Objective:Focusing on gynecological surgery, we constructed a prediction model for surgical duration by extracting features from unstructured surgical planning texts and integrating multimodal data via artificial intelligence technology. Methods:The clinical data of 34614 patients who underwent gynecologic surgeries at West China Second University Hospital, Sichuan University between January 2022 and October 2024 were collected. An embedding-transformer model was constructed to convert surgical planning texts into a one-dimensional numerical feature, referred to as the step feature. The predictive value of the step feature was assessed by comparing the performance improvements of linear regression, random forest, eXtreme Gradient Boosting (XGBoost), support vector regression, K-nearest neighbor regression, and artificial neural network algorithms in two scenarios-with and without the step feature as an input. The out-of-sample prediction accuracy of the models was assessed using mean absolute error (MAE), root mean squared error (RMSE), and R-squared (R 2). Furthermore, the model interpretability was examined using SHapley Additive exPlanations (SHAP) values. Results:SHAP results showed that the step feature had the highest predictive contribution. Temporal factors in surgical scheduling also influenced gynecological surgery duration. The XGBoost model demonstrated optimal performance on the test set, significantly improving prediction accuracy with a 40.43% increase in R 2, while reducing MAE and RMSE by 21.27% and 20.13%, respectively, compared to the baseline model without the step feature. Conclusion:The embedding-transformer model developed in this study effectively extracts features from surgical planning texts and enhances the predictive performance of machine learning models. The XGBoost prediction model can assist hospital administrators in implementing more refined management of gynecological surgeries and improving the utilization efficiency of surgical resources.

Background:Current lung cancer screening guidelines recommend annual low-dose computed tomography (LDCT) for high-risk individuals. However, the effectiveness of LDCT in non-high-risk individuals remains inadequately explored. With the incidence of lung cancer steadily increasing among non-high-risk individuals, this study aims to assess the risk of lung cancer in non-high-risk individuals and evaluate the potential of thin-section LDCT reconstruction combined with artificial intelligence (LDCT-TRAI) as a screening tool. Methods:A real-world cohort study on lung cancer screening was conducted at the West China Hospital of Sichuan University from January 2010 to July 2021. Participants were screened using either LDCT-TRAI or traditional thick-section LDCT without AI (traditional LDCT) . The AI system employed was the uAI-ChestCare software. Lung cancer diagnoses were confirmed through pathological examination. Results:Among the 259 121 enrolled non-high-risk participants, 87 260 (33.7%) had positive screening results. Within 1 year, 728 (0.3%) participants were diagnosed with lung cancer, of whom 87.1% (634/728) were never-smokers, and 92.7% (675/728) presented with stage I disease. Compared with traditional LDCT, LDCT-TRAI demonstrated a higher lung cancer detection rate (0.3% vs. 0.2%, P < 0.001), particularly for stage I cancers (94.4% vs. 83.2%, P < 0.001), and was associated with improved survival outcomes (5-year overall survival rate: 95.4% vs. 81.3%, P < 0.0001). Conclusion:These findings highlight the importance of expanding lung cancer screening to non-high-risk populations, especially never-smokers. LDCT-TRAI outperformed traditional LDCT in detecting early-stage cancers and improving survival outcomes, underscoring its potential as a more effective screening tool for early lung cancer detection in this population.

IntroductionWe aimed to explore the prognostic value of the aspartate aminotransferase to alanine aminotransferase (AST/ALT) ratio in non-surgical patients with type 2 diabetes hospitalized for heart failure.Material and methodsUsing a large electronic medical record-based cohort of diabetes in China (WECODe), we gathered data on non-surgical hospitalized patients with type 2 diabetes and heart failure from 2011 to 2019. Baseline AST/ALT ratio was calculated. The primary outcomes were all-cause death within 30 days after discharge, composite cardiac events, major acute kidney injury, and major systemic infection. A multivariable Cox proportional regression model was utilized to evaluate the association between the AST/ALT ratio and outcomes.ResultsThis retrospective cohort included 8,073 patients (39.4% women) with type 2 diabetes hospitalized for heart failure. The median age was 71 years. Higher AST/ALT ratio was associated with higher risks of poor endpoints (with per standard deviation increment in AST/ALT ratio, for death within 30 days after discharge: adjusted hazard ratio [HR], 1.35, 95% confidence interval [CI], 1.21 to 1.50; for composite cardiac events: HR, 1.18, 95% CI: 1.06 to 1.31). Compared to patients in the lowest quartile for the AST/ALT ratio, those in the highest quartile have elevated risk of death within 30 days after discharge and major systemic infection (HRs [95% CIs]: 1.61 [1.18 to 2.19] and 1.28 [1.06 to 1.56], respectively). Subgroup analyses and sensitivity analyses confirmed the robustness of the findings.ConclusionsType 2 diabetes patients hospitalized for heart failure with the AST/ALT ratio in the highest quartile face a poor short-term prognosis.

