Special Track on Generalizable & Explainable AI for the Care Continuum of Neurodegenerative Diseases

June 03-05-2026, Limassol, Cyprus

Advancing AI for Neurodegenerative Disease Care

Artificial intelligence (AI) is reshaping how we detect, stage, and monitor neurodegenerative diseases such as Alzheimer’s, Parkinson’s, Amyotrophic Lateral Sclerosis (ALS), and related conditions. From imaging to wearable devices and mobile platforms, digital biomarkers enable earlier detection, continuous real-world monitoring, and truly personalized interventions.

Yet translating these advances into clinical workflows remains challenging as models often behave as black boxes; performance can degrade under domain shift; multimodal data are messy, incomplete, or siloed; and regulatory-grade evidence requires calibrated uncertainty, fairness assessments, transparency, and reproducibility. For clinic-to-home care pathways, methods must be interpretable and robust across heterogeneous devices, sites, and populations, while preserving privacy and aligning with clinician workflows.

This special track aims to bring together researchers to advance clinically interpretable AI for digital biomarkers of neurodegeneration across the clinic–home continuum.

Research Themes

(Including but not limited to)

Interpretability & Uncertainty

explainable AI, causal reasoning, calibrated confidence.

Robustness & Domain Shift

generalization across scanners, sites, and sensors.

Annotation Scarcity

weakly or self-supervised learning; label noise mitigation.

Multimodality & Missing Data

integrating MRI, PET, EEG, speech, gait, ocular, and wearable data.

Fairness & Ethics

bias assessment, human-in-the-loop design, equitable performance.

Topics of Interest

(not limited to)

Interpretable Modeling for NDDs

Concept-based, prototype, counterfactual, causal, and rules-based approaches; global vs. local explanations for model transparency.

Learning with Limited or Weak Labels

Self-, semi-, and multiple-instance learning; curriculum and active learning; training under label noise and uncertain annotations.

Calibrated & Trustworthy AI

Uncertainty quantification, reliability diagrams, conformal prediction, and detection of out-of-distribution samples or domain shifts.

Domain Generalization & Adaptation

Site-robust modeling, harmonization, synthetic data augmentation, and test-time adaptation for cross-center reliability.

Digital Biomarkers & Sensing

Extraction and validation from speech, language, handwriting, gait and kinematics, ocular or retinal metrics, biosignals (EEG, ECG, EMG, GSR), and wearable or smartphone-based sensing — including passive monitoring.

Human-in-the-Loop & Interactive AI

Clinician- and patient-facing explanations, iterative model refinement, and UI/UX design to foster trust and adoption.

Medical Imaging for NDDs

Interpretable MRI/PET pipelines for early detection, differential diagnosis, disease staging, progression modeling, and treatment response prediction.

Prospective & Real-World Studies

Integration into clinical workflows (including telemedicine), remote monitoring, adherence analysis, and real-world utility assessment.

Multimodal Fusion & Missing-Modality Learning

Late, early, or intermediate fusion strategies using product-of-experts, mixture-of-experts, graph, or transformer architectures.

Benchmarks, Challenges & Datasets

Multi-site datasets, open-source tools, reproducible baselines, and reports of negative or neutral results with rigorous error analysis.

Proudly Featured at IEEE CBMS 2026 in Limassol, Cyprus

The 39th IEEE International Symposium on Computer-Based Medical Systems (CBMS 2026) will be held in Limassol, Cyprus, bringing together a leading international community at the intersection of medicine and computing. Each year, CBMS convenes clinicians, researchers, engineers, and developers to exchange innovative research and address emerging challenges in computer-based medical systems.

Expected Types of Contributions

Authors are invited to submit original, previously unpublished research papers. Papers should be written in English strictly following IEEE two-column format. For formatting instructions and templates see the IEEE Web page at https://www.ieee.org/conferences/publishing/templates.html. Papers can be submitted by means of EasyChair after the submission page has been released (see conference web page).

The following types of submissions are accepted:

    • Full research papers and experience papers with a maximum length of 6 pages, including references and appendices. It is possible to extend the paper length up to 8 pages by paying for each extra page.
    • Short papers and position papers with a minimum length of 3 pages and a maximum length of 4 pages, including references and appendices. It is not possible to extend the length of short papers.

Submission

Authors should follow the IEEE CBMS 2026 formatting and submission guidelines and select the “GEAI4ND” Special Track at submission time. We welcome:

Original research papers

  • Resource/benchmark papers (datasets, tools, leaderboards, reproducible baselines)
  • Application/clinical studies (pilot, prospective, multi-site, or real-world evaluations)
  • Negative/neutral results with thorough error and ablation analyses
  • Vision/position papers that crystallize standards, reporting, or governance for interpretable digital biomarkers

All submissions will be peer-reviewed for originality, technical quality, interpretability/clinical relevance, rigor of evaluation, and reproducibility.

Expected Audience

Researchers and practitioners in medical AI, biomedical engineering, neurology, neuroimaging, signal processing, HCI, clinical translation, including industry and healthcare partners.

General Deadlines

February 20, 2026 (AoE)

Paper submission deadline

April 10, 2026

Collaegues & Counting

April 24, 2026

Camera-ready due