PurposeThe purpose of this study was to evaluate image quality and status of lymph nodes in laryngeal and hypopharyngeal squamous cell carcinoma (SCC) patients using spectral CT imaging.Materials and MethodsThirty-eight patients with laryngeal and hypopharyngeal SCCs were scanned with spectral CT mode in venous phase. The conventional 140-kVp polychromatic images and one hundred and one sets of monochromatic images were generated ranging from 40 keV to 140 keV. The mean optimal keV was calculated on the monochromatic images. The image quality of the mean optimal keV monochromatic images and polychromatic images was compared with two different methods including a quantitative analysis method and a qualitative analysis method. The HU curve slope (λHU) in the target lymph nodes and the primary lesion was calculated respectively. The ratio of λHU was studied between metastatic and non-metastatic lymph nodes group.ResultsA total of 38 primary lesions were included. The mean optimal keV was obtained at 55±1.77 keV on the monochromatic images. The image quality evaluated by two different methods including a quantitative analysis method and a qualitative analysis method was obviously increased on monochromatic images than polychromatic images (p<0.05). The ratio of λHU between metastatic and non-metastatic lymph nodes was significantly different in the venous phase images (p<0.05).ConclusionThe monochromatic images obtained with spectral CT can be used to improve the image quality of laryngeal and hypopharyngeal SCC and the N-staging accuracy. The quantitative ratio of λHU may be helpful for differentiating between metastatic and non-metastatic cervical lymph nodes.

In response to negative impacts such as personal and property safety hazards caused by drivers being distracted while driving on the road, this article proposes a driver’s attention state-detection method based on the improved You Only Look Once version five (YOLOv5). Both fatigue and distracted behavior can cause a driver’s attention to be diverted during the driving process. Firstly, key facial points of the driver are located, and the aspect ratio of the eyes and mouth is calculated. Through the examination of relevant information and repeated experimental verification, threshold values for the aspect ratio of the eyes and mouth under fatigue conditions, corresponding to closed eyes and yawning, are established. By calculating the aspect ratio of the driver’s eyes and mouth, it is possible to accurately detect whether the driver is in a state of fatigue. Secondly, distracted abnormal behavior is detected using an improved YOLOv5 model. The backbone network feature extraction element is modified by adding specific modules to obtain different receptive fields through multiple convolution operations on the input feature map, thereby enhancing the feature extraction ability of the network. The introduction of Swin Transformer modules in the feature fusion network replaces the Bottleneck modules in the C3 module, reducing the computational complexity of the model while increasing its receptive field. Additionally, the network connection in the feature fusion element has been modified to enhance its ability to fuse information from feature maps of different sizes. Three datasets were created of distracting behaviors commonly observed during driving: smoking, drinking water, and using a mobile phone. These datasets were used to train and test the model. After testing, the mAP (mean average precision) has improved by 2.4% compared to the model before improvement. Finally, through comparison and ablation experiments, the feasibility of this method has been verified, which can effectively detect fatigue and distracted abnormal behavior.

No abstract

Generalized cross-spring pivots (CSPs) are widely used as revolute joints in precision machinery. However, pseudo-rigid-body (PRB) models cannot capture the parasitic motions of a generalized CSP exactly under combined loads; moreover, the characteristic parameters used in PRB methods must be recomputed using optimization techniques. In this study, we develop two simple and accurate PRB models for generalized CSPs. First, a PRB method for a beam is developed based on the beam constraint model and the instantaneous center model, where the beam is modeled as two rigid links joined at a pivot via a torsion spring. Subsequently, two PRB models of the generalized CSP, comprising a four-bar model for accuracy and a pin-joint model for stiffness, are constructed based on a kinematic analysis using the proposed PRB method. A deflection characteristic analysis is then conducted to determine the relationship between the proposed model and the existing models. Finally, the PRB models for the pivot under the action of combined loads are validated via finite element analysis. The error evaluation indicates that the proposed PRB models are more accurate than the results from existing methods. The PRB models proposed here can be used in parametric design of compliant mechanisms.