AIMS To investigate the sex-specific associations between predicted skeletal muscle mass index (pSMI) and incident type 2 diabetes in a retrospective longitudinal cohort of Chinese men and women. MATERIALS AND METHODS We enrolled Chinese adults without diabetes at baseline from WATCH (West chinA adulT health CoHort), a large health check-up-based database. We calculated pSMI to estimate skeletal muscular mass, and measured blood glucose variables and assessed self-reported history to identify new-onset diabetes. The nonlinear association between pSMI and incident type 2 diabetes was modelled using the penalized spline method. The piecewise association was estimated using segmented linear splines in weighted Cox proportional hazards regression models. RESULTS Of 47 885 adults (53.2% women) with a median age of 40 years, 1836 developed type 2 diabetes after a 5-year median follow-up. In women, higher pSMI was associated with a lower risk of incident type 2 diabetes (Pnonlinearity  = 0.09, hazard ratio [HR] per standard deviation increment in pSMI: 0.79 [95% confidence interval {CI} 0.68, 0.91]). A nonlinear association of pSMI with incident type 2 diabetes was detected in men (Pnonlinearity  < 0.001). In men with pSMI lower than 8.1, higher pSMI was associated with a lower risk of incident type 2 diabetes (HR 0.58 [95% CI 0.40, 0.84]), whereas pSMI was not significantly associated with incident diabetes in men with pSMI equal to or greater than 8.1 (HR 1.08 [95% CI 0.93, 1.25]). CONCLUSIONS In females, a larger muscular mass is associated with a lower risk of type 2 diabetes. For males, this association is significant only among those with diminished muscle mass.

目的 建立医联体统一的大数据中心,集成医联体管理及医疗相关数据,为医联体成员统一提供数据应用服务.方法 围绕应用需求,以及数据所需的时效性,建立不同的采集模式,对数据进行采集、清洗、集成后,形成医联体大数据中心,并为相关医联体应用提供数据服务.结果 建立了医联体大数据中心,采集数据量超过270TB,为11家医联体成员提供医联体指标集驾驶舱服务和统一的疾病诊断相关分组(DRG)指标计算服务,为5家医联体成员提供医联体患者全息视图服务,极大地提升了医联体的管理能力和医疗水平.结论 以应用为驱动的医联体数据统一应用平台,能够极大地节约医联体管理成本,同时有效推动区域大数据中心建设,提供统一的数据服务.

目的:为患者更及时、精准地匹配对应亚专业,开发结合人工智能的互联网智能分诊平台.方法:基于大数据和深度学习技术,依据疾病诊断分类和知识库,将患者与科室、医生实现精准匹配,实现智能化的分诊功能.结果:互联网智能分诊平台已经在某大型三甲医院运行,实现了患者的精准诊疗,极大地提升了诊疗效率和质量.结论:基于人工智能的互联网智能分诊平台作为一种新型的医院智能化、智慧化应用,不仅可显著提升互联网诊疗效率,提升医疗服务水平,同时提升了患者及医务人员的满意度,具有广泛的应用前景.

目的:探索大型医疗机构医疗健康大数据平台的建设路径,为医疗机构推进数据平台建设提供参考.方法:梳理、总结四川大学华西医院医疗健康大数据平台建设过程中的5个实施阶段,指出各个阶段的重要任务和实施难点;分析当前医疗健康大数据平台建设中存在的不足,明确未来持续优化的方向.结果:根据四川大学华西医院医疗健康大数据平台的建设经验,提供各个实施阶段的工作参考范式.结论:医疗健康大数据平台的建设有赖于多个学科的融合,需要各领域专家的深度合作.

