Human Pose Estimation (HPE) plays a critical role in medical applications, particularly within nursing robotics for patient monitoring. Despite its importance, HPE faces several challenges, including high rates of false positives and negatives, stringent real-time requirements, and limited computational resources, especially in complex backgrounds. In response, we introduce the HP-YOLO model, developed using the YOLOv8 framework, to effectively address these issues. We designed an Enhanced Large Separated Kernel Attention (ELSKA) mechanism and integrated it into the backbone network, thereby improving the model’s effective receptive field and feature separation capabilities, which enhances keypoint detection accuracy in challenging environments. Additionally, the Reparameterized Network with Cross-Stage Partial Connections and Efficient Layer Aggregation Network (RepNCSPELAN4) module was incorporated into the detection head, boosting accuracy in detecting small-sized targets through multi-scale convolution and reparameterization techniques while accelerating inference speed. On the COCO dataset, our HP-YOLO model outperformed existing lightweight methods by increasing average precision (AP) by 4.9%, while using 18% fewer parameters and achieving 1.4× higher inference speed. Our method significantly enhances the real-time performance and efficiency of human pose estimation while maintaining high accuracy, offering an optimal solution for applications in complex environments.

Fatigue significantly reduces productivity and cognitive performance by impairing the concentration and information-processing abilities of individuals. Existing fatigue detection methods often rely on invasive or privacy-disrupting techniques, limiting their applicability in real-world settings. To address these challenges, we propose a non-intrusive fatigue detection method using multidomain features of sitting pressure and machine learning. First, we design a wireless and flexible mat with an embedded pressure sensor array to acquire sitting pressure non-intrusively. Then, we extract multidomain features (time, frequency, and time-frequency domain) and apply feature selection techniques to find an optimal feature subset. We then detect fatigue using well-known machine learning methods based on full domain features and the optimal feature subset, respectively. Experimental results show that utilizing feature selection can improve the performance of fatigue detection and reduce computational complexity. Compared to single classifiers, ensemble learning models require more computational resources but achieve better performance in fatigue detection, with accuracies above 92%. Finally, we use the bootstrap method to evaluate the stability of each model, and the results show that random forest with the subset excels in terms of stability.