Object:To explore the feasibility of using artificial intelligence for quality control of echocardiographic images.Methods:Retrospectively，5 000 two-dimensional echocardiographic video images within the period from 2021 to 2023 were randomly retrieved from the echocardiography database of Nanjing Drum Tower Hospital，Affiliated Hospital of Medical School，Nanjing University. Among these selected images，1 559 of them were apical views. The physician team formulated the scoring rules，which specifically included four scoring criteria：gain，scaling ratio，cardiac axis angle，and structure. Subsequently，the data were labeled with view classification and image quality scores. The labeled data were further partitioned into the training set（ n = 643），the validation set（ n = 276），and the test set（ n = 640）. The training and validation sets were utilized for constructing the models for view classification and quality assessment，while the test set was employed to verify the models' effectiveness. The view classification module was implemented using the SlowFast model，and the quality assessment module involved algorithms such as ResNet，Video Swin Transformer，SSD，and U-Net. Results:The average accuracy，precision，recall rate and F1 score of the classification model in identifying each apical view were 0.987 1，0.983 0，0.987 1 and 0.984 9 respectively，and the inference time was（333.4 ± 105.4）ms. The average accuracies of the quality assessment module in terms of gain，scaling ratio，cardiac axis angle and display of main structures were 0.915 1，0.928 2，0.938 7 and 0.965 6 respectively，and the overall scoring accuracy was 0.912 7.Conclusions:The echocardiogram quality control system developed in this research can effectively classify and evaluate the quality of two-dimensional images of the apical views in echocardiograms. Moreover，it guarantees the objectivity，timeliness and high-efficiency of quality control，which has reference value for the establishment of the echocardiogram quality control system.

Object:To explore the feasibility of using artificial intelligence for quality control of echocardiographic images.Methods:Retrospectively，5 000 two-dimensional echocardiographic video images within the period from 2021 to 2023 were randomly retrieved from the echocardiography database of Nanjing Drum Tower Hospital，Affiliated Hospital of Medical School，Nanjing University. Among these selected images，1 559 of them were apical views. The physician team formulated the scoring rules，which specifically included four scoring criteria：gain，scaling ratio，cardiac axis angle，and structure. Subsequently，the data were labeled with view classification and image quality scores. The labeled data were further partitioned into the training set（ n = 643），the validation set（ n = 276），and the test set（ n = 640）. The training and validation sets were utilized for constructing the models for view classification and quality assessment，while the test set was employed to verify the models' effectiveness. The view classification module was implemented using the SlowFast model，and the quality assessment module involved algorithms such as ResNet，Video Swin Transformer，SSD，and U-Net. Results:The average accuracy，precision，recall rate and F1 score of the classification model in identifying each apical view were 0.987 1，0.983 0，0.987 1 and 0.984 9 respectively，and the inference time was（333.4 ± 105.4）ms. The average accuracies of the quality assessment module in terms of gain，scaling ratio，cardiac axis angle and display of main structures were 0.915 1，0.928 2，0.938 7 and 0.965 6 respectively，and the overall scoring accuracy was 0.912 7.Conclusions:The echocardiogram quality control system developed in this research can effectively classify and evaluate the quality of two-dimensional images of the apical views in echocardiograms. Moreover，it guarantees the objectivity，timeliness and high-efficiency of quality control，which has reference value for the establishment of the echocardiogram quality control system.

Objective To explore the value of deep learning(DL)models semi-automatic training system for automatic optimization of clinical image quality control of transthoracic echocardiography(TTE).Methods Totally 1 250 TTE videos from 402 patients were retrospectively collected,including 490 apical four chamber(A4C),310 parasternal long axis view of left ventricle(PLAX)and 450 parasternal short axis view of great vessel(PSAXGv).The videos were divided into development set(245 A4C,155 PLAX,225 PSAXGV),semi-automated training set(98 A4C,62 PLAX,90 PSAXGV)and test set(147 A4C,93 PLAX,135 PSAXGV)at the ratio of 5:2:3.Based on development set and semi-automatic training set,DL model of quality control was semi-automatically iteratively optimized,and a semi-automatic training system was constructed,then the efficacy of DL models for recognizing TTE views and assessing imaging quality of TTE were verified in test set.Results After optimization,the overall accuracy,precision,recall,and F1 score of DL models for recognizing TTE views in test set improved from 97.33％,97.26％,97.26％and 97.26％to 99.73％,99.65％,99.77％and 99.71％,respectively,while the overall accuracy for assessing A4C,PLAX and PSAXGV TTE as standard views in test set improved from 89.12％,83.87％and 90.37％to 93.20％,90.32％and 93.33％,respectively.Conclusion The developed DL models semi-automatic training system could improve the efficiency of clinical imaging quality control of TTE and increase iteration speed.

