Search

Multimodal emotion classification (MEC) has been extensively studied in human–computer interaction, healthcare, and other domains. Previous MEC research has utilized identical multimodal annotations (IMAs) to train unimodal models, hindering the learning of effective unimodal representations due to differences between unimodal expressions and multimodal perceptions. Additionally, most MEC fusion techniques fail to consider the unimodal–multimodal inconsistencies. This study addresses two important issues in MEC: learning satisfactory unimodal representations of emotion and accounting for unimodal–multimodal inconsistencies during the fusion process. To tackle these challenges, the authors propose the Two-Stage Conformer-based MEC model (Uni2Mul) with two key innovations: (1) in stage one, unimodal models are trained using independent unimodal annotations (IUAs) to optimize unimodal emotion representations; (2) in stage two, a Conformer-based architecture is employed to fuse the unimodal representations learned in stage one and predict IMAs, accounting for unimodal–multimodal differences. The proposed model is evaluated on the CH-SIMS dataset. The experimental results demonstrate that Uni2Mul outperforms baseline models. This study makes two key contributions: (1) the use of IUAs improves unimodal learning; (2) the two-stage approach addresses unimodal–multimodal inconsistencies during Conformer-based fusion. Uni2Mul advances MEC by enhancing unimodal representation learning and Conformer-based fusion. [ABSTRACT FROM AUTHOR]

The fabrication of various 3D tissue engineering tubular scaffolds with fibrous structures, to assist the human body in rapidly repairing a variety of ailments, is receiving more and more attention. Due to the inefficiency of the majority of fibrous preparation techniques, the question of how to rapidly produce the requisite three-dimensional tubular microfiber scaffold structures has become an urgent problem. In this study, an efficient polymer fiber preparation method was developed, using a high-speed airflow drive. Melt blending of polycaprolactone (PCL), polylactic acid (PLA), and tributyl citrate (TBC), was used for the printing material, to achieve the efficient preparation of tubular microfiber scaffolds with different structures. The scaffold diameter was as small as 2 mm, the wall thickness was up to 100 μm, and the fiber injection efficiency reached 15.48 g/h. By utilizing simulations to optimize the printing parameters and by adjusting the printing settings, it was possible to achieve a controlled fiber diameter in the range of 3 μm to 15 μm. In addition, plasma treatment was applied to the microfibers' surface, to increase their wettability, and the efficiency of the hydrophilic modification was demonstrated. Furthermore, the mechanical property test demonstrated that the fibers have a tensile strength of 1.36 ± 0.16 MPa and a tensile strain of 30.8 ± 3.5%. The radial compressive strain of the tubular scaffold could reach 60% of the original scaffold's diameter. Finally, the in vitro degradation of the fibers at various pH values was tested. The results showed that, under alkaline conditions, the surface of the fibers would be severely crushed and the rate of deterioration would increase. [ABSTRACT FROM AUTHOR]