Objective:To investigate the application value of major anatomical structure recognition model of minimally invasive liver resection based on deep learning.Methods:The retrospective and descriptive study was conducted. The 31 surgical videos of laparoscopic left lateral sectionectomy performed in Zhujiang Hospital of Southern Medical University from January 2019 to April 2023 were collected. Video clips containing the surgical procedure of left lateral lobe liver pedicle and left hepatic vein were screened by 2 liver surgeons. After quality control, screening and frame extraction, the major anatomical structures on the images of these clips were annotated. After pre-processing, these images were transported to the DeepLab v3+neural network framework for model training. Observation indicators: (1) video annotation and classification; (2) results of arti-ficial intelligence anatomical recognition model testing. Measurement data with normal distribution were represented as Mean± SD, and count data were described as absolute numbers. Results:(1) Video annotation and classification. A total of 4 130 frames of images were annotated in the 31 surgical videos, including 2 083 frames of annotated images for the left lateral lobe liver pedicle, 1 578 frames of annotated images for the left hepatic vein and 469 frames of annotated images for both the left lateral lobe liver pedicle and left hepatic vein. (2) Results of artificial intelligence anatomical recognition model testing. In four application scenarios (clean scene, bloodstain scene, partially obstruction by instrument scene, and small exposed area scene), the model was able to successfully recognize the left lateral lobe liver pedicle and left hepatic vein, with a recognition speed for anatomical markers >13 frames/s. When performing anatomical recognition on images with only the left lateral lobe liver pedicle, the Dice coefficient, intersection over union, accuracy, sensitivity and specificity of the model were 0.710±0.110, 0.560±0.120, 0.980±0.010, 0.640±0.030, and 0.980±0.010, respectively. The above indicators of the model were 0.670±0.180, 0.530±0.200, 0.980±0.010, 0.600±0.040, and 0.990±0.010 when performing anatomical recognition on images with only the left hepatic vein, and 0.580±0.180, 0.430±0.190, 0.980±0.010, 0.580±0.020, and 0.990±0.010 when per-forming anatomical recognition on images with both the left lateral lobe liver pedicle and left hepatic vein.Conclusion:The major anatomical structure recognition model of minimally invasive liver resection based on deep learning can be applied in identifying liver pedicle and hepatic vein.

Objective:To investigate the application value of augmented reality navigation combined with indocyanine green(ICG) fluorescence imaging technology in laparoscopic anatomical segment 8 liver resection.Methods:Clinical and pathological data from 8 patients with hepatocellular carcinoma located in segment 8 of the liver admitted to the First Department of Hepatobiliary Surgery,Zhujiang Hospital,Southern Medical University from October 2021 to October 2022 were collected restrospectively. Among them,there were 5 males and 3 females,aged between 40 and 72 years. During the operation,the self-developed laparoscopic augmented reality surgical navigation system was used to integrate the three-dimensional liver model with the laparoscopic scene,and ICG fluorescence imaging technology was used to guide the anatomical liver resection of segment 8. The predicted liver resection volume and actual liver resection volume,related surgical indicators and postoperative complications were analyzed.Results:Among the 8 patients, 4 underwent laparoscopic anatomical segment 8 liver resection,1 underwent laparoscopic anatomical ventral subsegment of segment 8 liver resection,2 underwent laparoscopic anatomical ventral subsegment combined with medial subsegment of segment 8 liver resection, and 1 underwent laparoscopic anatomical dorsal subsegment of segment 8 liver resection. All operations were completed under the guidance of augmented reality navigation combined with ICG fluorescence imaging,without conversion to open surgery. The operation time was (276.3±54.8)minutes(range:200 to 360 minutes). Intraoperative blood loss was (75.0±35.4)ml(range:50 to 150 ml). No blood transfusion was performed during the operation. The length of postoperative hospital stay was (7.6±0.8)days(range:7 to 9 days). There were no deaths or postoperative complications such as bleeding or biliary fistula during the perioperative period.Conclusion:Augmented reality navigation combined with ICG fluorescence imaging technology can guide the implementation of laparoscopic anatomical segment 8 liver resection.

