新乡医学院 2021 年教育教学改革研究项目立项课题
1.The Third Affiliated Hospital of Xinxiang Medical University;2.Xinxiang Medical University, College of Medical Engineering Xinxian;3.Xinxiang Medical university, college of medical engineering Xinxiang
爬虫是一类收集信息的自动化程序，当前越来越多的领域都在使用爬虫收集目标信息。由于 Python 具有可快速迭代的特性，在主要内容为图像处理与人工智能的医学影像中得到了广泛的应用。为了在保证程序运行效率的同时，减轻训练模型所需数据为计算机存储带来的压力，采用了能够大幅提高程序运行效率的异步式程序，并使用暂态文件保存数据。结果表明，异步式程序、暂态存储程序的运行效率分别是单线程的 4.722 倍、1.433 倍，在医学影像模型训练中使用爬虫可以降低对计算机存储性能的要求。
Crawlers are automated programs that collect information and are increasingly used in a wide range of fields. Python have been widely used in medical imaging, which mainly involves image processing and artificial intelligence, due to the rapid iteration capability of Python. In order to ensure the efficiency of program execution and reduce the pressure on computer storage caused by the required data for training models, an asynchronous design program that can significantly improve the efficiency of program execution is adopted, and transient files are used to store data. The results showed that the efficiency of asynchronous programs and transient storage programs were 4.722 times and 1.433 times higher than that of single threads, respectively. Using crawlers in medical imaging model training can reduce the requirements for computer storage performance.