重症大数据应用中国专家共识 2022

[1]Sanchez-PintoLN, LuoY, ChurpekMM. Big data and data science in critical Care[J]. Chest, 2018, 154(5):1239-1248. DOI: 10.1016/j.chest.2018.04.037.

[2]RumsfeldJS, JoyntKE, MaddoxTM. Big data analytics to improve cardiovascular care: promise and challenges[J]. Nat Rev Cardiol, 2016, 13(6):350-359. DOI: 10.1038/nrcardio.2016.42.

[3]YangS, StansburyLG, RockP, et al. Linking big data and prediction strategies: tools, pitfalls, and lessons learned[J]. Crit Care Med, 2019, 47(6):840-848. DOI: 10.1097/CCM.0000000000003739.

[4]CarraG, SalluhJ, da Silva RamosFJ, et al. Data-driven ICU management: using big data and algorithms to improve outcomes[J]. J Crit Care, 2020, 60:300-304. DOI: 10.1016/j.jcrc.2020.09.002.

[5]Le RouxP, MenonDK, CiterioG, et al. Consensus summary statement of the International Multidisciplinary Consensus Conference on Multimodality Monitoring in Neurocritical Care: a statement for healthcare professionals from the Neurocritical Care Society and the European Society of Intensive Care Medicine[J]. Neurocrit Care, 2014, 21Suppl 2:S1-S26. DOI: 10.1007/s12028-014-0041-5.

[6]SchmidtJM, De GeorgiaM. Multimodality monitoring: informatics, integration data display and analysis[J]. Neurocrit Care, 2014, 21Suppl 2:S229-S238. DOI: 10.1007/s12028-014-0037-1.

[7]

CiterioG, ParkS, SchmidtJM, et al. Data collection and interpretation[J]. Neurocrit Care, 2015, 22(3):360-368. DOI: 10.1007/s12028-015-0139-4.

[8]DochertyAB, LoneNI. Exploiting big data for critical care research[J]. Curr Opin Crit Care, 2015, 21(5):467-472. DOI: 10.1097/MCC.0000000000000228.

[9]ZampieriFG, SoaresM. Reply to: the Epimed Monitor ICU Database®: a cloud-based national registry for adult intensive care unit patients in Brazil[J]. Rev Bras Ter Intensiva, 2018, 30(3):398. DOI: 10.5935/0103-507x.20180048.

[10]KomorowskiM, CeliLA, BadawiO, et al. The artificial intelligence clinician learns optimal treatment strategies for sepsis in intensive care[J]. Nat Med, 2018, 24(11):1716-1720. DOI: 10.1038/s41591-018-0213-5.

[11]NematiS, HolderA, RazmiF, et al. An interpretable machine learning model for accurate prediction of sepsis in the ICU[J]. Crit Care Med, 2018, 46(4):547-553. DOI: 10.1097/CCM.0000000000002936.

[12]CeliLA, MarkRG, StoneDJ, et al. “Big data” in the intensive care unit. Closing the data loop[J]. Am J Respir Crit Care Med, 2013, 187(11):1157-1160. DOI: 10.1164/rccm.201212-2311ED.

[13]GoldbergerAL, AmaralLA, GlassL, et al. PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals[J]. Circulation, 2000, 101(23):E215-E220. DOI: 10.1161/01.cir.101.23.e215.

[14]PollardTJ, JohnsonA, RaffaJD, et al. The eICU Collaborative Research Database, a freely available multi-center database for critical care research[J]. Sci Data, 2018, 5:180178. DOI: 10.1038/sdata.2018.178.

[15]HylandSL, FaltysM, HüserM, et al. Early prediction of circulatory failure in the intensive care unit using machine learning[J]. Nat Med, 2020, 26(3):364-373. DOI: 10.1038/s41591-020-0789-4.

[16]ThoralPJ, PeppinkJM, DriessenRH, et al. Sharing ICU Patient Data Responsibly Under the Society of Critical Care Medicine/European Society of Intensive Care Medicine Joint Data Science Collaboration: the Amsterdam University Medical Centers Database (AmsterdamUMCdb) Example[J]. Crit Care Med, 2021, 49(6):e563-e577. DOI: 10.1097/CCM.0000000000004916.

[17]QiS, MaoZ, HuX, et al. Introduction of critical care database based on specialized information systems: a model of critical care medicine database in large Level Ⅲ Grade A hospital[J]. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue, 2020, 32(6):743-749. DOI: 10.3760/cma.j.cn121430-20200520-00393.

[18]XuP, ChenL, ZhuY, et al. Critical care database comprising patients with infection[J]. Front Public Health, 2022, 10:852410. DOI: 10.3389/fpubh.2022.852410.

[19]ZengX, YuG, LuY, et al. PIC, a paediatric-specific intensive care database[J]. Sci Data, 2020, 7(1):14. DOI: 10.1038/s41597-020-0355-4.