Objective:To investigate the application value of augmented reality navigation combined with indocyanine green(ICG) fluorescence imaging technology in laparoscopic anatomical segment 8 liver resection.Methods:Clinical and pathological data from 8 patients with hepatocellular carcinoma located in segment 8 of the liver admitted to the First Department of Hepatobiliary Surgery,Zhujiang Hospital,Southern Medical University from October 2021 to October 2022 were collected restrospectively. Among them,there were 5 males and 3 females,aged between 40 and 72 years. During the operation,the self-developed laparoscopic augmented reality surgical navigation system was used to integrate the three-dimensional liver model with the laparoscopic scene,and ICG fluorescence imaging technology was used to guide the anatomical liver resection of segment 8. The predicted liver resection volume and actual liver resection volume,related surgical indicators and postoperative complications were analyzed.Results:Among the 8 patients, 4 underwent laparoscopic anatomical segment 8 liver resection,1 underwent laparoscopic anatomical ventral subsegment of segment 8 liver resection,2 underwent laparoscopic anatomical ventral subsegment combined with medial subsegment of segment 8 liver resection, and 1 underwent laparoscopic anatomical dorsal subsegment of segment 8 liver resection. All operations were completed under the guidance of augmented reality navigation combined with ICG fluorescence imaging,without conversion to open surgery. The operation time was (276.3±54.8)minutes(range:200 to 360 minutes). Intraoperative blood loss was (75.0±35.4)ml(range:50 to 150 ml). No blood transfusion was performed during the operation. The length of postoperative hospital stay was (7.6±0.8)days(range:7 to 9 days). There were no deaths or postoperative complications such as bleeding or biliary fistula during the perioperative period.Conclusion:Augmented reality navigation combined with ICG fluorescence imaging technology can guide the implementation of laparoscopic anatomical segment 8 liver resection.

Primary liver cancer has the features of high malignancy, rapid progression, frequent recurrence/metastasis, and high mortality, and therefore, most patients have developed intrahepatic or extrahepatic metastasis when attending the hospital and thus lost the opportunity for surgical treatment. Chimeric antigen receptor T-cell (CAR-T) immunotherapy has achieved good efficacy in the treatment of B-cell acute lymphoblastic leukemia, and clinical trials have been initiated to explore its applications in solid tumors such as primary liver cancer, pancreatic cancer, gastric cancer, and prostate cancer. This article reviews the efficacy of CAR-T immunotherapy in the clinical trials for primary liver cancer and discusses the difficult issues that need to be solved in clinical practice, such as the lack of suitable tumor targets, the inhibitory effect of tumor microenvironment on CAR-T cells, and the poor infiltration of CAR-T cells in tumor tissue, so as to provide a reference for related clinical studies.

Objective:To investigate the clinical effect of the transverse rectus abdominismuscle (TRAM) on reconstruction of the breast.Methods:The clinical data of 23 patients receiving TRAM breast reconstruction in our department from Jan. 2018 to Dec. 2019 were retrospectively analyzed.Results:The operation time of 23 patients ranged from 240 to 360 mins, andthe average time was about 300 mins. Intraoperative bleeding was about 120 to 200 ml, with an average of 170 ml. All the flaps survived successfully, but 2 cases were complicated with local fat necrosis. The postoperative period was between 6 and 12 months. No local tumor recurrence or metastasis was found inall patients during postoperative follow-up, and the breast shape was maintained in good condition.Conclusion:TRAM can make up for the regret of breast loss caused by breast cancer in female patients. It can bring confidence in life and work to female patients, and the technology is safe and reliable, which is worthy of promotion.

Robot-assisted laparoscopy hepatectomy (RALH) is a new technique for surgical operation. Compared with conventional laparoscopic hepatectomy, RALH is more frequently used in complex liver tumor and liver tumor with special locations, but this technique is still under development and is limited by the burden of high costs and surgical devices. Meanwhile, there is a lack of generally accepted and confirmed clinical data, and therefore, the role of RALH is still under debate. This article reviews the surgical indication, learning curve, advantages, and limitations of RALH.

