Selected work

Representative Papers

Combine Virtual Reality and Machine-Learning to Identify the Presence of Dyslexia: A Cross-Linguistic Approach

Sep 02, 2025

This study addresses the challenge of automating cross-lingual (Italian/Spanish) dyslexia screening among university students, where linguistic and cultural variability complicates conventional clinical assessment. Method: We propose the first VR-AI integrated framework for dyslexia evaluation, leveraging silent reading tasks in immersive virtual reality to collect fine-grained behavioral and self-esteem data, coupled with supervised machine learning. Task completion time serves as the primary discriminative feature, while group-level differences are statistically validated via independent-samples t-tests and Mann–Whitney U tests. Contribution/Results: The model achieves 87.5% accuracy on Italian samples, 66.6% on Spanish samples, and 75.0% on the combined cohort. Critically, this work provides the first empirical evidence that VR-captured, language-agnostic reading behaviors—collected non-invasively and outside clinical settings—can robustly support AI-driven cross-lingual dyslexia identification. It establishes a scalable, low-cost, culturally adaptable paradigm for early, non-clinical screening.

0 citationsRead paper

Automatic Identification and Description of Jewelry Through Computer Vision and Neural Networks for Translators and Interpreters

Aug 30, 2025

Jewelry exhibits high intra-class visual diversity and structural complexity, making accurate identification and descriptive translation challenging for non-experts—particularly in cross-lingual terminology alignment. To address this, we propose a translation-oriented, fine-grained visual description framework for jewelry: a three-level natural language generation architecture (category → structure → material/technique) that emulates expert cognitive reasoning. Our method integrates CNN-based visual feature extraction with an encoder-decoder architecture, augmented by image captioning and object detection to enable end-to-end jewelry recognition and multi-granularity textual description generation. Evaluated on a curated jewelry image dataset, our model achieves 90.3% description accuracy, substantially outperforming baseline approaches. This work represents the first systematic application of fine-grained vision-language generation to support jewelry translation, establishing a scalable paradigm for domain-specific visual–linguistic co-understanding.

0 citationsRead paper

Design and evaluation of a serious game in virtual reality to increase empathy towards students with phonological dyslexia

Aug 28, 2025

Non-dyslexic individuals—particularly educators and the general public—often lack empathy and evidence-informed understanding of phonological dyslexia, hindering effective support for affected students. Method: This study designed and empirically validated a virtual reality (VR)-based serious game intervention to enhance empathic understanding of phonological dyslexia. The immersive VR experience simulates authentic reading challenges faced by individuals with phonological dyslexia, targeting educators and lay audiences to deepen comprehension of the disorder’s neurocognitive basis and the necessity of evidence-based instructional accommodations. Contribution/Results: Using standardized psychometric instruments (e.g., the Interpersonal Reactivity Index), pre- and post-intervention assessments with 101 participants revealed a statistically significant 20% average increase in empathy scores (p < 0.001), outperforming conventional awareness campaigns. This work establishes an empirically supported, scalable, and innovative technological approach to improving societal understanding of dyslexia.

0 citationsRead paper

Fostering Inclusion: A Virtual Reality Experience to Raise Awareness of Dyslexia-Related Barriers in University Settings

Feb 20, 2025Electronics

This study addresses the widespread lack of awareness and inadequate inclusive support for dyslexia in higher education institutions. To bridge this gap, we designed and empirically validated a novel first-person immersive virtual reality (VR) simulation system developed in Unity. The system embodies core dyslexic cognitive challenges—including dynamic text drifting, spatial-logical orientation disruption, and absence of contextual scaffolding—within an interactive virtual campus environment. Supporting both walking and teleportation locomotion, it transforms abstract neurocognitive difficulties into tangible, spatially grounded tasks. Evaluated across 30+ universities with faculty and students, the system significantly enhanced participants’ conceptual understanding and empathetic engagement with dyslexia: 92% rated it an effective tool for advancing inclusive education. By moving beyond conventional awareness-raising approaches, this work establishes a scalable technical framework and evidence-based paradigm for accessibility-oriented pedagogy.