AIMS:This study aimed to investigate the influencing factors of frailty in elderly patients with multimorbidity and to develop a predictive risk model for frailty in elderly patients with multimorbidity. METHODS:In total, 3836 elderly patients with multimorbidity who were admitted to the medical wards of five grade A tertiary hospitals in Sichuan Province from March 2020 to June 2021 were selected. Based on the general data of patients with multimorbidity, the independent risk factors for frailty were obtained using logistic analysis, and a risk prediction model of frailty was developed. RESULTS:Independent risk factors for frailty in patients with multimorbidity were age, types of medication, and comorbidity with chronic heart failure (CHF), chronic obstructive pulmonary disease (COPD) and chronic cerebrovascular disease (CCVD); and the protective factors for frailty were body mass index (BMI), exercise and education level. The expression of the model was Z = -2.054 + 0.016 × age - 0.029 × BMI - 0.153 × education level-1.059 × exercise + 0.203 × types of medication + 0.788 × comorbidity with CHF + 0.950 × comorbidity with COPD + 0.363 × comorbidity with CCVD. CONCLUSION:Age, BMI, education level, exercise, types of medication, and comorbidity with CHF, COPD and CCVD can affect frailty risk in elderly patients with multimorbidity, which may be helpful to predict the frailty risk of elderly patients with multimorbidity. Geriatr Gerontol Int 2022; 22: 471-476.

AimThe aim of this study was to compare the safety and overall effect of robotic distal pancreatectomy (RDP) to laparoscopic distal pancreatectomy (LDP) after the learning curve, especially in perioperative outcome and short-term oncological outcome.MethodsA literature search was performed by two authors independently using PubMed, Embase, and Web of Science to identify any studies comparing the results of RDP versus LDP published until 5 January 2022. Only the studies where RDP was performed in more than 35 cases were included in this study. We performed a meta-analysis of operative time, blood loss, reoperation, readmission, hospital stay, overall complications, major complications, postoperative pancreatic fistula (POPF), blood transfusion, conversion to open surgery, spleen preservation, tumor size, R0 resection, and lymph node dissection.ResultsOur search identified 15 eligible studies, totaling 4,062 patients (1,413 RDP). It seems that the RDP group had a higher rate of smaller tumor size than the LDP group (MD: −0.15; 95% CI: −0.20 to −0.09; p < 0.00001). Furthermore, compared with LPD, RDP was associated with a higher spleen preservation rate (OR: 2.19; 95% CI: 1.36–3.54; p = 0.001) and lower rate of conversion to open surgery (OR: 0.43; 95% CI: 0.33–0.55; p < 0.00001). Our study revealed that there were no significant differences in operative time, overall complications, major complications, blood loss, blood transfusion, reoperation, readmission, POPF, and lymph node dissection between RDP and LDP.ConclusionsRDP is safe and feasible for distal pancreatectomy compared with LDP, and it can reduce the rate of conversion to open surgery and increase the rate of spleen preservation, which needs to be further confirmed by quality comparative studies with large samples.Systematic Review Registrationhttps://www.crd.york.ac.uk/PROSPERO/#recordDetails.

Background: With the advent of data-intensive science, a full integration of big data science and health care will bring a cross-field revolution to the medical community in China The concept big data represents not only a technology but also a resource and a method. Big data are regarded as an important strategic resource both at the national level and at the medical institutional level, thus great importance has been attached to the construction of a big data platform for health care. Objective: We aimed to develop and implement a big data platform for a large hospital, to overcome difficulties in integrating, calculating, storing, and governing multisource heterogeneous data in a standardized way, as well as to ensure health care data security. Methods: The project to build a big data platform at West China Hospital of Sichuan University was launched in 2017. The West China Hospital of Sichuan University big data platform has extracted, integrated, and governed data from different departments and sections of the hospital since January 2008. A master-slave mode was implemented to realize the real-time integration of multisource heterogeneous massive data, and an environment that separates heterogeneous characteristic data storage and calculation processes was built. A business-based metadata model was improved for data quality control, and a standardized health care data governance system and scientific closed-loop data security ecology were established. Results: After 3 years of design, development, and testing, the West China Hospital of Sichuan University big data platform was formally brought online in November 2020. It has formed a massive multidimensional data resource database, with more than 12.49 million patients, 75.67 million visits, and 8475 data variables. Along with hospital operations data, newly generated data are entered into the platform in real time. Since its launch, the platform has supported more than 20 major projects and provided data service, storage, and computing power support to many scientific teams, facilitating a shift in the data support model-from conventional manual extraction to self-service retrieval (which has reached 8561 retrievals per month). Conclusions: The platform can combine operation systems data from all departments and sections in a hospital to form a massive high-dimensional high-quality health care database that allows electronic medical records to be used effectively and taps into the value of data to fully support clinical services, scientific research, and operations management. The West China Hospital of Sichuan University big data platform can successfully generate multisource heterogeneous data storage and computing power. By effectively governing massive multidimensional data gathered from multiple sources, the West China Hospital of Sichuan University big data platform provides highly available data assets and thus has a high application value in the health care field. The West China Hospital of Sichuan University big data platform facilitates simpler and more efficient utilization of electronic medical record data for real-world research.

