By in situ hybridization histochemistry, the changes of preprotachykinin (PPT) mRNA expression were examined in the neurons of adjacent thoracal dorsal root ganglion (DRG) after a strong electric stimulation to an int...By in situ hybridization histochemistry, the changes of preprotachykinin (PPT) mRNA expression were examined in the neurons of adjacent thoracal dorsal root ganglion (DRG) after a strong electric stimulation to an intact dorsal cutaneous branch and the cut distal part of left T 9 thoracal spinal nerve of rat. There was a significant increase of the number of neurons expressing PPT mRNA in the ipsilateral T 8, T 9 and T 10 DRG of the animals given electric stimulation to intact spinal nerve branch 24 h after the electric stimulation. The same increase was found in the ipsilateral T 8 and T 10 DRG of the animals given electric stimulation to the distal part of spinal nerve branch. While no change was found in the DRG of the contralateral side of these animals. The present results showed that the antidromic electric stimulation strengthened the biosynthesis of PPT mRNA in adjacent DRG. These findings suggested that there was information transmission across segments between two sensory nerve endings and some bioactive substances such as SP might play important roles in the information transmission across segments of spinal cord.展开更多
[目的/意义]根系是植物组成的重要部分,其生长发育至关重要。根系图像分割是根系表型分析的重要方法,受限于图像质量、复杂土壤环境、低效传统方法,根系图像分割存在一定挑战。[方法]为提高根系图像分割的准确性和鲁棒性,本研究以UNet...[目的/意义]根系是植物组成的重要部分,其生长发育至关重要。根系图像分割是根系表型分析的重要方法,受限于图像质量、复杂土壤环境、低效传统方法,根系图像分割存在一定挑战。[方法]为提高根系图像分割的准确性和鲁棒性,本研究以UNet模型为基础,提出了一种多尺度特征提取根系分割算法,并结合数据增强和迁移学习进一步提高改进UNet模型的泛化性和通用性。首先,获取棉花根系单一数据集和开源多作物混合数据集,基于单一数据集的消融试验测试多尺度特征提取模块(Conv_2+Add)的有效性,与UNet、PSPNet、SegNet、DeeplabV3Plus算法对比验证其优势。基于混合数据集验证改进算法(UNet+Conv_2+Add)在迁移学习的优势。[结果和讨论] UNet+Conv_2+Add相比其他算法(UNet、PSPNet、SegNet、DeeplabV3Plus),mIoU、mRecall和根系F_1调和平均值分别为81.62%、86.90%和78.39%。UNet+Conv_2+Add算法的迁移学习相比于普通训练在根系的交并比(Intersection over Union,IoU)值提升1.25%,根系的Recall值提升1.79%,F_1调和平均值提升0.92%,且模型的整体收敛速度快。[结论]本研究采用的多尺度特征提取策略能准确、高效地分割根系,为作物根系表型研究提供重要的研究基础。展开更多
文摘By in situ hybridization histochemistry, the changes of preprotachykinin (PPT) mRNA expression were examined in the neurons of adjacent thoracal dorsal root ganglion (DRG) after a strong electric stimulation to an intact dorsal cutaneous branch and the cut distal part of left T 9 thoracal spinal nerve of rat. There was a significant increase of the number of neurons expressing PPT mRNA in the ipsilateral T 8, T 9 and T 10 DRG of the animals given electric stimulation to intact spinal nerve branch 24 h after the electric stimulation. The same increase was found in the ipsilateral T 8 and T 10 DRG of the animals given electric stimulation to the distal part of spinal nerve branch. While no change was found in the DRG of the contralateral side of these animals. The present results showed that the antidromic electric stimulation strengthened the biosynthesis of PPT mRNA in adjacent DRG. These findings suggested that there was information transmission across segments between two sensory nerve endings and some bioactive substances such as SP might play important roles in the information transmission across segments of spinal cord.
文摘[目的/意义]根系是植物组成的重要部分,其生长发育至关重要。根系图像分割是根系表型分析的重要方法,受限于图像质量、复杂土壤环境、低效传统方法,根系图像分割存在一定挑战。[方法]为提高根系图像分割的准确性和鲁棒性,本研究以UNet模型为基础,提出了一种多尺度特征提取根系分割算法,并结合数据增强和迁移学习进一步提高改进UNet模型的泛化性和通用性。首先,获取棉花根系单一数据集和开源多作物混合数据集,基于单一数据集的消融试验测试多尺度特征提取模块(Conv_2+Add)的有效性,与UNet、PSPNet、SegNet、DeeplabV3Plus算法对比验证其优势。基于混合数据集验证改进算法(UNet+Conv_2+Add)在迁移学习的优势。[结果和讨论] UNet+Conv_2+Add相比其他算法(UNet、PSPNet、SegNet、DeeplabV3Plus),mIoU、mRecall和根系F_1调和平均值分别为81.62%、86.90%和78.39%。UNet+Conv_2+Add算法的迁移学习相比于普通训练在根系的交并比(Intersection over Union,IoU)值提升1.25%,根系的Recall值提升1.79%,F_1调和平均值提升0.92%,且模型的整体收敛速度快。[结论]本研究采用的多尺度特征提取策略能准确、高效地分割根系,为作物根系表型研究提供重要的研究基础。