[20]ZhangZ, CaoL, ChenR, et al. Electronic healthcare records and external outcome data for hospitalized patients with heart failure[J]. Sci Data, 2021, 8(1):46. DOI: 10.1038/s41597-021-00835-9.

[21]许杰, 周瑜, 夏星球, 等. 重症医学专科大数据平台的建设及应用[J]. 中华急诊医学杂志, 2022, 31(1):129-132. DOI: 10.3760/cma.j.issn.1671-0282.2022.01.028.

[22]MorrisonJL, CaiQ, DavisN, et al. Clinical and economic outcomes of the electronic intensive care unit: results from two community hospitals[J]. Crit Care Med, 2010, 38(1):2-8. DOI: 10.1097/CCM.0b013e3181b78fa8.

[23]ElbersPW, GirbesA, MalbrainML, et al. Right dose, right now: using big data to optimize antibiotic dosing in the critically ill[J]. Anaesthesiol Intensive Ther, 2015, 47(5):457-463. DOI: 10.5603/AIT.a2015.0061.

[24]KindleRD, BadawiO, CeliLA, et al. Intensive care unit telemedicine in the era of big data, artificial intelligence, and computer clinical decision support systems[J]. Crit Care Clin, 2019, 35(3):483-495. DOI: 10.1016/j.ccc.2019.02.005.

[25]NoshadM, RoseCC, ChenJH. Signal from the noise: a mixed graphical and quantitative process mining approach to evaluate care pathways applied to emergency stroke care[J]. J Biomed Inform, 2022, 127:104004. DOI: 10.1016/j.jbi.2022.104004.

[26]BossJM, NarulaG, StraessleC, et al. ICU Cockpit: a platform for collecting multimodal waveform data, AI-based computational disease modeling and real-time decision support in the intensive care unit[J]. J Am Med Inform Assoc, 2022, 29(7):1286-1291. DOI: 10.1093/jamia/ocac064.

[27]TaglangG, JacksonDB. Use of “big data” in drug discovery and clinical trials[J]. Gynecol Oncol, 2016, 141(1):17-23. DOI: 10.1016/j.ygyno.2016.02.022.

[28]ZhuY, YinH, ZhangR, et al. The effect of dobutamine vs milrinone in sepsis: a big data, real-world study[J]. Int J Clin Pract, 2021, 75(11):e14689. DOI: 10.1111/ijcp.14689.

[29]HuangX, ShanS, KhanYA, et al. Risk assessment of ICU patients through deep learning technique: a big data approach[J]. J Glob Health, 2022, 12:04044. DOI: 10.7189/jogh.12.04044.

[30]XiaY, WangX, WuW, et al. Rehabilitation of sepsis patients with acute kidney injury based on intelligent medical big data[J]. J Healthc Eng, 2022, 2022:8414135. DOI: 10.1155/2022/8414135.

[31]VergetisV, SkaltsasD, GorgoulisVG, et al. Assessing Drug Development Risk Using Big Data and Machine Learning[J]. Cancer Res, 2021, 81(4):816-819. DOI: 10.1158/0008-5472.CAN-20-0866.

[32]FleurenLM, KlauschT, ZwagerCL, et al. Machine learning for the prediction of sepsis: a systematic review and meta-analysis of diagnostic test accuracy[J]. Intensive Care Med, 2020, 46(3):383-400. DOI: 10.1007/s00134-019-05872-y.

[33]DelahantyRJ, AlvarezJ, FlynnLM, et al. Development and evaluation of a machine learning model for the early identification of patients at risk for sepsis[J]. Ann Emerg Med, 2019, 73(4):334-344. DOI: 10.1016/j.annemergmed.2018.11.036.

[34]GohKH, WangL, YeowA, et al. Artificial intelligence in sepsis early prediction and diagnosis using unstructured data in healthcare[J]. Nat Commun, 2021, 12(1):711. DOI: 10.1038/s41467-021-20910-4.

[35]WardiG, BriceJ, CorreiaM, et al. Demystifying lactate in the emergency department[J]. Ann Emerg Med, 2020, 75(2):287-298. DOI: 10.1016/j.annemergmed.2019.06.027.

[36]Ozrazgat-BaslantiT, LoftusTJ, RenY, et al. Advances in artificial intelligence and deep learning systems in ICU-related acute kidney injury[J]. Curr Opin Crit Care, 2021, 27(6):560-572. DOI: 10.1097/MCC.0000000000000887.

[37]XiaoZ, HuangQ, YangY, et al. Emerging early diagnostic methods for acute kidney injury[J]. Theranostics, 2022, 12(6):2963-2986. DOI: 10.7150/thno.71064.

[38]ChurpekMM, CareyKA, EdelsonDP, et al. Internal and external validation of a machine learning risk score for acute kidney injury[J]. JAMA Netw Open, 2020, 3(8):e2012892. DOI: 10.1001/jamanetworkopen.2020.12892.

