ABSTRACT
Objective:To develop a preoperative CT image segmentation algorithm based on artificial intelligence deep learning technology for total hip arthroplasty (THA) revision surgery, and to verify and preliminarily apply it.Methods:A total of 706 revision cases with clear CT data from April 2019 to October 2022 in Chinese PLA General Hospital were retrospectively analyzed, including 520 males, aged 58.45±18.13 years, and 186 females, aged 52.23±16.23 years. All of them were unilateral, and there were 402 hips on the left and 304 hips on the right. The transformer_unet convolutional neural network was constructed and trained using Tensorflow 1.15 to achieve intelligent segmentation of the revision THA CT images. Based on the developed three-dimensional planning system of total hip arthroplasty, an intelligent planning system for revision hip arthroplasty was preliminarily constructed. Dice overlap coefficient (DOC), average surface distance (ASD) and Hausdorff distance (HD) parameters were used to evaluate the segmentation accuracy of transformer_unet, full convolution network (FCN), 2D U-shaped Net and Deeplab v3 +, and segmentation time was used to evaluate the segmentation efficiency of these networks.Results:Compared with the FCN, 2D U-Net, and Deeplab v3+ learning curves, the transformer_unet network could achieve better training effect with less training amount.The DOC of transformer_unet was 95%±4%, the HD was 3.35±1.03 mm, and the ASD was 1.38±0.02 mm; FCN was 94%±4%, 4.83±1.90 mm, 1.42±0.03 mm; 2D U-Net was 93%±5%, 5.27±2.20 mm, and 1.46±0.02 mm, respectively. Deeplab v3+ was 92%±4%, 6.12±1.84 mm, 1.52±0.03 mm, respectively. The transformer_unet coefficients were better than those of the other three convolutional neural networks, and the differences were statistically significant (all P<0.05). The segmentation time of transformer_unet was 0.031±0.001 s, FCN was 0.038±0.002 s, 2D U-Net was 0.042±0.001 s, Deeplab v3+ was 0.048±0.002 s. The segmentation time of transformer_unet was less than that of the other three convolutional neural networks, and the difference was statistically significant ( P<0.05). Based on the results of previous studies, an artificial intelligence assisted preoperative planning system for THA revision surgery was initially constructed. Conclusion:Compared with FCN, 2D U-Net and Deeplab v3+, the transformer_unet convolutional neural network can complete the segmentation of the revision THA CT image more accurately and efficiently, which is expected to provide technical support for preoperative planning and surgical robots.
ABSTRACT
OBJECTIVE：To establish the fingerprint of wine-processed Schisandra chinensis ，and to conduct cluster analysis and principal component analysis. METHODS ：HPLC method was adopted. The determination was performed on Diamonsil C 18（2） column with mobile phased consisted of methanol-water （gradient elution ）at the flow rate of 1 mL/min. The detection wavelength was set at 250 nm，and the column temperature was 30 ℃；the injection volume was 10 μL. With schisandrol A as the reference peak，HPLC fingerprints of 15 batches of samples were drawn and their similarity were evaluated with Similarity Evaluation System of TCM Chromatographic Fingerprint （2012 edition）. The common peaks were determined. Cluster analysis and principal component analysis were performed by using SPSS 22.0 statistical software. RESULTS ：There were 20 common peaks in 15 batches of samples ，and the similarities were 0.983-0.999；a total of 8 common peaks were identified ，namely 5-hydroxymethyl furfural，schisandrol A ，schisandrol B ，schisantherin A ，schisantherin B ，deoxyschizandrin，γ-schizandrin，pseudo-γ-schizandrin. The results of cluster analysis showed that 15 batches of wine-processed S. chinensis could be clustered into 4 categories. Among them，S1-S4 and S 14 were clustered into one category ，S9-S11 were clustered into one category ，S5，S7-S8，S12-S13 were clustered into one category ，and S 6 and S 15 were clustered into one category. The results of principal component analysis showed that the cumulative variance contribution rate of first four principal component s was 85.381％；the classification results were basically consistent with the results of cluster analysis. Compared with S. chinensis ，5-hydroxymethyl furfural was newly found in S. chinensis after wine-processing ，with high content ；but there was no significant difference in the other chromatographic peaks. CONCLUSIONS：The established HPLC fingerprint is simple and easy to operate ，combined with cluster analysis and principal component analysis ，can be used for quality control of wine-processed S. chinensis decoction pieces.
ABSTRACT
<p><b>OBJECTIVE</b>To evaluate the application of tendon-derived stem cells (TDSC) and bone marrow-derived mesenchymal stem cells (BMSC) for patellar tendon injury repair in rat model.</p><p><b>METHODS</b>TDSCs and BMSCs were isolated from patellar tendons or bone marrow of healthy SD rats. The patellar tendon injury model was induced in 60 SD rats, then the animals were divided into 3 groups with 20 in each group: rats in TDSC group received transplantation of TDSC with fibrin glue in defected patellar tendon, rats in BMSC group received BMSC with fibrin glue for transplantation and those in control group received fibrin glue only. The gross morphology, histology and biomechanics of the patellar tendon were examined at 1, 2, 4, 6 and 8 weeks after the treatment.</p><p><b>RESULTS</b>Gross observation showed that the tendon defects in TDSC group and BMSC group almost disappeared in week 8, while the boundary of tendon defects in control group was still visible. Histology examination showed that the neo-tendon formation in TDSC group and BMSC group was observed at week 8, while there was no neo-tendon formation in control group. Biomechanics study showed that the ultimate stress and Young Modulus, relative ultimate stress and relative Young Modulus increased with the time going in all groups(all P<0.05); the ultimate stress and Young Modulus, relative ultimate stress and relative Young Modulus of TDSC and BMSC groups were significantly higher than those in control group at week 4, 6 and 8(all P<0.05). There was no difference in ultimate stress and Young Modulus between TDSC group and BMSC group(P>0.05), however, the relative Young Modulus of TDSC group was significantly higher than that in BMSC group at week 8(P<0.05).</p><p><b>CONCLUSION</b>Allogeneic TDSC and BMSC transplantation facilitates the repair of tendon injury and improves the biomechanics of tendon. TDSC is more suitable for in vivo tendon regeneration than BMSC.</p>