Objective To evaluate the value of MRI and MDCT in detecting both inferior vena cava tumor thrombus and vena cava wall invasion in renal cell carcinoma. Methods Databases including PubMed, EMbase, The Cochrane Library, MEDLINE (Ovid), CBM, CNKI, VIP and WanFang Data were searched from January 2000 to February 2012. Relevant studies were screened on the basis of the inclusion and exclusion criteria, and then quality assessment and data extraction were conducted. Then heterogeneity test and meta-analysis were conducted using RevMan 5 and Meta-disc 1.4. Results A total of 6 trials involving 244 patients and 246 cases of renal cell carcinoma were included. The results of meta-analysis showed that, for the MRI group and the MDCT group, the sensitivity was 0.963 and 0.952, the specificity was 0.969 and 0.979, the value of +LR was 9.759 and 15.57, the value of ?LR was 0.091 and 0.108, and the dOR was 198.71 and 251.54, respectively. There were no significant differences in pooled effect-size among groups (Pgt;0.05). The area under curve (AUC) of summary ROC curve analysis as well as Q index of the MDCT group were 0.981 8 and 0.940 7, respectively. Conclusion There is no significant difference in the value of MRI and MDCT in detecting inferior vena cava tumor thrombus induced by renal cell carcinoma. More original studies on vena cava wall invasion by tumor thrombus should be conducted in the future due to the limitation of current materials.
Objective To investigate the relation between the sites of colorectal cancer and liver metastatic distribution. Methods The enhanced multiple-slice spiral CT images and clinical data of 105 cases diagnosed colorectalcancer with liver metastases admitted from January 2010 to April 2012 were analyzed retrospectively. Primary site of the tumor， numbers of the metastases on CT images， and the anatomical position of the inferior mesenteric vein （IMV） terminates were recorded. Results ①The ratio of metastases in the right and left hemiliver was 2∶1 for 38 right-sided primary tumors as compared with 1.2∶1 for 67 left-sided primary tumors. The pattern of lobar distribution was significantly different in the two groups （χ2=8.709， P=0.003）. ②In the left-sided colon cancer group， the ratio of metastases in the right and left hemiliver was 65∶98 for 28 patients with IMV terminating in splenic vein （SpV）， 116∶52 for 36 patients with IMV terminating in superior mesenteric vein （SMV）， and 13∶15 for 3 patients with IMV terminating in the junction of SMV and SpV. The pattern of lobar distribution was significantly different among the three groups （χ2=28.575， P=0.000）. Further comparison between the former two groups， the difference was statistically significant （χ2=27.951， P=0.000）. ③In 25 patients with IMV terminating in SpV， the metastases of 19 cases were mainly distributed in the left lobe of liver （P=0.001）；In 34 patients with IMV terminating in SMV， the metastases of 25 cases located mainly in the right hepatic lobe （P=0.000）. Conclusions Right-sided colon cancers selectively involve the right lobe of liver， while left-sided tumors selectively involve the right lobe of liver when its IMV terminates in SMV and involve the left lobe when its IMV terminates in SpV， respectively. The discovery may help shorting the diagnostic workup in patients presenting with liver metastases from an unknown primary site， and may improve the detection rate of metastases in initial diagnosis and follow-up.
Objective To explore clinical effect of failure mode and effect analysis in improving the submission rate of pathogen examination in counterpart supported high-altitude county hospitals, and formulate practical measures and methods suitable for high-altitude county hospitals to improve the submission rate of pathogen examination. Methods Patients admitted to the People’s Hospital of Ganzi County between January and December 2024 were selected. The data of hospitalized patients between January and June 2024 were as the control group, and the data of hospitalized patients between July and December 2024 were as the intervention group. The study analyzed and compared the submission rate of pathogen testing and the pass rate of microbiological test specimens before antimicrobial treatment between the two groups. Results A total of 3 984 patients were included. Among them, there were 1 748 cases in the control group and 2 236 cases in the intervention group. A total of 10 risk factors and 2 high-risk points were identified. There were statistically significant differences in the submission rate of pathogen specimens before antibiotic treatment [36.21% (633/1 748) vs. 49.33% (1 103/2 236); χ2=68.646, P<0.001] and the qualified rate of microbiological test specimens [26.75% (122/456) vs. 36.45% (261/716); χ2=11.910, P=0.001] between the control group and the intervention group. Conclusions Failure mode and effect analysis can effectively find out the weak points in low pathogen examination submission rate in high-altitude county hospitals. According to the high-risk points to guide the formulation of relevant measures, the pathogen submission rate in the region can be effectively improved.