[39]KateRJ, PearceN, MazumdarD, et al. A continual prediction model for inpatient acute kidney injury[J]. Comput Biol Med, 2020, 116:103580. DOI: 10.1016/j.compbiomed.2019.103580.

[40]LeS, PellegriniE, Green-SaxenaA, et al. Supervised machine learning for the early prediction of acute respiratory distress syndrome (ARDS)[J]. J Crit Care, 2020, 60:96-102. DOI: 10.1016/j.jcrc.2020.07.019.

[41]MayampurathA, ChurpekMM, SuX, et al. External validation of an acute respiratory distress syndrome prediction model using radiology reports[J]. Crit Care Med, 2020, 48(9):e791-e798. DOI: 10.1097/CCM.0000000000004468.

[42]SjodingMW, TaylorD, MotykaJ, et al. Deep learning to detect acute respiratory distress syndrome on chest radiographs: a retrospective study with external validation[J]. Lancet Digit Health, 2021, 3(6):e340-e348. DOI: 10.1016/S2589-7500(21)00056-X.

[43]LinnenDT, EscobarGJ, HuX, et al. Statistical modeling and aggregate-weighted scoring systems in prediction of mortality and ICU transfer: a systematic review[J]. J Hosp Med, 2019, 14(3):161-169. DOI: 10.12788/jhm.3151.

[44]MuralitharanS, NelsonW, DiS, et al. Machine learning-based early warning systems for clinical deterioration: systematic scoping review[J]. J Med Internet Res, 2021, 23(2):e25187. DOI: 10.2196/25187.

[45]TisdaleJE, JaynesHA, OverholserBR, et al. Enhanced response to drug-induced QT interval lengthening in patients with heart failure with preserved ejection fraction[J]. J Card Fail, 2020, 26(9):781-785. DOI: 10.1016/j.cardfail.2020.06.008.

[46]Broch PorcarMJ, Rodríguez CubilloB, Domínguez-RoldánJM, et al. Practical document on the management of hyponatremia in critically ill patients[J]. Med Intensiva (Engl Ed), 2019, 43(5):302-316. DOI: 10.1016/j.medin.2018.12.002.

[47]LeviR, CarliF, ArévaloAR, et al. Artificial intelligence-based prediction of transfusion in the intensive care unit in patients with gastrointestinal bleeding[J]. BMJ Health Care Inform, 2021, 28(1)DOI: 10.1136/bmjhci-2020-100245.

[48]RyanL, MatarasoS, SiefkasA, et al. A machine learning approach to predict deep venous thrombosis among hospitalized patients[J]. Clin Appl Thromb Hemost, 2021, 27:1076029621991185. DOI: 10.1177/1076029621991185.

[49]YuanKC, TsaiLW, LeeKH, et al. The development an artificial intelligence algorithm for early sepsis diagnosis in the intensive care unit[J]. Int J Med Inform, 2020, 141:104176. DOI: 10.1016/j.ijmedinf.2020.104176.

[50]BurdickH, PinoE, Gabel-ComeauD, et al. Effect of a sepsis prediction algorithm on patient mortality, length of stay and readmission: a prospective multicentre clinical outcomes evaluation of real-world patient data from US hospitals[J]. BMJ Health Care Inform, 2020, 27(1)DOI: 10.1136/bmjhci-2019-100109.

[51]BoseSN, GreensteinJL, FacklerJC, et al. Early prediction of multiple organ dysfunction in the pediatric intensive care unit[J]. Front Pediatr, 2021, 9:711104. DOI: 10.3389/fped.2021.711104.

[52]Romero-BrufauS, WhitfordD, JohnsonMG, et al. Using machine learning to improve the accuracy of patient deterioration predictions: Mayo Clinic Early Warning Score (MC-EWS)[J]. J Am Med Inform Assoc, 2021, 28(6):1207-1215. DOI: 10.1093/jamia/ocaa347.

[53]BedoyaAD, ClementME, PhelanM, et al. Minimal impact of implemented early warning score and best practice alert for patient deterioration[J]. Crit Care Med, 2019, 47(1):49-55. DOI: 10.1097/CCM.0000000000003439.

[54]ArdilaD, KiralyAP, BharadwajS, et al. Author correction: end-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography[J]. Nat Med, 2019, 25(8):1319. DOI: 10.1038/s41591-019-0536-x.

[55]QinZZ, SanderMS, RaiB, et al. Using artificial intelligence to read chest radiographs for tuberculosis detection: a multi-site evaluation of the diagnostic accuracy of three deep learning systems[J]. Sci Rep, 2019, 9(1):15000. DOI: 10.1038/s41598-019-51503-3.

[56]ReamaroonN, SjodingMW, GryakJ, et al. Automated detection of acute respiratory distress syndrome from chest X-Rays using Directionality Measure and deep learning features[J]. Comput Biol Med, 2021, 134:104463. DOI: 10.1016/j.compbiomed.2021.104463.

