Abstract
This research topic explores the advancements in artificial intelligence (AI) in the realm of bioimaging and bio-signal processing. It encompasses a diverse range of studies spanning various bio-signal modalities and subspecialties. These modalities aim to monitor physiological events in patients, enabling applications such as fetal distress diagnosis, bone mineral density prediction, portable ECG measurements, pulse wave velocity estimation, drowsiness detection, knowledge extraction, and semantic understanding of bio-signals. The studies included in this compilation were specifically chosen for their focus on AI applications that have the potential to revolutionize these respective fields.Yefei Zhang et al. [1] introduced a novel multi-modal information fusion (MMIF) framework aimed at enhancing fetal well-being evaluation in late pregnancy and preventing sudden fetal death. This study encompasses several innovative contributions. Firstly, the authors employed the Category Constrained-Parallel ViT (CCPViT) approach to model unimodal representations for Gramian Angular Field (GAF)-based 2D images and label texts. This allowed for effective representation learning within each modality. Secondly, to address the challenge of misalignment between the 33 modalities, the authors proposed the Multimodal Representation Alignment Network (MRAN