1.Nodule candidate detection algorithm based on deep learning
Caidi ZHANG ; Yueyang LI ; Fangzheng CUI ; Haichi LUO ; Zhongxuan GU
Chinese Journal of Medical Physics 2024;41(9):1177-1184
A nodule candidate detection algorithm based on 3DSCANet utilizing deep learning techniques is proposed to improve nodule candidate detection performance.The algorithm employs a strengthen coordinate attention(SCA)module which improves upon the basic coordinate attention mechanism to enable it to extract three-dimensional(3D)features,and incorporates adaptive convolution to extract cross-channel features,thereby enhancing the feature extraction capability of the SCA mechanism.Additionally,a method to convert 3D rectangular anchor boxes into 3D spheres is proposed,along with the introduction of a sphere based intersection over union loss function(SIoUX)to fully leverage the morphological characteristics of lung nodules which are spherical in shape.During the experimental phase,the method is tested on the LUNA16 dataset using ten-fold cross-validation,and it achieves an average recall rate of 0.94.
2.Research progress on surface treatment technology of zirconia implant
Jing LI ; Caidi JIANG ; Hui YE ; Yue WANG ; Yingchun ZHU ; Dalin WANG
Chinese Journal of Stomatology 2020;55(9):691-695
Zirconia material has a color closer to human natural teeth, and has excellent mechanical properties and good biocompatibility, therefore it is expected to become an ideal material for dental implants. In order to improve the osteogenic properties of zirconia implants, a variety of surface treatment techniques emerge in two categories: non-coating method and coating method. Different surface treatment technologies have their advantages and disadvantages. In this paper, the recent progress of zirconia surface treatment technology is reviewed.
3.Effects of rehabilitation training combined with thunder-fire moxibustion on affected upper limb spasm after stroke
Na ZHANG ; Shurui GAO ; Haiyun LIANG ; Qiong WU ; Hang XIONG ; Yan LIU ; Ying XIE ; Lijie ZHANG ; Lyu ZHANG ; Wenjuan WANG ; Caidi LI
Chinese Journal of Modern Nursing 2018;24(15):1821-1824
Objective To explore the effects of thunder-fire moxibustion in improving affected upper limb hypermyotonia of patients after stroke, so as to provide a clinical basis for improving affected upper limb hypermyotonia (spasm) after stroke. Methods From June 2016 to February 2017, we selected 100 stroke patients from Fastern Area of Dongzhimen Hospital, Beijing University of Chinese Medicine. All of the patients were divided into control group and experimental group according to the order of admission, 50 cases in each group. Patients of control group accepted routine rehabilitation training. On that basis, patients of experimental group were also treated with thunder-fire moxibustion. We compared the rehabilitation efficacy of two groups with the activity of daily living (Barthel) and the Modified Ashworth Scale (MAS). Results The scores of MAS of experimental group were (1.83±0.64) and (1.27±0.85) respectively 14 and 28 days after intervention, lower than those of control group [(2.60±0.79), (1.99±0.85)]with significant differences (t=-2.711, -2.808; P<0.01). There was a significant difference in the score of activity of daily living between experimental group (55.44±19.77) and control group (48.85±13.73), (t=2.112, P=0.037). Conclusions The thunder-fire moxibustion combined with rehabilitation could improve the activity of daily living of patients after stroke and reduce the degree of affected upper limb spasm of patients.