Objective:To analyze the OCT angiography (OCTA) features of eyes affected with angioid streaks.Methods:Retrospective analysis of 26 patients (52 eyes) diagnosed as angioid streaks by multimodal imaging in Ophthalmology Department of Yunnan Second People’s Hospital from May 2017 to February 2019 were included in this study. There were 18 males and 8 females, with the mean age of 50.8±6.9 years. All the patients were binocular. There were 34 eyes in 20 patients with CNV. Among them, 13 eyes had a course of disease within 1 month, 16 eyes had a course of disease over 1 month, and 5 eyes were treated with anti-VEGF drugs. All patients were examined by ultra-wide-angle fundus photography, infrared fundus imaging (IR), spectral-domain OCT, FAF and FFA. At the same time, the Heidelberg OCTA instrument was used for OCTA examination. The central wavelength was 840 nm, the acquisition speed was 85,000 times/s, and the width was 45 nm. A 3 mm × 3 mm scan was obtained. Each cube consisted of two 304 B scans of repeated volumes, and motion correction was performed using two orthogonally captured image volume registration. The results of fundus color photography, IR, FAF, FFA, OCT and OCTA were compared and analyzed to summarize the image features of AS and its secondary CNV in OCTA.dus color photography, IR, FAF, FFA, OCT and OCTA were compared and analyzed to summarize the image features of AS and its secondary CNV in OCTA.Results:Among 52 eyes, 40 eyes showed choroidal capillary shadows in OCTA, and no obvious abnormal OCTA images in 12 eyes. Ten eyes OCTA showed a vascular network beseide the optic disc, FFA showed fluorescent staining instead of fluorescein leakage. Among the 52 eyes, 34 eyes had secondary macular CNV, and 34 eyes had different forms of CNV. Different forms had a certain correlation with the length of medical history and treatment history. Among them, 13 eyes with short course of disease (less than 1 month) and untreated patients had smaller flower-ring morphology, 16 eyes with longer course (more than 1 month) but no treatment had larger fan-shaped segments, and 5 eyes with anti-VEGF therapy were trimmed after the dendritic shape.Conclusions:AS streaks appeared no blood flow signal area in OCTA, and repairing vascular network beside the optic disc can be observed in some AS patients. The CNV performance with different disease course and treatment experience is different.

Objective To evaluate the effect of nucleolar protein 14 (NOP14) on angiogenesis in melanoma.Methods Melanoma tissues were collected from 40 patients with pathologically diagnosed melanoma in Guangzhou First People's Hospital from January 2016 to December 2018,and immunohistochemical study was conducted to determine the expression of NOP14 and CD31 (expressed as microvessel density [MVD]).Melanoma cell lines A375 and SK-MEL-1 were both divided into 4 groups:empty vector group transfected with the empty vector,NOPI4 group transfected with a NOP14-overexpressing vector,siNOP14 group transfected with the siRNA targeting NOP14,and siNC group transfected with a negative control siRNA.Fluorescence-based quantitative PCR and Western blot analysis were performed to determine the mRNA and protein expression of NOP14 respectively,and Western blot analysis and enzyme-linked immunosorbent assay (ELISA) to measure the expression of vascular endothelial growth factor (VEGF) and VEGF receptor (VEGFR) in cells and their culture media.Coculture models of human umbilical vein endothelial cells (HUVECs) and A375/SK-MEL-1 cells in the above groups were established in Transwell chambers,and cell counting kit-8 (CCK8) assay,Transwell migration and invasion assays and Matrigel-based vasculogenic mimicry assay were performed to evaluate the cellular proliferative,migratory,invasive activity and tube formation capacity respectively.A linear regression model was used to analyze the relationship between NOP14 expression and MVD in melanoma tissues,multi-way analysis of variance to analyze the difference in cellular proliferative activity,and independent-sample t test to compare other experimental indices between 2 groups.Results The expression of CD31 (MVD) was 44 ± 13 in the group with high NOP14 expression (n =20),58 ± 16 in that with moderate NOP14 expression (n =17),and 62 ± 11 in that with low NOP14 expression (n =3).The NOP14 expression was negatively correlated with MVD (r =-0.525,P =0.017).Compared with the empty vector group,the expression of VEGF and VEGFR in A375 and SK-MEL-1 cells and their culture media significantly decreased in the NOP14 group (all P ＜ 0.05).Compared with the siNC group,the expression of VEGF and VEGFR in the A375 and SK-MEL-1 cells and their culture media significantly increased in the siNOP14 group(all P ＜ 0.05).In the co-culture models of A375 cells and HUVECs,the NOP14 group showed significantly decreased proliferative activity of HUVECs (F =131.85,P ＜ 0.05),and numbers of migratory cells (22 ± 5 vs.63 ± 8,t =7.07,P =0.002),invasive cells (14 ± 5 vs.45 ± 10,t =4.94,P =0.008) and branch points (8 ± 2 vs.14 ± 3,t =5.06,P ＜ 0.001) compared with the empty vector group;compared with the siNC group,the siNOP14 group showed significantly increased proliferative activity of HUVECs (F =79.92,P ＜ 0.01),and numbers of migratory cells (152 ± 30 vs.59 ± 4,t =5.36,P =0.006),invasive cells (134 ± 21 vs.50 ± 8,t =6.40,P ＜ 0.001) and branch points (27 ± 3 vs.15 ± 4,t =6.10,P ＜ 0.001).In the co-culture models of SK-MEL-1 cells and HUVECs,the 4 groups showed the same trend of changes in the cellular proliferative,migratory,invasive activity and tube formation capacity of HUVECs as the above groups in the co-culture models of A375 cells and HUVECs.Conclusion The NOP14 expression is negatively correlated with MVD in melanoma tissues,and NOP14 can inhibit angiogenesis in melanoma.

