Acute kidney injury prediction for non-critical care patients: a retrospective external and internal validation study
University of Florida(佛罗里达大学);University of Pittsburgh(匹兹堡大学)
Esra Adiyeke, Yuanfang Ren, Benjamin Shickel, Matthew M. Ruppert, Ziyuan Guan, Sandra L. Kane-Gill, Raghavan Murugan, Nabihah Amatullah, Britney A. Stottlemyer,… 展开作者
Esra Adiyeke, Yuanfang Ren, Benjamin Shickel, Matthew M. Ruppert, Ziyuan Guan, Sandra L. Kane-Gill, Raghavan Murugan, Nabihah Amatullah, Britney A. Stottlemyer, Tiffany L. Tran, Dan Ricketts, Christopher M Horvat, Parisa Rashidi, Azra Bihorac, Tezcan Ozrazgat-Baslanti
背景:急性肾损伤(AKI)是肾脏排泄功能下降,发生在高达18%的住院患者中。AKI的进展可能导致不可逆的肾脏损害。方法:这项回顾性队列研究纳入了在匹兹堡大学医学中心(UPMC)非重症监护病房(n = 46,815)和佛罗里达大学健康中心(UFH)(n = 127,202)住院的成年患者。我们开发并比较了深度学习和传统机器学习模型,以预测未来48小时内进展至2期或更高阶段AKI的情况。我们为每个站点训练了本地模型(UFH模型在UFH上训练,UPMC模型在UPMC上训练),并使用来自两个站点的患者开发队列训练了一个独立模型(UFH-UPMC模型)。我们在每个站点对模型进行了内部和外部验证,并按性别和种族进行了亚组分析。结果:UFH和UPMC患者中分别有3%(n=3,257)和8%(n=2,296)发生2期或更高阶段AKI。UFH测试队列的受试者工作特征曲线下面积(AUROC)值介于0.77(UPMC模型)和0.81(UFH模型)之间,而UPMC测试队列的AUROC值介于0.79(UFH模型)和0.83(UPMC模型)之间。UFH-UPMC模型在UFH测试队列中实现了0.81(95%置信区间[CI] [0.80, 0.83])的AUROC,在UPMC测试队列中实现了0.82(95% CI [0.81, 0.84])的AUROC;精确率-召回率曲线下面积(AUPRC)在UFH测试队列中为0.6(95% CI [0.05, 0.06]),在UPMC测试队列中为0.13(95% CI [0.11, 0.15])。动力学估算的肾小球滤过率、肾毒性药物负担和血尿素氮在各模型和医疗中心中仍然是影响最大的前三个特征。结论:本地开发的模型在另一机构测试时显示出略微降低的判别能力,而影响最大的特征集在各模型和站点中保持一致。
英文摘要:
Background: Acute kidney injury (AKI), the decline of kidney excretory function, occurs in up to 18% of hospitalized admissions. Progression of AKI may lead to irreversible kidney damage. Methods: This retrospective cohort study includes adult patients admitted to a non-intensive care unit at the University of Pittsburgh Medical Center (UPMC) (n = 46,815) and University of Florida Health (UFH) (n = 127,202). We developed and compared deep learning and conventional machine learning models to predict progression to Stage 2 or higher AKI within the next 48 hours. We trained local models for each site (UFH Model trained on UFH, UPMC Model trained on UPMC) and a separate model with a development cohort of patients from both sites (UFH-UPMC Model). We internally and externally validated the models on each site and performed subgroup analyses across sex and race. Results: Stage 2 or higher AKI occurred in 3% (n=3,257) and 8% (n=2,296) of UFH and UPMC patients, respectively. Area under the receiver operating curve values (AUROC) for the UFH test cohort ranged between 0.77 (UPMC Model) and 0.81 (UFH Model), while AUROC values ranged between 0.79 (UFH Model) and 0.83 (UPMC Model) for the UPMC test cohort. UFH-UPMC Model achieved an AUROC of 0.81 (95% confidence interval [CI] [0.80, 0.83]) for UFH and 0.82 (95% CI [0.81,0.84]) for UPMC test cohorts; an area under the precision recall curve values (AUPRC) of 0.6 (95% CI, [0.05, 0.06]) for UFH and 0.13 (95% CI, [0.11,0.15]) for UPMC test cohorts. Kinetic estimated glomerular filtration rate, nephrotoxic drug burden and blood urea nitrogen remained the top three features with the highest influence across the models and health centers. Conclusion: Locally developed models displayed marginally reduced discrimination when tested on another institution, while the top set of influencing features remained the same across the models and sites.
