基于物联网技术的肺部辅助诊断平台的设计与实现

Design and Implementation of a Lung-assisted Diagnosis Platform Based on the Internet of Things Technology

本文设计的肺部辅助诊断平台以医用CT机为载体,结合人工智能技术和云存储技术,旨在提升医院诊疗效率、关注医疗影像隐私保护和影像高效存储问题。本文提出了改进后的3D-U-Net神经网络,用于完成肺部影像的分割任务,实验结果显示Dice系数达到了0.956的高水平。在影像数据存储与隐私保护方面,本文运用了铅笔盒加密隐写技术将切片进行加密隐私保护,获得了更高的容量为3.0bpp,平均PSNR为43.25dB。

In this paper, the lung auxiliary diagnostic platform is designed to take the medical CT machine as a carrier, combined with artificial intelligence technology and cloud storage technology, aiming to improve the diagnosis and treatment efficiency in hospitals, focusing on the privacy protection of medical images and the problem of efficient storage of images. Improved 3D-U-Net neural network is proposed in this paper to complete the segmentation task of lung images, and the experimental results show that the Dice coefficient reaches a high level of 0.956. In terms of image data storage and privacy protection, this paper employs pencil box cryptographic steganography to encrypt the slices for privacy protection, obtaining a higher capacity of 3.0 bpp and an average PSNR of 43.25 dB.