[57]RueckelJ, KunzWG, HoppeBF, et al. Artificial intelligence algorithm detecting lung infection in supine chest radiographs of critically ill patients with a diagnostic accuracy similar to board-certified radiologists[J]. Crit Care Med, 2020, 48(7):e574-e583. DOI: 10.1097/CCM.0000000000004397.

[58]ZhangK, LiuX, ShenJ, et al. Clinically applicable ai system for accurate diagnosis, quantitative measurements, and prognosis of COVID-19 pneumonia using computed tomography[J]. Cell, 2020, 182(5):1360. DOI: 10.1016/j.cell.2020.08.029.

[59]FarzanehN, WilliamsonCA, JiangC, et al. Automated segmentation and severity analysis of subdural hematoma for patients with traumatic brain injuries[J]. Diagnostics (Basel), 2020, 10(10)DOI: 10.3390/diagnostics10100773.

[60]van SlounR, DemiL. Localizing B-lines in lung ultrasonography by weakly supervised deep learning, in-vivo results[J]. IEEE J Biomed Health Inform, 2020, 24(4):957-964. DOI: 10.1109/JBHI.2019.2936151.

[61]SilvaS, Ait AissaD, CocquetP, et al. Combined thoracic ultrasound assessment during a successful weaning trial predicts postextubation distress[J]. Anesthesiology, 2017, 127(4):666-674. DOI: 10.1097/ALN.0000000000001773.

[62]LvY, HuangZ. Account of deep learning-based ultrasonic image feature in the diagnosis of severe sepsis complicated with acute kidney injury[J]. Comput Math Methods Med, 2022, 2022:8158634. DOI: 10.1155/2022/8158634.

[63]YingF, ChenS, PanG, et al. Artificial intelligence pulse coupled neural network algorithm in the diagnosis and treatment of severe sepsis complicated with acute kidney injury under ultrasound image[J]. J Healthc Eng, 2021, 2021:6761364. DOI: 10.1155/2021/6761364.

[64]StrodthoffN, StrodthoffC, BecherT, et al. Inferring respiratory and circulatory parameters from electrical impedance tomography with deep recurrent models[J]. IEEE J Biomed Health Inform, 2021, 25(8):3105-3111. DOI: 10.1109/JBHI.2021.3059016.

[65]BosLD, SchoutenLR, van VughtLA, et al. Identification and validation of distinct biological phenotypes in patients with acute respiratory distress syndrome by cluster analysis[J]. Thorax, 2017, 72(10):876-883. DOI: 10.1136/thoraxjnl-2016-209719.

[66]ChaudharyK, VaidA, DuffyÁ, et al. Utilization of deep learning for subphenotype identification in sepsis-associated acute kidney injury[J]. Clin J Am Soc Nephrol, 2020, 15(11):1557-1565. DOI: 10.2215/CJN.09330819.

[67]BhavaniSV, CareyKA, GilbertER, et al. Identifying novel sepsis subphenotypes using temperature trajectories[J]. Am J Respir Crit Care Med, 2019, 200(3):327-335. DOI: 10.1164/rccm.201806-1197OC.

[68]SinhaP, ChurpekMM, CalfeeCS. Machine learning classifier models can identify acute respiratory distress syndrome phenotypes using readily available clinical data[J]. Am J Respir Crit Care Med, 2020, 202(7):996-1004. DOI: 10.1164/rccm.202002-0347OC.

[69]KudoD, GotoT, UchimidoR, et al. Coagulation phenotypes in sepsis and effects of recombinant human thrombomodulin: an analysis of three multicentre observational studies[J]. Crit Care, 2021, 25(1):114. DOI: 10.1186/s13054-021-03541-5.

[70]SeymourCW, KennedyJN, WangS, et al. Derivation, validation, and potential treatment implications of novel clinical phenotypes for Sepsis[J]. JAMA, 2019, 321(20):2003-2017. DOI: 10.1001/jama.2019.5791.

[71]WiersemaR, JukarainenS, VaaraST, et al. Two subphenotypes of septic acute kidney injury are associated with different 90-day mortality and renal recovery[J]. Crit Care, 2020, 24(1):150. DOI: 10.1186/s13054-020-02866-x.

[72]SciclunaBP, van VughtLA, ZwindermanAH, et al. Classification of patients with sepsis according to blood genomic endotype: a prospective cohort study[J]. Lancet Respir Med, 2017, 5(10):816-826. DOI: 10.1016/S2213-2600(17)30294-1.

[73]CalfeeCS, DelucchiK, ParsonsPE, et al. Subphenotypes in acute respiratory distress syndrome: latent class analysis of data from two randomised controlled trials[J]. Lancet Respir Med, 2014, 2(8):611-620. DOI: 10.1016/S2213-2600(14)70097-9.

