RESUMEN
BACKGROUND: It is very important to detect mandibular fracture region. However, the size of mandibular fracture region is different due to different anatomical positions, different sites and different degrees of force. It is difficult to locate and recognize fracture region accurately. METHODS: To solve these problems, M3YOLOv5 model is proposed in this paper. Three feature enhancement strategies are designed, which improve the ability of model to locate and recognize mandibular fracture region. Firstly, Global-Local Feature Extraction Module (GLFEM) is designed. By effectively combining Convolutional Neural Network (CNN) and Transformer, the problem of insufficient global information extraction ability of CNN is complemented, and the positioning ability of the model to the fracture region is improved. Secondly, in order to improve the interaction ability of context information, Deep-Shallow Feature Interaction Module (DSFIM) is designed. In this module, the spatial information in the shallow feature layer is embedded to the deep feature layer by the spatial attention mechanism, and the semantic information in the deep feature layer is embedded to the shallow feature layer by the channel attention mechanism. The fracture region recognition ability of the model is improved. Finally, Multi-scale Multi receptive-field Feature Mixing Module (MMFMM) is designed. Deep separate convolution chains are used in this modal, which is composed by multiple layers of different scales and different dilation coefficients. This method provides richer receptive field for the model, and the ability to detect fracture region of different scales is improved. RESULTS: The precision rate, mAP value, recall rate and F1 value of M3YOLOv5 model on mandibular fracture CT data set are 97.18%, 96.86%, 94.42% and 95.58% respectively. The experimental results show that there is better performance about M3YOLOv5 model than the mainstream detection models. CONCLUSION: The M3YOLOv5 model can effectively recognize and locate the mandibular fracture region, which is of great significance for doctors' clinical diagnosis.
Asunto(s)
Fracturas Mandibulares , Humanos , Fracturas Mandibulares/diagnóstico por imagen , Almacenamiento y Recuperación de la Información , Redes Neurales de la Computación , SemánticaRESUMEN
Apricot (Prunus armeniaca L.) cultivars originated in China display a typical S-RNase-based gametophytic self-incompatibility (GSI). 'Katy', a natural self-compatible cultivar belonging to the European ecotype group, was used as a useful material for breeding new cultivars with high frequency of self-compatibility by hybridizing with Chinese native cultivars. In this work, the pollen-S genes (S-haplotype-specific F-box gene, or SFB gene) of 'Katy' were first identified as SFB1 and SFB (8), and the S-genotype was determined as S1 S8. Genetic analysis of 'Katy' progenies under controlled pollination revealed that the stylar S1-RNase and S8-RNase have a normal function in rejecting wild-type pollen with the same S-haplotype, while the pollen grains carrying either the SFB1 or the SFB8 gene are both able to overcome the incompatibility barrier. However, the observed segregation ratios of the S-genotype did not fit the expected ratios under the assumption that the pollen-part mutations are linked to the S-locus. Moreover, alterations in the SFB1 and SFB8 genes and pollen-S duplications were not detected. These results indicated that the breakdown of SI in 'Katy' occurred in pollen, and other factors not linked to the S-locus, which caused a loss of pollen S-activity. These findings support a hypothesis that modifying factors other than the S-locus are required for GSI in apricot.