Background Myocardial infarction can lead to malignant arrhythmia, heart failure, and sudden death. Clinical studies have shown that early identification of and timely intervention for acute MI can significantly reduce mortality. The traditional MI risk assessment models are subjective, and the data that go into them are difficult to obtain. Generally, the assessment is only conducted among high-risk patient groups. Objective To construct an artificial intelligence-based risk prediction model of myocardial infarction (MI) for continuous and active monitoring of inpatients, especially those in noncardiovascular departments, and early warning of MI. Methods The imbalanced data contain 59 features, which were constructed into a specific dataset through proportional division, upsampling, downsampling, easy ensemble, and w-easy ensemble. Then, the dataset was traversed using supervised machine learning, with recursive feature elimination as the top-layer algorithm and random forest, gradient boosting decision tree (GBDT), logistic regression, and support vector machine as the bottom-layer algorithms, to select the best model out of many through a variety of evaluation indices. Results GBDT was the best bottom-layer algorithm, and downsampling was the best dataset construction method. In the validation set, the F1 score and accuracy of the 24-feature downsampling GBDT model were both 0.84. In the test set, the F1 score and accuracy of the 24-feature downsampling GBDT model were both 0.83, and the area under the curve was 0.91. Conclusion Compared with traditional models, artificial intelligence-based machine learning models have better accuracy and real-time performance and can reduce the occurrence of in-hospital MI from a data-driven perspective, thereby increasing the cure rate of patients and improving their prognosis.

To evaluate the impact of stress hyperglycemia on the in-hospital prognosis in non-surgical patients with heart failure and type 2 diabetes. We identified non-surgical hospitalized patients with heart failure and type 2 diabetes from a large electronic medical record-based database of diabetes in China (WECODe) from 2011 to 2019. We estimated stress hyperglycemia using the stress hyperglycemia ratio (SHR) and its equation, say admission blood glucose/[(28.7 × HbA1c)− 46.7]. The primary outcomes included the composite cardiac events (combination of death during hospitalization, requiring cardiopulmonary resuscitation, cardiogenic shock, and the new episode of acute heart failure during hospitalization), major acute kidney injury (AKI stage 2 or 3), and major systemic infection. Of 2875 eligible Chinese adults, SHR showed U-shaped associations with composite cardiac events, major AKI, and major systemic infection. People with SHR in the third tertile (vs those with SHR in the second tertile) presented higher risks of composite cardiac events ([odds ratio, 95

目的 通过集成院内疫情防控相关业务系统的数据,建设疫情防控大数据中心和疫情防控平台,支撑常态化疫情防控工作.方法 针对疫情防控需要的人员信息、诊疗信息、位移信息及物资信息,整合医院信息系统、探视系统、门禁系统、医院资源规划等业务系统的数据资源,经过数据治理,形成疫情防控大数据中心,依托数据中心,建设疫情防控平台,支撑疫前、疫中、疫后各类情况的预警、定位和管理等.结果 基于院内大数据平台,分别在门诊部、医务部、院长办公室、护理部部署了常态化疫情防控平台的相关应用,提前向管理部门发布健康码异常人员信息4667条,其中红色健康码异常人员信息522条,2020年12月7日,定位院内一名疑似新冠患者移动轨迹,反馈近100名该患者在院内的接触人员信息.结论 通过建设统一数据中心的疫情防控平台,能够针对性的整合所有的信息资源,在常态化疫情防控中提供更加有效的手段.