Objective:To develop a new method based on neuraminidase (NA) activity for detec-ting virus titers of cell culture-based influenza vaccines and preliminarily analyze its application.Methods:Reaction conditions including the substrate concentration for enzymatic reaction, stop solution, the number of initially infected target cells and cell lysis buffer were optimized. The titers of cell culture-based influenza vaccine strains were detected by the established method and the results were compared with those by the traditional viral titration test.Results:The optimal substrate concentration for enzymatic reaction was 25 μmol/L, and the optimal stop solution was 0.2 mol/L Na 2CO 3. In the detection of NA activity in infected cells, the maximum relative fluorescence value was obtained by infecting 4×10 4 cells/well with influenza virus for 48 h and using 0.5% TritonX-100 for lysis. The developed method showed no significant differences with the traditional virus titration test in detecting the titers of four batches of influenza vaccine virus strains ( P>0.05), indicating that the two methods had a good consistency. Conclusions:This study established a new method based on NA activity to detect virus titers of cell culture-based influenza vaccines. The method could be used for the detection of virus strains used in the production of cell cultured-based influenza vaccines.

Background and purpose:Diffusion-weighted imaging (DWI) is a non-invasive technique of breast magnetic resonance imaging (MRI). DWI is an alternative to dynamic contrast-enhanced (DCE) MRI for differentiating malignant from benign lesions in breast screening or not. This study aimed to evaluate the potential role of DWI in differentiating malignant breast lesions from benign lesions.Methods:Seventy-four patients underwent digital mammography, DCE and DWI (49 patients’b-value of 0, 400, 600 and 800 s/mm2). The detectability, sensitivity and speciifcity of DWI and DCE were compared. Absolute apparent diffusion coefifcient (ADC) was compared with standardized ADC for quantitative analysis.Results:Sixty-four of 74 patients had positive pathologic findings (38 malignant, 26 benign). All of the malignant lesions were detected on DWI and DCE. The sensitivity of DWI was 83.33%, 90.00% and 93.33%, and the specificity was 85.91%, 76.19% and 72.72%, forb-value of 400, 600 and 800 s/mm2, respectively. The sensitivity and speciifcity of DCE were 86.61% and 90.48%. There was no signiifcant difference between absolute and standardized ADC in detecting breast cancer (P>0.05).Conclusion:DWI is an important complemented technique to DCE-MRI for differentiating malignant from benign lesions in breast MRI.