplaces, as women experience significantly higher digital transformation stress (DTS) due to entrenched technological stereotypes. Method: Addressing the lack of objective behavioral data in prior research, we introduce a novel mixed-methods approach integrating natural language processing (NLP)-based sentiment analysis of IT helpdesk ticket texts with standardized psychological self-report scales—validated across dual data sources: real-world service desk logs and employee surveys. Contribution/Results: Results robustly confirm a gender gap in DTS that persists after controlling for job role, tenure, and other covariates. We advance the “gendered technical support demand” perspective, arguing that organizational DT governance must explicitly incorporate targeted technological adaptation support for women. This provides empirically grounded insights and actionable intervention pathways for fostering inclusive digital work environments.

0 citationsRead paper

IoT and Older Adults: Towards Multimodal EMG and AI-Based Interaction with Smart Home

Jul 25, 2025

This study addresses the limited accessibility of standard interfaces (e.g., voice, touch) for older adults and people with disabilities in smart home environments. We propose a non-invasive, multimodal biosignal interaction paradigm integrating surface electromyography (EMG), electrooculography (EOG), and speech signals. A lightweight AI model enables real-time intent recognition, while participatory design iteratively refines human–machine collaboration logic, implemented in the Sagacity prototype system. Our key contribution is the first integration of low-cost EMG/EOG sensing with edge AI to deliver a contactless, low-cognitive-load control solution tailored to vulnerable users. Experimental evaluation in realistic domestic settings confirms feasibility and robustness, identifies critical technical bottlenecks, and elicits authentic user requirements. The work provides a reproducible design framework and empirical evidence for accessible intelligent interaction.

0 citationsRead paper

Immersive Technologies in Training and Healthcare: From Space Missions to Psychophysiological Research

Jun 30, 2025

This study addresses suboptimal human cognitive performance and health management efficacy in high-stakes, highly regulated domains—including space missions, clinical psychological assessment, and medical education—by proposing a cross-domain integrated extended reality (XR) framework. Methodologically, it pioneers the systematic integration of immersive simulation, multimodal physiological–psychological measurement, and real-time health monitoring to establish closed-loop interactive systems for astronaut autonomous health diagnosis, medical student surgical training, and patient rehabilitation interventions. Key contributions include enhanced ecological validity in psychological experimentation, improved accuracy in dynamic spaceflight health assessment, and increased situational fidelity in medical education and patient adherence in rehabilitation. Empirical results demonstrate significant improvements in astronauts’ self-diagnostic accuracy, medical students’ procedural proficiency, and patients’ rehabilitation engagement—validating XR’s effectiveness and broad applicability in high-pressure, high-precision operational contexts.

0 citationsRead paper

Towards AI-based Depression and Social Anxiety Screening Through Eye Tracking

Mar 22, 2025International journal of marketing, communication and new media

This study addresses the need for early, non-invasive screening of depression and social anxiety disorder. Methodologically, it introduces a novel AI-assisted assessment paradigm that converts individual gaze scanpaths into grayscale trajectory images—encoding spatiotemporal attention dynamics—and feeds them into a lightweight ResNet architecture to discriminate affective disorder subtypes via visual attention pattern analysis. Key contributions include: (1) the first formulation of scanpath image representation for eye-tracking data, and (2) an end-to-end deep learning framework for multi-class affective disorder classification. Empirical evaluation yields 48% accuracy in three-way classification (depression vs. social anxiety vs. healthy controls) and 62% in binary classification (clinical disorder vs. healthy), significantly validating the discriminative power of oculomotor features. The approach offers a scalable, ecologically valid, and low-cost solution for preliminary mental health screening.

0 citationsRead paper