深度学习作为人工智能领域最为活跃的研究分支,近年来在计算机视觉、自然语言处理、语音识别等领域取得丰硕成果.同时,深度学习在医疗领域中的应用也逐渐成为研究热点,并且在医学图像和信号处理、计算机辅助检测与诊断、临床决策支持、医疗信息挖掘和检索等方面取得了一些成功,展现出了极大的应用前景.本文在介绍深度学习原理和常用深度神经网络的同时,结合相关文献和应用实践,对深度学习在医疗系统中的应用场景和研究进展做了全面系统性地介绍.同时,本文还探讨了深度学习在现代医疗领域实施的难点和挑战,并针对性地给出了一些解决方案或解决思路.

AbstractObjectiveTo assess the efficacy and safety of hypotensive anesthesia (HA) combined with tranexamic acid (TXA) for reducing perioperative blood loss in simultaneous bilateral total hip arthroplasty (SBTHA).MethodsIn this retrospective cohort study, a total of 183 eligible patients (15 females and 168 males, 44.01 ± 9.29 years old) who underwent SBTHA from January 2015 to September 2020 at our medical center were enrolled for analysis. Fifty‐nine patients received standard general anesthesia (Std‐GA group), the other 85 and 39 patients received HA with an intraoperative mean arterial pressure between 70 and 80 mmHg (70–80 HA group) and below 70 mmHg (<70 HA group), respectively. TXA was administrated to all patients. Perioperative blood loss (total, dominant, and hidden), transfusion rate and volume, hemoglobin and hematocrit reduction, duration of operation and anesthesia, length of hospitalization, range of hip motion as well as postoperative complications were collected from hospital's electronic records and compared between groups.ResultsAll patients were followed for more than 3 months. Total blood loss in the two HA groups (1390.25 ± 595.67 ml and 1377.74 ± 423.46 ml, respectively) was significantly reduced compared with that in Std‐GA group (1850.83 ± 800.73 ml, P < 0.001). Both dominant and hidden blood loss were dramatically decreased when HA was applied (both P < 0.001). Accordingly, the transfusion rate along with volume in 70–80 HA group (14.1%, 425.00 ± 128.81 ml) and <70 HA group (12.8%, 340.00 ± 134.16 ml) were reduced in comparison with those in Std‐GA group (37.3%, 690.91 ± 370.21ml; P = 0.001 and P = 0.014, respectively). The maximal hemoglobin and hematocrit reduction in both HA groups were significantly less than those in Std‐GA group (both P < 0.001). Of note, 70–80 and <70 HA groups exhibited comparable efficacy with no significant differences between them. Besides, significant difference in duration of surgery was found among groups (P = 0.044 and P < 0.001), while no differences in anesthesia time and postoperative range of hip motion were observed. Regarding complications, the incidence of both acute kidney injury and postoperative hypotension in <70 HA group was significantly higher than that in 70–80 HA and Std‐GA groups (P = 0.014 and P < 0.001). Incidence of acute myocardial injury was similar among groups (P = 0.099) and no other severe complications or mortality were recorded.ConclusionThe combination of HA with a mean arterial pressure (MAP) of 70–80 mmHg and TXA could significantly reduce blood loss and transfusion during SBTHA, in addition to shortening operation time and length of hospitalization, and with no increase in complications.

Our aim was to investigate the association of glycated haemoglobin A1c (HbA1c) variability score (HVS) with estimated glomerular filtration rate (eGFR) slope in Chinese adults living with type 2 diabetes. This cohort study included adults with type 2 diabetes attending outpatient clinics between 2011 and 2019 from a large electronic medical record-based database of diabetes in China (WECODe). We estimated the individual-level visit-to-visit HbA1c variability using HVS, a proportion of changes in HbA1c of ≥0.5% (5.5 mmol/mol). We estimated the odds of people experiencing a rapid eGFR annual decline using a logistic regression and differences across HVS categories in the mean eGFR slope using a mixed-effect model. The analysis involved 2397 individuals and a median follow-up of 4.7 years. Compared with people with HVS ≤ 20%, those with HVS of 60% to 80% had 11% higher odds of experiencing rapid eGFR annual decline, with an extra eGFR decline of 0.93 mL/min/1.73 m2 per year on average; those with HVS > 80% showed 26% higher odds of experiencing a rapid eGFR annual decline, with an extra decline of 1.83 mL/min/1.73 m2 per year on average. Chinese adults with type 2 diabetes and HVS > 60% could experience a more rapid eGFR decline.