Objective:To investigate the relapse risk factors of anti-aquaporin 4 (AQP4)-IgG positive neuromyelitis optica spectrum disorders (NMOSD) patients treated with immunosuppressant.Methods:Data (from January 2011 to June 2021) of AQP4-IgG positive NMOSD patients treated with immunosuppressant for longer than 5 years from MSNMObase, a hospital-based electronic registry for multiple sclerosis and related disorders in Peking Union Medical College Hospital, were collected. Clinical features and risk factor differences between patients with and without relapse under the immunosuppressive therapy were analyzed.Results:One hundred and twelve patients with AQP4-IgG positive NMOSD were included, 105 (93.8%) of which were female. The disease onset age was (34.9±11.3) years, 13(11.6%) had an older disease onset age than 50 years (late onset), and the disease duration was 8.1 (6.6, 11.4) years. Sixty-four (57.1%) patients had relapse, and the proportion of late onset patients was significantly lower in relapse group than in non-relapse group [4/64(6.3%) vs 9/48(18.8%), χ2=4.18, P=0.041]. Compared with those without relapse, both the annualized relapse rate (ARR) before treatment [1.07 (0.36, 2.25) vs 0.34 (0, 1.11), Z=2.92, P=0.003] and the proportion of patients with relapse before treatment [54/64(84.4%) vs 33/48(68.8%), χ2=3.86, P=0.049] were significantly higher for patients in relapse group. Multivariate Logistic regression analysis revealed the relapse risk of late-onset patients was lower than that of early-onset patients ( HR=0.26, 95% CI 0.10-0.73, P=0.010) and patients with higher ARR before treatment showed a higher risk of relapse under the immunosuppressive therapy ( HR=1.55,95% CI 1.26-1.91, P<0.001). Conclusion:AQP4-IgG positive NMOSD patients with younger disease onset age than 50 years or with frequent relapses before treatment had a higher relapse risk under the immunosuppressive therapy, and they may need highly effective treatments.

Objective:Longitudinally extensive transverse myelitis (LETM) could be seen in patients with connective tissue disease (CTD), especially systemic lupus erythematosus (SLE) or primary Sj?gren′s syndrome (pSS). Some patients are combined with neuromyelitis optica spectrum disorders (NMOSD)(termed CTD-LETM-NMOSD) while others without (termed CTD-LETM-non-NMOSD). The aim of this study is to compare the clinical characteristics of CTD-LETM-NMOSD patients to CTD-LETM-non-NMOSD patients.Methods:We retrospectively collected data from 40 CTD patients with LETM who were admitted to the Department of Neurology or Rheumatology at Peking Union Medical College Hospital from Jan, 2006 to Dec, 2016. They were divided into CTD-LETM-NMOSD and CTD-LETM-non-NMOSD two groups. Demographic characteristics, clinical and laboratory features were obtained from the database. Relapse rates and clinical outcome were analyzed by Kaplan-Meier method.Results:Among 40 patients with CTD, 28 (70.0%) were NMOSD while 12 (30.0%) were not. The positivity rates of anti-SSA, antibodies to aquaporin-4 (anti-AQP4) were significantly higher in patients with NMOSD than those in patients with non-NMOSD ( P<0.05). Age, gender, clinical features, disease duration, anti-double-stranded DNA antibody, anti-ribosomal P antibody, antiphospholipid antibodies, expanded disability status scale (EDSS) scores, and magnetic resonance imaging (MRI) features were all comparable between two groups. CTD-NMOSD patients had significantly higher disease relapse rate (75.0% vs. 3/12, P<0.01). Conclusion:Anti-SSA and anti-AQP4 positivity is associated with NMOSD and higher relapse rates, which suggests that NMOSD in CTD-LETM patients may represent distinct characteristics and pathogenesis from patients with CTD-LETM-non NMOSD.