A Method for Characterizing Disease Progression from Acute Kidney Injury to Chronic Kidney Disease
Department of Biomedical Informatics, Columbia University(哥伦比亚大学生物医学信息学系);Department of Medicine, Division of Nephrology, Columbia University(哥伦比亚大学医学系肾脏病科);Department of Medicine, Division of General Medicine, Columbia University(哥伦比亚大学医学系全科医学)
Sum of previous inpatient serum creatinine measurements predicts acute kidney injury in rehospitalized patients
University of Rochester School of Medicine and Dentistry(罗切斯特大学医学与牙科学院);University of Rochester Medical Center(罗切斯特大学医学中心);Rochester Center for Health Informatics(罗切斯特健康信息学中心);University of Rochester(罗切斯特大学);University of Rochester Information Systems Division(罗切斯特大学信息系统部)
Causal-Transformer with Adaptive Mutation-Locking for Early Prediction of Acute Kidney Injury
School of Electrical and Information Engineering, Tianjin University(天津大学电气与信息工程学院);Department of Nephrology, TEDA Hospital, Tianjin University(天津大学TEDA医院肾内科);Tianjin Medical University General Hospital(天津医科大学总医院);Department of Nephrology, First Teaching Hospital of Tianjin University of Traditional Chinese Medicine(天津中医药大学第一教学医院肾内科)
Automated Dynamic Bayesian Networks for Predicting Acute Kidney Injury Before Onset
University of California, Los Angeles(加州大学洛杉矶分校);UCLA Clinical and Translational Science Institute(加州大学洛杉矶分校临床与转化科学研究所);David Geffen School of Medicine at UCLA(加州大学洛杉矶分校大卫·格芬医学院);Providence Health Care(普罗维登斯医疗保健系统)
COVID-Net Biochem: An Explainability-driven Framework to Building Machine Learning Models for Predicting Survival and Kidney Injury of COVID-19 Patients from Clinical and Biochemistry Data
Stony Brook University(石溪大学);University of Waterloo(滑铁卢大学);Waterloo Artificial Intelligence Institute(滑铁卢人工智能研究所);DarwinAI Corp.(达尔文人工智能公司);McGill University(麦吉尔大学)
Causal prediction models for medication safety monitoring: The diagnosis of vancomycin-induced acute kidney injury
Amsterdam UMC(阿姆斯特丹大学医学中心);University of Amsterdam(阿姆斯特丹大学);Amsterdam Public Health(阿姆斯特丹公共卫生研究所);Institute for Logic, Language and Computation(逻辑、语言与计算研究所);Pacmed(Pacmed公司)
Precisely Predicting Acute Kidney Injury with Convolutional Neural Network Based on Electronic Health Record Data
The School of Computer Science and Technology, Xi’an Jiaotong University(西安交通大学计算机科学与技术学院);The School of Life Science and Technology, Xi’an Jiaotong University(西安交通大学生命科学与技术学院);Lishui People’s Hospital(丽水市人民医院)
急性肾损伤(Acute Kidney Injury,AKI)是重症监护病房(Intensive Care Unit,ICU)患者的常见病,尤其是成人患者,是影响近期和远期死亡率的独立危险因素。虽然近年来研究者们重视对AKI的早期预测,但现有模型的性能还不够精确。本研究的目的是通过卷积神经网络对电子健康记录(EHR)数据进行精确预测AKI。本研究中使用的数据集是两个公共电子健康记录(EHR)数据库:MIMIC-III和eICU数据库。在本研究中,我们采用了几个卷积神经网络模型进行训练和测试我们的AKI预测器,它可以准确地预测某一患者是否会发生AKI后,入住ICU根据最后测量的16个血气和人口统计学特征。该研究基于肾脏疾病改善全球结局(KDIGO)标准定义AKI。我们的工作大大提高了AKI的预测精度,最好的AUROC在MIMIC-III数据集上达到0.988,在eICU数据集上达到0.936,两者都优于最先进的预测器。该预测器所用的输入向量的维数远小于已有的研究中所用的维数。与现有的AKI预测器相比,该预测器采用卷积神经网络结构,输入向量更加简洁,大大提高了AKI早期预测的精度。早期、准确地预测AKI的发生,对指导治疗具有重要意义,具有重要的临床应用价值。
Continual Prediction from EHR Data for Inpatient Acute Kidney Injury
University of Wisconsin-Milwaukee(威斯康星大学密尔沃基分校);Aurora Research Institute(奥罗拉研究所);Aurora Health Care(奥罗拉医疗保健);Milwaukee Kidney Associates(密尔沃基肾脏协会);Allina Health System(阿利纳医疗系统)
Machine Learning-Based Prediction of ICU Mortality in Sepsis-Associated Acute Kidney Injury Patients Using MIMIC-IV Database with Validation from eICU Database
University of Southern California(南加州大学);Iran University of Medical Sciences(伊朗医学科学大学);California State University(加利福尼亚州立大学);California State Polytechnic University(加利福尼亚州_polytechnic大学)