目的 总结急性心肌梗死合并室间隔穿孔(AMI-VSR)患者的临床特点及预后.方法 回顾性分析四川大学华西医院2012年1月—2021年8月收治的AMI-VSR患者65例,描述其基线资料特征、穿孔性质和治疗情况;进行电话随访,收集其预后情况.并对以上指标进行多因素logistic回归分析.结果 65例AMI-VSR患者,继发于前壁心肌梗死患者55例(84.6％),其AMI-VSR时间为0.25～15.00 d[(3.9±3.4)d];其中院内死亡40例(61.5％),出院后1年内死亡11例(16.9％),好转出院并且存活至今11例(16.9％),失访3例(4.6％).其中手术治疗6例,院内死亡0例,出院后1年内死亡0例,全部6例患者存活至今.接受外科手术或介入治疗患者院内死亡率以及出院后1年内死亡率显著低于保守治疗患者(P<0.05).多因素logistic回归结果显示AMI患者的TIMI评分≥7分是AMI-VSR患者院内以及出院后1年内死亡的独立危险因素(OR=30.034,95％CI 1.65～546.70,P<0.05).结论 VSR是AMI严重的并发症;AMI的TIMI评分≥7分是AMI-VSR患者院内及出院后1年内死亡的独立危险因素.外科手术或介入封堵治疗可显著改善患者的近期以及远期预后,AMI后2周以上行外科手术治疗成功率高,对于极其危重的患者,可适当提前外科手术时间.具体更有说服力的结论仍需进一步的临床研究.

Background Artificial intelligence-based disease prediction models have a greater potential to screen COVID-19 patients than conventional methods. However, their application has been restricted because of their underlying black-box nature. Objective To addressed this issue, an explainable artificial intelligence (XAI) approach was developed to screen patients for COVID-19. Methods A retrospective study consisting of 1,737 participants (759 COVID-19 patients and 978 controls) admitted to San Raphael Hospital (OSR) from February to May 2020 was used to construct a diagnosis model. Finally, 32 key blood test indices from 1,374 participants were used for screening patients for COVID-19. Four ensemble learning algorithms were used: random forest (RF), adaptive boosting (AdaBoost), gradient boosting decision tree (GBDT), and extreme gradient boosting (XGBoost). Feature importance from the perspective of the clinical domain and visualized interpretations were illustrated by using local interpretable model-agnostic explanations (LIME) plots. Results The GBDT model [area under the curve (AUC): 86.4%; 95% confidence interval (CI) 0.821–0.907] outperformed the RF model (AUC: 85.7%; 95% CI 0.813–0.902), AdaBoost model (AUC: 85.4%; 95% CI 0.810–0.899), and XGBoost model (AUC: 84.9%; 95% CI 0.803–0.894) in distinguishing patients with COVID-19 from those without. The cumulative feature importance of lactate dehydrogenase, white blood cells, and eosinophil counts was 0.145, 0.130, and 0.128, respectively. Conclusions Ensemble machining learning (ML) approaches, mainly GBDT and LIME plots, are efficient for screening patients with COVID-19 and might serve as a potential tool in the auxiliary diagnosis of COVID-19. Patients with higher WBC count, higher LDH level, or higher EOT count, were more likely to have COVID-19.

四川大学华西医院基于大数据集成平台,设计了患者全程随访管理系统,以患者为中心,整合了患者院内院外的健康数据,建立了全生命周期的医疗服务体系,管理系统为患者提供"健康-疾病-愈后-随访"的精细化诊疗服务,优化患者体验.本文从需求分析、数据集成、系统架构、功能模块、应用效果5个方面对随访系统进行介绍,并总结了在院内肺结节项目组的应用成效,证实了该系统确实提升了医疗服务的专业性和患者的就医体验,以及全程随访管理平台的现状和未来需改进的要点.