The integration of Multimodal Artificial Intelligence (MAI) in healthcare has become increasingly vital, as healthcare systems and technologies evolve toward more personalized, precise, and efficient patient care. MAI combines data from various sources, such as imaging, electronic health records, wearable sensors, and genomics, offering a holistic view of patient health. This approach enables more accurate diagnostics, predictive modeling, and treatment personalization, making it a cornerstone of modern healthcare innovation.

However, developing and deploying Multimodal AI in healthcare presents significant challenges. The heterogeneity of data from different modalities—ranging from structured clinical data to unstructured text, images, and physiological signals—requires sophisticated integration methods. Moreover, each data source presents unique issues related to data quality, missing information, standardization, and interpretability. Ethical concerns such as data privacy, bias in AI models, and ensuring equitable access to AI-driven tools also present ongoing challenges.

The special track Multimodal Artificial Intelligence in Healthcare aims to showcase recent advances in Multimodal Learning approaches for healthcare.

In particular, the track will focus on integrating heterogeneous data sources for diverse applications and fostering collaboration and knowledge sharing among researchers with expertise in data fusion techniques and multimodal learning.

The Special Track is organized in collaboration with the Future AI Research (FAIR) Project.

Topics of interest

Example areas include but are not limited to:

  • Data Fusion Techniques for Healthcare: Novel algorithms and methods for integrating heterogeneous data sources such as imaging, genomics, EHRs, and wearable sensors.
  • Multimodal AI for Disease Diagnosis and Prognosis: AI-driven approaches combining various data types for more accurate diagnosis, disease progression prediction, and personalized treatment strategies.
  • Natural Language Processing (NLP) and Multimodal Data Integration: The use of NLP to extract insights from clinical notes and combine them with structured and unstructured data (e.g., EHRs, imaging) for improved decision support.
  • Ethical and Fairness Challenges in Multimodal AI: Addressing bias, data privacy, and the ethical challenges that arise when using multimodal datasets in healthcare applications.
  • Multimodal AI for Personalized Medicine: Leveraging AI to integrate genomic, phenotypic, and clinical data for individualized treatment plans and drug discovery.
  • AI in Multimodal Medical Imaging: Techniques for combining imaging data (e.g., MRI, CT, X-ray) with other modalities for enhanced diagnostic accuracy and clinical insights.
  • Multimodal AI in Remote and Telemedicine Applications: AIdriven integration of data from telemedicine platforms, remote sensors, and patient-reported outcomes for long-distance clinical care.
  • Resilient Multimodal Artificial Intelligence: Developing systems that operate effectively in challenging, noisy, incomplete, and uncertain real-world biomedical settings.
  • Explainable AI (XAI) in Multimodal Healthcare Systems: Methods for enhancing the transparency and interpretability of multimodal AI models, ensuring that clinicians and patients can trust AI-driven decisions.
  • Performance Evaluation: Methods and metrics for assessing the performance of multimodal learning models in biomedicine.

Expected types of contributions

Submitted papers must be unpublished and not considered elsewhere for publication. Submissions will undergo a rigorous review process handled by the Technical Program Committee. This Special Track accepts two types of submissions:

  • Regular papers: The length of the contribution is limited to 6 pages, but it is possible to extend the paper length up to 8 pages by paying for each extra page. Check the conference website for further information (https://2025.cbms-conference.org/)
  • Short papers: The length of the contribution is limited to 4 pages and no less than 3 pages, not being possible to extend the paper length. The duration of the oral presentation of short posters will be less than regular ones.

Organizers

  • Prof. Consuelo Gonzalo-Martín, Ph.D., Universidad Politécnica de Madrid, Spain
  • Prof. Angel Mario García-Pedrero, Ph.D., Universidad Politécnica de Madrid, Spain
  • Eng. Michela Gravina, Ph.D., University of Naples Federico II, Italy
  • Eng. Antonio Galli, Ph.D., University of Naples Federico II, Italy
  • Eng. Valerio Guarrasi, Ph.D., University Campus Bio-Medico of Rome, Italy
  • Eng. Meryeme Boumahdi, Universidad Politécnica de Madrid, Spain

Program Committee

  • Vincenzo Moscato, University of Naples Federico II, Italy
  • Carlo Sansone, University of Naples Federico II, Italy
  • Vito Paolo Pastore, University of Genova, Italy
  • Giuseppe Pontillo, Vrije Universiteit Amsterdam, The Netherlands
  • Ferrán Marques. Universitat Politécnica de Catalunya, Spain
  • Antonino Ferraro, Pegaso University, Italy
  • Marco Postiglione, Northwestern University, United States
  • Valerio La Gatta, Northwestern University, United States
  • Domiziana Santucci, University Campus Bio-Medico Roma, Italy
  • Simona Parisi, Università della Campania Luigi Vanvitelli, Italy

Contact

For more information, please visit the IEEE CBMS 2025 website at https://2025.cbms-conference.org/ or contact the track chairs at michela [dot] gravina [at] unina [dot] it.

Call for papers

You can download Call for papers here.