Objective To summerize our experience in the diagnosis and treatment of pancreatic cystic neoplasms.Methods A retrospective analysis was conducted on the clinical data of 126 patients with pancreatic cystic neoplasms seen between January 2003 and December 2012 in the Fuzhou General Hospital.Results There was no special clinical manifestation in this series of 126 patients with pancreatic cystic neoplasms.The diagnostic accuracies of ultrasound,CT and MRI were 90.8％ (109/120),93.4％ (114/122) and 96.3％ (103/107) respectively.Eighteen and ten patients were observed to have elevated serum CA19 9 and CEA respectively.One hundred and five patients received conventional open surgery,while 21 patients underwent laparoscopic operations.The operations in cluded enucleation of tumor (n=11),pancreatic segment resection (n=5),spleen preserving distal pancreatectomy (n=30),distal pancreatectomy (n=26),duodenum-preserving pancreatic head resection (n=4),pylorus-preserving Whipple resection (n=5),Whipple procedure (n=38),total pancre atectomy (n=2) and exploratory laparotomy and biopsy (n=5).The perioperative mortality rate was 0.8％ (1/126).Pancreatic fistula (B and C) occurred in 16 patients (12.7％,16/126),and postoperative hemorrhage (B and C) occurred in 9 patients (7.1％,9/126).One hundred and fourteen patients were followed up from 6 to 72 months (median 52.4 months).The five-year survival rates for all the pancreatic cystic neoplasms,non-invasive and invasive neoplasms were 80.5％,96.4％ and 40.7％ respectively.Conclusions Pancreatic cystic neoplasms are a rare subset of pancreatic tumor,being increasingly detected due to the widespread use of abdominal imaging and improved imaging techniques.Most of them are benign,but some are malignant or they have malignant transformation and metastatic potential.Patients with asymptomatic benign pancreatic cystic neoplasms,especially small size,are candidates for observation.Patients with clinical symptoms or having a high suspicion of malignancy should be offered surgical resection.Non-invasive neoplasms should receive pancreatic preserving operations.Pancreatectomy plus regional lymph nodes dissection with or without combined resection of adjacent multi organs should be adopted for invasive neoplasms.

Objective To study the anatomy of the caudate lobe and to apply the knowledge in caudate lobe resection and piggyback liver transplantation. Methods Dissection of the caudate lobe was carried out in 35 fresh adult livers. Before dissection, the portal venous system was injected with barium sulfate emulsion to show the portal branches of the caudate lobe. Results The caudate lobe was shown to consiste of 3 portions: (1) the Spigelian lobe, the paracaval portion and the caudate process. The Spigelian lobe was pyramidal in shape with 3 faces, a peak and a base; (2) the paracaval portion was situated to the right and to the front of the vena cava; (3) the caudal process was the liver parenchyma between the hepatic hilum and the anterior aspect of the vena cava. Conclusions The entire caudate lobe is situated at the posterior part of the liver and it surrounds the vena cava. The key point in resecting the entire caudate lobe or in performing a piggyback liver transplantation is adequate mobilization and division of all the outflowing veins of the caudate lobe.

Objective To observe the effects of Wu-Yao,Da-Huang combination on severe acute pancreatitis gut barrier protection,and on the clinical treatment of severe acute pancreatitis.Methods In the past 10 months,45 casea of severe pancreatitis those treated in our department were randomly divided into 3 groups.group A:on the basis of conventional treatment of inhibiting trypsin secretion and anti-infection,50%magnesium sulfate 50 mL were insected via stomach tube,twice a day;group B:on the basis of conventional treatment,Da-Huang decoction 100 mL high retention enema,twice a day;group C:on the basis of conventional treatment were given by Wu-Yao decoction 50 mL.as well as 100 mL Da-Huang decoction enema,twice a day.Three groups were observed the release of abdominal distension,abdominal(bladder pressure measurement),ACAPHE-Ⅱ score,indicators of neutrophils(N),C-reactive protein(CRP),and the blood levels of D-lactic acid in 1,3,5,7,9 days etc.Results The abdomenal circumference,abdominal pressure,APACHE-Ⅱ score,N and other indicators of the three groups after treatment had the significantly different(P<0.05),with group C the most significant decrease,while no significant difference in CRP.Endotoxin content in three groups was significantly different(P<0.05),and each diminished with time(P<0.05).Conclusion Wu-Yao,Da-Huang combination on gut barrier dysfunction in severe acute pancreatitis showed synergistic protective effect.