Temporal Fusion Nexus: A task-agnostic multi-modal embedding model for clinical narratives and irregular time series in post-kidney transplant care
Hahn-Schickard(哈恩-施克特德研究所);University of Freiburg(弗赖堡大学);Charité University Medical Center(查理医院大学医学中心);Hasso Plattner Institute for Digital Engineering, University of Potsdam(哈索·platner数字工程研究所,波茨坦大学);DFKI(德意志联邦防务研究院);University Hospital Essen(埃森大学医院);University Hospital Erlangen(埃尔兰根大学医院)
Prediction of Significant Creatinine Elevation in First ICU Stays with Vancomycin Use: A retrospective study through Catboost
University of Southern California(美国南加州大学);California State Polytechnic University Pomona(加州州立理工大学庞纳分校);California State University Long Beach(加州州立大学长滩分校)
Population stratification for prediction of mortality in post-AKI patients
Institute of Applied Health Sciences, University of Aberdeen(阿伯丁大学应用卫生科学学院);Institute of Mathematics and Statistics, University of Sao Paulo(圣保罗大学数学与统计学院)
Development and Validation of a Dynamic Kidney Failure Prediction Model based on Deep Learning: A Real-World Study with External Validation
Peking University(北京大学);National University of Singapore(新加坡国立大学);Peking University First Hospital(北京大学第一医院);Chinese Academy of Medical Sciences(中国医学科学院);Ningbo Yinzhou No.2 Hospital(宁波市鄞州第二医院);Yinzhou District Center for Disease Control and Prevention(鄞州区疾病预防控制中心)
Causally-informed Deep Learning towards Explainable and Generalizable Outcomes Prediction in Critical Care
Tsinghua University(清华大学);Institute of Brain and Cognitive Sciences, Tsinghua University(清华大学脑与认知科学研究院);Chinese PLA General Hospital(中国人民解放军总医院);Shanghai Artificial Intelligence Laboratory(上海人工智能实验室)
Yet Another ICU Benchmark: A Flexible Multi-Center Framework for Clinical ML
Hasso Plattner Institute, University of Potsdam(波茨坦大学哈索·普拉特纳研究所);Amsterdam UMC, Vrije Universiteit(阿姆斯特丹自由大学医学中心);Charité - Universitätsmedizin Berlin(柏林夏里特医学院)
A Self-Correcting Deep Learning Approach to Predict Acute Conditions in Critical Care
Nanyang Technological University(南洋理工大学);National University Health System(新加坡国立大学医疗系统);National University of Singapore(新加坡国立大学);Weill Cornell Medical College(威尔康奈尔医学院);Cornell University(康奈尔大学)
Rolling Day-Wise Mortality Prediction in Critically Ill Patients With AKI on CRRT Utilizing Machine Pressure Waveforms
University of Alabama at Birmingham(阿拉巴马大学伯明翰分校);University of Cincinnati(辛辛那提大学);Icahn School of Medicine at Mount Sinai(西奈山伊坎医学院);UT Southwestern Medical Center(德克萨斯大学西南医学中心)
Artificial Intelligence (AI) Based Prediction of Mortality, for COVID-19 Patients
Imam Abdulrahman Bin Faisal University(伊玛目阿卜杜勒拉赫曼·本·费萨尔大学);Ahsanullah University of Science and Technology(阿赫萨努拉科技大学);Health Cluster(卫生集群);Ajman University(阿治曼大学)
Predicting Hyperkalemia in the ICU and Evaluation of Generalizability and Interpretability
The Hong Kong University of Science and Technology(香港科技大学);Massachusetts Institute of Technology(麻省理工学院);Beth Israel Deaconess Medical Center(贝斯以色列女执事医疗中心);Harvard Medical School(哈佛医学院);The Chinese University of Hong Kong(香港中文大学);Philips Research North America(飞利浦北美研究院);Harvard T.H. Chan School of Public Health(哈佛陈曾熙公共卫生学院);University of Helsinki(赫尔辛基大学)
Identifying acute illness phenotypes via deep temporal interpolation and clustering network on physiologic signatures
University of Florida(佛罗里达大学);Intelligent Critical Care Center(智能重症监护中心);College of Pharmacy(药学院);Center for Natural Products, Drug Discovery and Development(天然产物、药物发现与开发中心);J. Crayton Pruitt Family Department of Biomedical Engineering(J.克雷顿·普鲁伊特家族生物医学工程系)