[74]FamousKR, DelucchiK, WareLB, et al. Acute respiratory distress syndrome subphenotypes respond differently to randomized fluid management strategy[J]. Am J Respir Crit Care Med, 2017, 195(3):331-338. DOI: 10.1164/rccm.201603-0645OC.

[75]CalfeeCS, DelucchiKL, SinhaP, et al. Acute respiratory distress syndrome subphenotypes and differential response to simvastatin: secondary analysis of a randomised controlled trial[J]. Lancet Respir Med, 2018, 6(9):691-698. DOI: 10.1016/S2213-2600(18)30177-2.

[76]ZhangZ, HoKM, HongY. Machine learning for the prediction of volume responsiveness in patients with oliguric acute kidney injury in critical care[J]. Crit Care, 2019, 23(1):112. DOI: 10.1186/s13054-019-2411-z.

[77]ParrecoJ, HidalgoA, ParksJJ, et al. Using artificial intelligence to predict prolonged mechanical ventilation and tracheostomy placement[J]. J Surg Res, 2018, 228:179-187. DOI: 10.1016/j.jss.2018.03.028.

[78]FabregatA, MagretM, FerréJA, et al. A machine learning decision-making tool for extubation in intensive care unit patients[J]. Comput Methods Programs Biomed, 2021, 200:105869. DOI: 10.1016/j.cmpb.2020.105869.

[79]HurS, MinJY, YooJ, et al. Development and validation of unplanned extubation prediction models using intensive care unit data: retrospective, comparative, machine learning study[J]. J Med Internet Res, 2021, 23(8):e23508. DOI: 10.2196/23508.

[80]ChenH, MaY, HongN, et al. Early warning of citric acid overdose and timely adjustment of regional citrate anticoagulation based on machine learning methods[J]. BMC Med Inform Decis Mak, 2021, 21(Suppl 2):126. DOI: 10.1186/s12911-021-01489-8.

[81]SuL, LiuC, LiD, et al. Toward optimal heparin dosing by comparing multiple machine learning methods: retrospective study[J]. JMIR Med Inform, 2020, 8(6):e17648. DOI: 10.2196/17648.

[82]LiD, GaoJ, HongN, et al. A clinical prediction model to predict heparin treatment outcomes and provide dosage recommendations: development and validation study[J]. J Med Internet Res, 2021, 23(5):e27118. DOI: 10.2196/27118.

[83]MavigliaR, MichiT, PassaroD, et al. Machine learning and antibiotic management[J]. Antibiotics (Basel), 2022, 11(3)DOI: 10.3390/antibiotics11030304.

[84]WangY, ZhangH, FanY, et al. Propofol anesthesia depth monitoring based on self-attention and residual structure convolutional neural network[J]. Comput Math Methods Med, 2022, 2022:8501948. DOI: 10.1155/2022/8501948.

[85]van de SandeD, van GenderenME, HuiskensJ,et al. Moving from bytes to bedside: a systematic review on the use of artificial intelligence in the intensive care unit[J]. Intensive Care Med, 2021, 47(7):750-760. DOI: 10.1007/s00134-021-06446-7.

[86]BarchittaM, MaugeriA, FavaraG, et al. Early prediction of seven-day mortality in intensive care unit using a machine learning model: results from the SPIN-UTI project[J]. J Clin Med, 2021, 10(5)DOI: 10.3390/jcm10050992.

[87]LiK, ShiQ, LiuS, et al. Predicting in-hospital mortality in ICU patients with sepsis using gradient boosting decision tree[J]. Medicine (Baltimore), 2021, 100(19):e25813. DOI: 10.1097/MD.0000000000025813.

[88]García-GalloJE, Fonseca-RuizNJ, CeliLA, et al. A machine learning-based model for 1-year mortality prediction in patients admitted to an intensive care unit with a diagnosis of sepsis[J]. Med Intensiva (Engl Ed), 2020, 44(3):160-170. DOI: 10.1016/j.medin.2018.07.016.

[89]NieX, CaiY, LiuJ, et al. Mortality prediction in cerebral hemorrhage patients using machine learning algorithms in intensive care units[J]. Front Neurol, 2020, 11:610531. DOI: 10.3389/fneur.2020.610531.

[90]HalonenKI, LeppäniemiAK, LundinJE, et al. Predicting fatal outcome in the early phase of severe acute pancreatitis by using novel prognostic models[J]. Pancreatology, 2003, 3(4):309-315. DOI: 10.1159/000071769.

[91]DingN, GuoC, LiC, et al. Anartificial neural networks model for early predicting in-hospital mortality in acute pancreatitis in MIMIC-Ⅲ[J]. Biomed Res Int, 2021, 2021:6638919. DOI: 10.1155/2021/6638919.

[92]WeissCH. Why do we fail to deliver evidence-based practice in critical care medicine?[J]. Curr Opin Crit Care, 2017, 23(5):400-405. DOI: 10.1097/mcc.0000000000000436.