0 citationsRead paper

Recent publications

Latest Papers

Combine Virtual Reality and Machine-Learning to Identify the Presence of Dyslexia: A Cross-Linguistic Approach

Sep 02, 2025

This study addresses the challenge of automating cross-lingual (Italian/Spanish) dyslexia screening among university students, where linguistic and cultural variability complicates conventional clinical assessment. Method: We propose the first VR-AI integrated framework for dyslexia evaluation, leveraging silent reading tasks in immersive virtual reality to collect fine-grained behavioral and self-esteem data, coupled with supervised machine learning. Task completion time serves as the primary discriminative feature, while group-level differences are statistically validated via independent-samples t-tests and Mann–Whitney U tests. Contribution/Results: The model achieves 87.5% accuracy on Italian samples, 66.6% on Spanish samples, and 75.0% on the combined cohort. Critically, this work provides the first empirical evidence that VR-captured, language-agnostic reading behaviors—collected non-invasively and outside clinical settings—can robustly support AI-driven cross-lingual dyslexia identification. It establishes a scalable, low-cost, culturally adaptable paradigm for early, non-clinical screening.

0 citationsRead paper

Automatic Identification and Description of Jewelry Through Computer Vision and Neural Networks for Translators and Interpreters

Aug 30, 2025

Jewelry exhibits high intra-class visual diversity and structural complexity, making accurate identification and descriptive translation challenging for non-experts—particularly in cross-lingual terminology alignment. To address this, we propose a translation-oriented, fine-grained visual description framework for jewelry: a three-level natural language generation architecture (category → structure → material/technique) that emulates expert cognitive reasoning. Our method integrates CNN-based visual feature extraction with an encoder-decoder architecture, augmented by image captioning and object detection to enable end-to-end jewelry recognition and multi-granularity textual description generation. Evaluated on a curated jewelry image dataset, our model achieves 90.3% description accuracy, substantially outperforming baseline approaches. This work represents the first systematic application of fine-grained vision-language generation to support jewelry translation, establishing a scalable paradigm for domain-specific visual–linguistic co-understanding.

0 citationsRead paper

Design and evaluation of a serious game in virtual reality to increase empathy towards students with phonological dyslexia

Aug 28, 2025

Non-dyslexic individuals—particularly educators and the general public—often lack empathy and evidence-informed understanding of phonological dyslexia, hindering effective support for affected students. Method: This study designed and empirically validated a virtual reality (VR)-based serious game intervention to enhance empathic understanding of phonological dyslexia. The immersive VR experience simulates authentic reading challenges faced by individuals with phonological dyslexia, targeting educators and lay audiences to deepen comprehension of the disorder’s neurocognitive basis and the necessity of evidence-based instructional accommodations. Contribution/Results: Using standardized psychometric instruments (e.g., the Interpersonal Reactivity Index), pre- and post-intervention assessments with 101 participants revealed a statistically significant 20% average increase in empathy scores (p < 0.001), outperforming conventional awareness campaigns. This work establishes an empirically supported, scalable, and innovative technological approach to improving societal understanding of dyslexia.

0 citationsRead paper

Fostering Inclusion: A Virtual Reality Experience to Raise Awareness of Dyslexia-Related Barriers in University Settings

Feb 20, 2025Electronics

This study addresses the widespread lack of awareness and inadequate inclusive support for dyslexia in higher education institutions. To bridge this gap, we designed and empirically validated a novel first-person immersive virtual reality (VR) simulation system developed in Unity. The system embodies core dyslexic cognitive challenges—including dynamic text drifting, spatial-logical orientation disruption, and absence of contextual scaffolding—within an interactive virtual campus environment. Supporting both walking and teleportation locomotion, it transforms abstract neurocognitive difficulties into tangible, spatially grounded tasks. Evaluated across 30+ universities with faculty and students, the system significantly enhanced participants’ conceptual understanding and empathetic engagement with dyslexia: 92% rated it an effective tool for advancing inclusive education. By moving beyond conventional awareness-raising approaches, this work establishes a scalable technical framework and evidence-based paradigm for accessibility-oriented pedagogy.

0 citationsRead paper