近十年，在藥品不良反應監測領域，基于醫療保健數據庫的安全信號檢測方法受到越來越多的關注，已成為彌補自發報告固有局限性的重要手段。目前數據挖掘方法主要基于比值失衡分析法（disproportionality analysis）、傳統藥物流行病學設計（如自身對照設計）、序列對稱分析（sequence symmetry analysis，SSA）、序貫統計檢驗（sequential statistical testing）、時序關聯規則（temporal association rules）、監督機器學習（supervised machine learning，SML）、樹狀掃描統計量方法（tree-based scan statistic）等。本文從應用場景和實用性角度對醫療保健數據庫中安全信號檢測方法及其性能進行介紹。
混合模型框架下的模型，如潛變量增長混合模型（latent growth mixture modeling，LGMM）或潛類別增長分析（latent class growth analysis，LCGA），因估算過程中涉及多個決策過程，導致潛變量軌跡分析結果的報告呈現多樣性。為解決這一問題，指南制訂小組按照系統化的制訂流程，通過 4 輪德爾菲法調查，遵循專家小組意見，提出了各領域報告潛變量軌跡分析結果時需采用統一的標準，最終確定了報告軌跡研究結果必要的關鍵條目，發布了潛變量軌跡研究報告規范（guidelines for reporting on latent trajectory studies，GRoLTS），并利用 GRoLTS 評價了 38 篇使用 LGMM 或 LCGA 研究創傷后應激軌跡的論文的報告情況。
目的 探討軀體感覺誘發電位（SEP）在頸脊髓損傷術前、術中監測的意義。 方法 納入2010年1月－2012年4月治療的241例頸脊髓損傷患者，術前按美國脊柱脊髓損傷協會（ASIA）評分并分級，確定損傷平面。術前與術中SEP監測，分析不同損傷分級以及不同損傷平面術前的波幅及潛伏期的差異，術中SEP監測以波幅下降＞50%和或潛伏期延長＞10%為預警標準。 結果 各損傷分級組術前SEP監測：A級組SEP波消失，呈一直線，而B、C、D、E級組均測出SEP波形，根據是否可測出SEP波形，可將A級與B、C、D、E及組區別。B、C、D級組之間波幅和潛伏期均無統計學意義（P＞0.05）。E級組較B、C、D級組波幅增高、潛伏期縮短，差異有統計學意義（P＜0.05）；不完全性頸脊髓損傷組內不同損傷平面組之間波幅和潛伏期差異均無統計學意義（P＞0.05）。術中SEP對脊髓功能損傷監測的靈敏度83.3%、特異度98.7%。其中術中：SEP陽性8例，真陽性5例，4例術者處理后波幅及潛伏期回復至正常范圍，術后無新的神經功能損傷，另1例術者采取各種處理后波幅及潛伏期無恢復，術后神經功能損傷較術前加重；假陽性3例，1例麻醉師給予升高血壓后波形恢復至正常，另2例經麻醉師調整麻醉深度后波形恢復正常，此3例術后無新的神經功能損傷。SEP陰性233例，真陰性232例，術后無新的神經功能損傷；假陰性1例，患者術中、術后波形未見異常，術后運動功能損傷程度較術前加重。 結論 ① SEP能準確評估完全性和不完性頸脊髓損傷，但對不完全性頸脊髓損傷的損傷程度不能作出準確評估、也不能區分頸脊髓損傷的損傷平面；② 術中SEP監測能較好地反映頸脊髓功能完整性，對減少頸脊髓損傷術中發生醫源性頸脊髓損傷風險具有重要意義。
The accurate segmentation of breast ultrasound images is an important precondition for the lesion determination. The existing segmentation approaches embrace massive parameters, sluggish inference speed, and huge memory consumption. To tackle this problem, we propose T2KD Attention U-Net (dual-Teacher Knowledge Distillation Attention U-Net), a lightweight semantic segmentation method combined double-path joint distillation in breast ultrasound images. Primarily, we designed two teacher models to learn the fine-grained features from each class of images according to different feature representation and semantic information of benign and malignant breast lesions. Then we leveraged the joint distillation to train a lightweight student model. Finally, we constructed a novel weight balance loss to focus on the semantic feature of small objection, solving the unbalance problem of tumor and background. Specifically, the extensive experiments conducted on Dataset BUSI and Dataset B demonstrated that the T2KD Attention U-Net outperformed various knowledge distillation counterparts. Concretely, the accuracy, recall, precision, Dice, and mIoU of proposed method were 95.26%, 86.23%, 85.09%, 83.59%and 77.78% on Dataset BUSI, respectively. And these performance indexes were 97.95%, 92.80%, 88.33%, 88.40% and 82.42% on Dataset B, respectively. Compared with other models, the performance of this model was significantly improved. Meanwhile, compared with the teacher model, the number, size, and complexity of student model were significantly reduced (2.2×106 vs. 106.1×106, 8.4 MB vs. 414 MB, 16.59 GFLOPs vs. 205.98 GFLOPs, respectively). Indeedy, the proposed model guarantees the performances while greatly decreasing the amount of computation, which provides a new method for the deployment of clinical medical scenarios.