[93]RosaRG, TeixeiraC, SjodingM. Novel approaches to facilitate the implementation of guidelines in the ICU[J]. J Crit Care, 2020, 60:1-5. DOI: 10.1016/j.jcrc.2020.07.014.

[94]GiulianoKK, LecardoM, StaulL. Impact of protocol watch on compliance with the surviving sepsis campaign[J]. Am J Crit Care, 2011, 20(4):313-321. DOI: 10.4037/ajcc2011421.

[95]LiuVX, MorehouseJW, MarelichGP, et al. Multicenter implementation of a treatment bundle for patients with sepsis and intermediate lactate values[J]. Am J Respir Crit Care Med, 2016, 193(11):1264-1270. DOI: 10.1164/rccm.201507-1489OC.

[96]EslamiS, Abu-HannaA, SchultzMJ, et al. Evaluation of consulting and critiquing decision support systems: effect on adherence to a lower tidal volume mechanical ventilation strategy[J]. J Crit Care, 2012, 27(4):425.e1-e8. DOI: 10.1016/j.jcrc.2011.07.082.

[97]TrogrlićZ, van der JagtM, LingsmaH, et al. Improved guideline adherence and reduced brain dysfunction after a multicenter multifaceted implementation of ICU delirium guidelines in 3, 930 patients[J]. Crit Care Med, 2019, 47(3):419-427. DOI: 10.1097/CCM.0000000000003596.

[98]BourdeauxC, GhoshE, AtallahL, et al. Impact of a computerized decision support tool deployed in two intensive care units on acute kidney injury progression and guideline compliance: a prospective observational study[J]. Crit Care, 2020, 24(1):656. DOI: 10.1186/s13054-020-03343-1.

[99]American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference: definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis[J]. Crit Care Med, 1992, 20(6):864-874.

[100]LevyMM, FinkMP, MarshallJC,et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference[J]. Crit Care Med, 2003, 31(4):1250-1256. DOI: 10.1097/01.Ccm.0000050454.01978.3b.

[101]SingerM, DeutschmanCS, SeymourCW, et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3)[J]. JAMA, 2016, 315(8):801-810. DOI: 10.1001/jama.2016.0287.

[102]朱志勇, 陈一昕, 李建功. 大数据技术在医疗急重症领域的应用[J]. 邮电设计技术, 2016, (8):28-32. DOI: 10.16463/j.cnki.issn1007-3043.2016.08.006.

[103]刘再毅, 石镇维, 梁长虹. 推进联邦学习技术在医学影像人工智能中的应用 [J]. 中华医学杂志, 2022, 102(5): 318-320. DOI: 10.3760/cma.j.cn112137-20210619-01389.

[104]RiekeN, HancoxJ, LiW, et al. The future of digital health with federated learning[J]. NPJ Digit Med, 2020, 3:119. DOI: 10.1038/s41746-020-00323-1.

[105]MahmoudiE, KamdarN, KimN, et al. Use of electronic medical records in development and validation of risk prediction models of hospital readmission: systematic review[J]. BMJ, 2020, 369:m958. DOI: 10.1136/bmj.m958.

[106]LiangH, TsuiBY, NiH, et al. Evaluation and accurate diagnoses of pediatric diseases using artificial intelligence[J]. Nat Med, 2019, 25(3):433-438. DOI: 10.1038/s41591-018-0335-9.

[107]RusinCG, AcostaSI, VuEL, et al. Automated prediction of cardiorespiratory deterioration in patients with single ventricle[J]. J Am Coll Cardiol, 2021, 77(25):3184-3192. DOI: 10.1016/j.jacc.2021.04.072.

[108]NarulaG, HaeberlinM, BalsigerJ, et al. Detection of EEG burst-suppression in neurocritical care patients using an unsupervised machine learning algorithm[J]. Clin Neurophysiol, 2021, 132(10):2485-2492. DOI: 10.1016/j.clinph.2021.07.018.

[109]MaddaliMV, ChurpekM, PhamT, et al. Validation and utility of ARDS subphenotypes identified by machine-learning models using clinical data: an observational, multicohort, retrospective analysis[J]. Lancet Respir Med, 2022, 10(4):367-377. DOI: 10.1016/S2213-2600(21)00461-6.

[110]NiaziM, ParwaniAV, GurcanMN. Digital pathology and artificial intelligence[J]. Lancet Oncol, 2019, 20(5):e253-e261. DOI: 10.1016/S1470-2045(19)30154-8.

[111]MassionPP, AnticS, AtherS, et al. Assessing the accuracy of a deep learning method to risk stratify indeterminate pulmonary nodules[J]. Am J Respir Crit Care Med, 2020, 202(2):241-249. DOI: 10.1164/rccm.201903-0505OC.

