Workshop Overview
Extended Reality technologies are increasingly being explored as interactive, immersive, and data-rich environments for neurocognitive assessment, rehabilitation, and health-related intervention. Compared with traditional tools, XR systems can reproduce complex real-world situations while capturing fine-grained behavioral data — including movement patterns, task performance, gaze, posture, interaction traces, physiological signals, and subjective responses.
This opens new opportunities for observing cognition, motor behavior, perception, attention, memory, spatial orientation, pain, fatigue, and functional performance in controlled yet ecologically meaningful scenarios. However, the field still lacks shared methodological frameworks for designing XR-based tasks, validating behavioral measures, interpreting telemetry data, and translating immersive interaction patterns into clinically meaningful digital biomarkers.
XR-NeuroHealth'26 brings together XR researchers, HCI scholars, clinicians, rehabilitation experts, cognitive scientists, AI/data science researchers, and digital health practitioners. The workshop positions XR not merely as a visualization technology, but as an ecologically valid, data-rich environment for observing, supporting, and measuring neurocognitive and rehabilitation-related processes.
Submission
All submissions must be written in English and formatted according to the IEEE Computer Society VGTC conference format (templates: LaTeX · Word). Accepted papers are intended for publication in the ISMAR 2026 Adjunct Proceedings and IEEE Xplore, subject to the final publication policy of ISMAR 2026.
Each submission will receive at least two reviews by program committee members. Evaluation criteria include relevance, originality, methodological clarity, interdisciplinary value, ethical awareness, potential to stimulate discussion, and alignment with ISMAR topics.
Keynote
Fibromyalgia syndrome remains challenging to diagnose because of its heterogeneous symptoms and clinical overlap with other rheumatic and musculoskeletal diseases. In 2024, a proof-of-concept study showed that large language model analysis of patient narratives could distinguish fibromyalgia from other chronic pain conditions with 87% accuracy and an AUC of 0.86. Building on these results, the 2026 fAI-BRO study integrated psycholinguistic transcript analysis with video-derived visual descriptors.
In 100 patients, this multimodal system achieved 91% accuracy, 94% sensitivity, 88% specificity and an AUC of 0.96. All inflammatory arthritis and mechanical conditions were correctly classified, while false positives were restricted to dermatomyositis and polymyalgia rheumatica, revealing an important overlap in muscle-pain narratives.
Together, these studies demonstrate the evolution from transcript-only analysis to explainable, privacy-preserving multimodal decision support. Although fAI-BRO may facilitate earlier recognition and referral, it should be used as a human-in-the-loop screening tool and not as a replacement for clinical assessment and subsequent diagnosis. Prospective external validation remains necessary before implementation in routine practice.
Venerito V, et al. RMD Open 2026;12:e006995.
Venerito V, Iannone F. RMD Open 2024;10:e004367.
Organizing Committee
Department of Computer Science, Italy
Institute of New Imaging Technologies (INIT), Spain
Centre de Visualització Interactiva (CEVI), Spain
Department of Computer Science, Italy
Workshop Schedule
Program Committee
Covering XR, HCI, neurorehabilitation, clinical assessment, cognitive science, digital health, AI, accessibility, and ethics.