Location Algorithm of Conus Medullaris End Based on Convolution Neural Network
for Lumbar Anesthesia
LUO Yan1, LI Xin2, QIN Xiaoyong3, LIANG Jingjing1, GUAN Yanshun2, YANG Jianwei4
1. Department of Anesthesiology, Mianyang Traditional Chinese Medicine Hospital, Mianyang Sichuan 621000, China;
2. Mudanjiang Traditional Chinese Medicine Hospital, Mudanjiang Heilongjiang 157000, China;
3. Sichuan Jiuzhou Electrical Appliance Group Co., Ltd., Mianyang Sichuan 621000, China;
4. Nangang Branch of Heilongjiang Academy of Traditional Chinese Medicine, Harbin Heilongjiang 150006, China
Abstract:There is no visual technical support for conventional conus medullaris end anesthesia, mainly through the puncture
location of the anesthesiologist according to the osseous characteristics of the patient’s body surface. The degree of difficulty is
closely related to the patient’s current state and anesthesiologist’s experience. In this paper, a recognition and localization algorithm
based on deep learning was proposed. The algorithm used the deep learning model to extract the features of the normalized object
area, and enhanced the representation ability of the features. Finally, the SVM classification model was used to get the final detection
result of the conus medullaris end. The results of qualitative and quantitative experiments showed that the performance of the deep
learning detection model proposed in this paper was good, and had certain feasibility and application value.
罗燕1,李欣2,秦小勇3,梁晶晶1,关延顺2,杨建威4. 腰麻领域基于卷积神经网络的脊髓圆锥末端位置定位算法[J]. 中国医疗设备, 2020, 35(1): 61-63.
LUO Yan1, LI Xin2, QIN Xiaoyong3, LIANG Jingjing1, GUAN Yanshun2, YANG Jianwei4. Location Algorithm of Conus Medullaris End Based on Convolution Neural Network
for Lumbar Anesthesia. China Medical Devices, 2020, 35(1): 61-63.