As artificial intelligence advances, speech-emotion recognition (SER) has become a critical research area. Traditional SER methods typically rely on homogeneous domain data for training and testing. This practice requires adaptation when confronted with real-world data's heterogeneous linguistic, methodological, and speaker-related attributes. These variances can degrade the accuracy and generalization of SER models. To address this gap, we introduce a novel domain adaptation technique, multi domain emotion recognition enhancement (MDERE), which utilizes a non-negative matrix to reduce the inflexibility of the conventional binary label matrix for source domain data. This process yields a label matrix that better adapts to the nuances of the source labels while preserving their original structure. This framework refines SER methods by fine-tuning a transformation matrix for enhanced emotion discrimination. Elastic net regularization, which combines L1 and L2 penalties, enriches the transformation matrix, selectively emphasizing relevant features to enhance the robustness of emotion detection. The framework constructs customized similarity and dissimilarity graphs to reconcile the differences between source and target domains, enabling nuanced cross-domain data analysis. Extensive testing on multiple cross-domain SER tasks has shown that MDERE substantially improves recognition accuracy, surpassing several state-of-the-art algorithms. These results demonstrate MDERE's ability to effectively align domain variations enhances the generalizability of SER systems.

Recent advances in electrochemical devices have sparked exciting opportunities in the healthcare, environment, and food industries. These devices can be fabricated at low costs and are capable of multiplex monitoring. This overcomes challenges presnted in traditional sensors for biomolecules and provides us a unique gateway toward comprehensive analyses. The advantages of electrochemical sensors are derived from their direct integration with electronics and their high selectivity along with sensitivity to sense a wide range of ionic analytes at an economical cost. This review paper aims to summarize recent innovations of a wide variety of electrochemical sensors for ionic analytes for health care and industrial applications. Many of these ionic analytes are important biomarkers to target for new diagnostic tools for medicine, food quality monitoring, and pollution detection. In this paper, we will examine various fabrication techniques, sensing mechanisms, and will also discuss various future opportunities in this research direction.

Tooth decay is a chronic disease resulting in pain, infection, and tooth loss. This illness is common because many factors, such as poor oral hygiene, sugar consumption, and microbial flora, contribute to dental cavities. Untreated or undetected tooth decay often escalates to a more severe stage, emphasizing the importance of early detection and intervention. We propose a portable, low-cost, and ergonomic optoelectronic device to provide a possible solution for the early detection of dental cavities before annual or regular dental checkups for the first time. This device integrates mini cameras on top of a dental impression tray to capture images of the teeth, and the photos can be transmitted via Wi-Fi to the cloud for real-time cavities detection through a YOLO (You Only Look Once) algorithm that is based on a convolutional neural network (CNN). Our results show the precision, recall, and (mean average precision)@0.5:0.95 for YOLOv5 (0.72, 0.70, 0.75), YOLOv6 (0.59, 0.50, 0.58), and YOLOv7 (0.93, 0.94, 0.82). We also compared the YOLO algorithm with traditional techniques such as support vector machine and k-nearest neighbor algorithms. This intra-oral cavity detection system paves the way for early detection of dental cavities with quick accessibility and affordable cost. We foresee that this optoelectronic device will play a role in advancing biomedical technologies, ultimately promoting the long-term well-being of individuals.

Nano-dendritic structures have gained increasing popularity in electrochemical sensors. However, it is still rare to generate a 3-dimensional model in a short period of time to understand the structure-function relationship of the sensors. Here, we report the construction of a 3-dimensional model for nano-dendritic, metallic structures frequently grown on top of bioelectronics. This is achieved by merging two sources of 2-dimensional dendritic information, which includes top-view images from scanning electron microscopy and side-view visualization from Monte-Carlo simulations. The microscopy images provide the boundary conditions to tune the Monte-Carlo simulations to construct the 3-dimensional dendritic morphology. We validated the 3-dimensional model by comparing the dendritic area densities predicted via this model with those computed from microscopy images. Additionally, tuning the simulation parameters in the 3-dimensional model can be used to find the optimized dendritic density, which is an essential indicator for sensitivity enhancement. The success of this model provides a means to understand the sensitivity limits of bio-electronics through dendritic growth without the need for the time-consuming sensor fabrication and testing. Further, our SEM-guided Monte-Carlo technique provides a dendritic model with a significant resemblance to experimental images. It possesses the potential for applications in 3-dimensional morphological investigations for future biosensor design.