Editorial: Computational intelligence for multimodal biomedical data fusion

This Research Topic, Computational Intelligence for Multimodal Biomedical Data Fusion, was established to explore innovative computational intelligence techniques capable of integrating multiple biomedical data modalities. By leveraging advanced machine learning, deep learning architectures, and intelligent data fusion strategies, researchers are developing models that extract meaningful patterns from diverse biomedical datasets. Such systems are increasingly enabling early disease detection, accurate diagnosis, improved patient monitoring, and the advancement of personalized healthcare.A central motivation behind this Research Topic was to address key challenges in multimodal data integration. Biomedical datasets often differ in temporal resolution, noise characteristics, format, and scale. Furthermore, ensuring interpretability, reliability, and security of AI-based healthcare systems remains essential for clinical adoption. The articles included in this Topic demonstrate innovative solutions that address these challenges while presenting practical applications across different biomedical domains.Across the five contributions included in this Research Topic, several common methodological threads emerge that highlight the evolving landscape of computational intelligence for multimodal biomedical data fusion. A central theme is the integration of heterogeneous biomedical modalities-such as physiological signals, medical imaging, electronic health records, and other clinical data-through advanced machine learning and deep learning architectures. The studies collectively demonstrate the growing adoption of multimodal learning frameworks that combine complementary data sources to produce richer feature representations and improve predictive performance in healthcare analytics. Techniques such as ensemble learning, graph-based modeling, deep neural networks, and fusion-driven time-series analysis illustrate the diversity of computational strategies used to capture complex relationships within biomedical datasets. These approaches reflect a broader shift toward data-driven healthcare systems where intelligent fusion of multiple modalities enables more comprehensive patient modeling and supports improved diagnostic and prognostic decision-making.While the contributions in this Research Topic demonstrate promising advances in multimodal biomedical analytics, several translational challenges remain for their effective adoption in realworld healthcare environments. One key issue is the heterogeneity of biomedical data modalities, which often differ in format, scale, temporal resolution, and data quality. Integrating such diverse data sources requires robust preprocessing, standardized data representations, and scalable computational frameworks. In addition, the deployment of AI-driven healthcare systems must address concerns related to model validation, interpretability, and reliability to ensure that predictions can be trusted by clinicians and healthcare practitioners. Data privacy and security are also critical considerations, particularly when dealing with sensitive patient information across distributed healthcare infrastructures. Consequently, future research must focus not only on improving predictive performance but also on ensuring transparency, ethical data governance, and robust clinical validation to enable safe and effective integration of multimodal AI systems into clinical workflowsIn summary, this Research Topic provides valuable insights into the emerging landscape of computational intelligence for multimodal biomedical data fusion. The contributions demonstrate how innovative AI-driven methodologies can enhance biomedical data analysis and support improved healthcare outcomes. We hope that the research presented in this collection will inspire further interdisciplinary collaboration and encourage continued advancements in intelligent biomedical data integration.