[112]ChengN, RenY, ZhouJ, et al. Deep learning-based classification of hepatocellular nodular lesions on whole-slide histopathologic images[J]. Gastroenterology, 2022, 162(7):1948-1961.e7. DOI: 10.1053/j.gastro.2022.02.025.

[113]WalshS, HumphriesSM, WellsAU, et al. Imaging research in fibrotic lung disease; applying deep learning to unsolved problems[J]. Lancet Respir Med, 2020, 8(11):1144-1153. DOI: 10.1016/S2213-2600(20)30003-5.

[114]WalshS, CalandrielloL, SilvaM, et al. Deep learning for classifying fibrotic lung disease on high-resolution computed tomography: a case-cohort study[J]. Lancet Respir Med, 2018, 6(11):837-845. DOI: 10.1016/S2213-2600(18)30286-8.

[115]WinslowCJ, EdelsonDP, ChurpekMM, et al. The impact of a machine learning early warning score on hospital mortality: a multicenter clinical intervention trial[J]. Crit Care Med, 2022, 50(9):1339-1347. DOI: 10.1097/CCM.0000000000005492.

[116]YuanS, SunY, XiaoX, et al. Using machine learning algorithms to predict candidaemia in ICU patients with new-onset systemic inflammatory response syndrome[J]. Front Med (Lausanne), 2021, 8:720926. DOI: 10.3389/fmed.2021.720926.

[117]DastiderAG, SadikF, FattahSA. An integrated autoencoder-based hybrid CNN-LSTM model for COVID-19 severity prediction from lung ultrasound[J]. Comput Biol Med, 2021, 132:104296. DOI: 10.1016/j.compbiomed.2021.104296.

[118]TiwariP, ColbornKL, SmithDE, et al. Assessment of a machine learning model applied to harmonized electronic health record data for the prediction of incident atrial fibrillation[J]. JAMA Network Open, 2020, 3(1): e1919396.

[119]PappL, SpielvogelCP, GrubmüllerB, et al. Supervised machine learning enables non-invasive lesion characterization in primary prostate cancer with [(68)Ga]Ga-PSMA-11 PET/MRI[J]. Eur J Nucl Med Mol Imaging, 2021, 48(6):1795-1805. DOI: 10.1007/s00259-020-05140-y.

[120]LiangW, YaoJ, ChenA, et al. Early triage of critically ill COVID-19 patients using deep learning[J]. Nat Commun, 2020, 11(1):3543. DOI: 10.1038/s41467-020-17280-8.

[121]SuL, XuZ, ChangF, et al. Early prediction of mortality, severity, and length of stay in the intensive care unit of sepsis patients based on sepsis 3.0 by machine learning models[J]. Front Med (Lausanne), 2021, 8:664966. DOI: 10.3389/fmed.2021.664966.

[122]SuL, ZhangZ, ZhengF, et al. Five novel clinical phenotypes for critically ill patients with mechanical ventilation in intensive care units: a retrospective and multi database study[J]. Respir Res, 2020, 21(1):325. DOI: 10.1186/s12931-020-01588-6.

[123]LiuS, SuL, LiuX, et al. Recognizing blood pressure patterns in sedated critically ill patients on mechanical ventilation by spectral clustering[J]. Ann Transl Med, 2021, 9(18):1404. DOI: 10.21037/atm-21-2806.

[124]LandiI, GlicksbergBS, LeeHC, et al. Deep representation learning of electronic health records to unlock patient stratification at scale[J]. NPJ Digit Med, 2020, 3:96. DOI: 10.1038/s41746-020-0301-z.

[125]HyunS, KaewpragP, CooperC, et al. Exploration of critical care data by using unsupervised machine learning[J]. Comput Methods Programs Biomed, 2020, 194:105507. DOI: 10.1016/j.cmpb.2020.105507.

[126]YangH, ShanC, BouwmanA, et al. Medical instrument segmentation in 3D US by hybrid constrained semi-supervised learning[J]. IEEE J Biomed Health Inform, 2022, 26(2):762-773. DOI: 10.1109/JBHI.2021.3101872.

[127]HellerN, IsenseeF, Maier-HeinKH, et al. The state of the art in kidney and kidney tumor segmentation in contrast-enhanced CT imaging: results of the KiTS19 challenge[J]. Med Image Anal, 2021, 67:101821. DOI: 10.1016/j.media.2020.101821.

[128]GrecoM, CarusoPF, CecconiM. Artificial intelligence in the intensive care unit[J]. Semin Respir Crit Care Med, 2021, 42(1):2-9. DOI: 10.1055/s-0040-1719037.

[129]RichensJG, LeeCM, JohriS. Improving the accuracy of medical diagnosis with causal machine learning[J]. Nat Commun, 2020, 11(1):3923. DOI: 10.1038/s41467-020-17419-7.

[130]WeiT, FengF, ChenJ, et al. Model-agnostic counterfactual reasoning for eliminating popularity bias in recommender system[EB/OL].[2022-02-01]. https://arxiv.org/abs/2010.15363.