Objective To explore the clinical effect of failure mode and effect analysis (FMEA) combined with PDCA cycle management model in the prevention and control of multidrug-resistant organisms (MDROs) in intensive care unit (ICU), and provide evidences for drawing up improvement measures in healthcare-associated MDRO infections in ICU. Methods In January 2020, a risk assessment team was established in the Department of Critical Care Medicine, the First People’s Hospital of Longquanyi District of Chengdu, to analyze the possible risk points of MDRO infections in ICU from then on. FMEA was used to assess risks, and the failure modes with high risk priority numbers were selected to evaluate the high-risk points of MDRO infections. The causes of the high-risk points were analyzed, and improvement measures were formulated to control the risks through PDCA cycle management model. The incidence of healthcare-associated MDRO infections in ICU, improvement of high-risk events, and satisfaction of doctors and nurses after the implementation of intervention measures (from January 2020 to June 2021) were retrospectively collected and compared with those before the implementation of intervention measures (from January 2018 to December 2019). Results Six high-risk factors were screened out, namely single measures of isolation, unqualified cleaning and disinfection of bed units, irrational use of antimicrobial agents, weak consciousness of isolation among newcomers of ICU, weak awareness of pathogen inspection, and untimely disinfection. The incidence of healthcare-associated MDRO infections was 2.71% (49/1800) before intervention and 1.71% (31/1808) after intervention, and the difference between the two periods was statistically significant (χ2=4.224, P=0.040). The pathogen submission rate was 56.67% (1020/1800) before intervention and 61.23% (1107/1808) after intervention, and the difference between the two periods was statistically significant (χ2=7.755, P=0.005). The satisfaction rate of doctors and nurses was 75.0% (30/40) before intervention and 95.0% (38/40) after intervention, and the difference between the two periods was statistically significant (χ2=6.275, P=0.012). Conclusions FMEA can effectively find out the weak points in the prevention and treatment of MDRO infections in ICU, while PDCA model can effectively formulate improvement measures for the weak points and control the risks. The combined application of the two modes provides a scientific and effective guarantee for the rational prevention and treatment of MDRO infections in ICU patients.