[131]GoudetO, KalainathanD, CaillouP, et al. Learning functional causal models with generative neural networks[EB/OL].[2022-02-01]. https://arxiv.org/abs/1709. 05321.

[132]AtheyS, TibshiraniJ, WagerS. Generalized random forests[J]. Ann Stat, 2019, 47(2):1148-1178. DOI: 10.1214/18-AOS1709.

[133]TanJ, XuS, GeY, et al. Counterfactual explainable recommendation[EB/OL].[2022-02-01]. https://arxiv.org/abs/2108.10539.

[134]TharwatA. Classification assessment methods[J]. Appl Comput Inform, 2021, 17(1):168-192. DOI: 10.1016/j.aci.2018.08.003.

[135]JungY, HuJ. A K-fold averaging cross-validation procedure[J]. J Nonparametr Stat, 2015, 27(2):167-179. DOI: 10.1080/10485252.2015.1010532.

[136]ZhangZ, ChenL, XuP, et al. Predictive analytics with ensemble modeling in laparoscopic surgery: a technical note[J]. Laparosc Endosc Robot Surg, 2022, 5(1):25-34. DOI: 10.1016/j.lers.2021.12.003.

[137]ZhangZ, BeckMW, WinklerDA, et al. Opening the black box of neural networks: methods for interpreting neural network models in clinical applications[J]. Ann Transl Med, 2018, 6(11):216. DOI: 10.21037/atm.2018.05.32.

[138]ShimabukuroDW, BartonCW, FeldmanMD, et al. Effect of a machine learning-based severe sepsis prediction algorithm on patient survival and hospital length of stay: a randomised clinical trial[J]. BMJ Open Respir Res, 2017, 4(1):e000234. DOI: 10.1136/bmjresp-2017-000234.

[139]SemlerMW, WeavindL, HooperMH, et al. An electronic tool for the evaluation and treatment of sepsis in the ICU: a randomized controlled trial[J]. Crit Care Med, 2015, 43(8):1595-1602. DOI: 10.1097/CCM.0000000000001020.

[140]WilsonFP, MartinM, YamamotoY, et al. Electronic health record alerts for acute kidney injury: multicenter, randomized clinical trial[J]. BMJ, 2021, 372:m4786. DOI: 10.1136/bmj.m4786.

[141]SelbyNM, CasulaA, LammingL, et al. An organizational-level program of intervention for AKI: a pragmatic stepped wedge cluster randomized trial[J]. J Am Soc Nephrol, 2019, 30(3):505-515. DOI: 10.1681/ASN.2018090886.

[142]WuY, ChenY, LiS, et al. Value of electronic alerts for acute kidney injury in high-risk wards: a pilot randomized controlled trial[J]. Int Urol Nephrol, 2018, 50(8):1483-1488. DOI: 10.1007/s11255-018-1836-7.

[143]WilsonFP, ShashatyM, TestaniJ, et al. Automated, electronic alerts for acute kidney injury: a single-blind, parallel-group, randomised controlled trial[J]. Lancet, 2015, 385(9981):1966-1974. DOI: 10.1016/S0140-6736(15)60266-5.

[144]EscobarGJ, LiuVX, KipnisP. Automated identification of adults at risk for in-hospital clinical deterioration. reply[J]. N Engl J Med, 2021, 384(5):486. DOI: 10.1056/NEJMc2034836.

[145]齐霜, 毛智, 胡新, 等: 基于专科信息系统建立的重症医学数据库:大型三甲医院重症医学数据库的模式[J]. 中华危重病急救医学2020, 32(6):743-749

[146]张素珍, 唐素娟, 戎珊, 等. 基于机器学习的重症监护病房脓毒性休克患者早期发生急性肾损伤风险的预测模型构建[J]. 中华危重病急救医学, 2022, 34(3):255-259. DOI: 10.3760/cma.j.cn121430-20211126-01790.

[147]FlechetM, FaliniS, BonettiC, et al. Machine learning versus physicians′ prediction of acute kidney injury in critically ill adults: a prospective evaluation of the AKI predictor[J]. Crit Care, 2019, 23(1):282. DOI: 10.1186/s13054-019-2563-x.

[148]SjodingMW, HoferTP, CoI, et al. Interobserver reliability of the Berlin ARDS definition and strategies to improve the reliability of ARDS diagnosis[J]. Chest, 2018, 153(2):361-367. DOI: 10.1016/j.chest.2017.11.037.

[149]FleurenLM, ThoralP, ShillanDet al. Machine learning in intensive care medicine: ready for take-off?[J]. Intensive Care Med, 2020, 46(7):1486-1488. DOI: 10.1007/s00134-020-06045-y.

[150]KomorowskiM, CeliLA. Will artificial intelligence contribute to overuse in healthcare?[J]. Crit Care Med, 2017, 45(5):912-913. DOI: 10.1097/ccm.0000000000002351.