FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4383199473> ?p ?o ?g. }

• W4383199473 endingPage "1854" @default.
• W4383199473 startingPage "1852" @default.
• W4383199473 abstract "In the 2004 movie I, Robot, the humanoid robot dives into the water in an attempt to triage casualties of a car crash, resulting in 2 passengers trapped in submerged motor vehicles. It quickly calculates the survival odds of the adult and child victims. The robot determines that the adult victim has a higher chance of survival and saves the adult, leaving the child to drown. After this scene, robots are viewed as unemotional and uncompromising while calculating who gets to live. Can machines, with their ability to rapidly incorporate multiple data sources without emotional bias, help make healthcare decisions? Is the world ready for that? Complex and multifactor problems may be ideal for artificial intelligence (AI). Acute kidney injury (AKI) after cardiac surgery is an intricate problem, and clinicians have yet to identify all the elements that induce it. Acute kidney injury after cardiac surgery occurs in approximately 1% to 16% of patients, with increased incidence in those with preexisting impaired renal function.1Zanardo G Michielon P Paccagnella A et al.Acute renal failure in the patient undergoing cardiac operation. Prevalence, mortality rate, and main risk factors.J Thorac Cardiovasc Surg. 1994; 107: 1489-1495Abstract Full Text PDF PubMed Google Scholar,2Mangos GJ Brown MA Chan WY et al.Acute renal failure following cardiac surgery: Incidence, outcomes and risk factors.Aust N Z J Med. 1995; 25: 284-289Crossref PubMed Google Scholar It can increase morbidity and mortality, length of hospital stay, and the cost of healthcare. The need for postoperative dialysis has been identified as an independent risk factor for death,3Chertow GM Levy EM Hammermeister KE et al.Independent association between acute renal failure and mortality following cardiac surgery.Am J Med. 1998; 104: 343-348Abstract Full Text Full Text PDF PubMed Scopus (1058) Google Scholar with mortality rates reaching >60% in cardiac surgery patients who develop postoperative acute renal failure requiring renal replacement therapy.4Mangano CM Diamondstone LS Ramsay JG et al.Renal dysfunction after myocardial revascularization: Risk factors, adverse outcomes, and hospital resource utilization. The Multicenter Study of Perioperative Ischemia Research Group.Ann Intern Med. 1998; 128: 194-203Crossref PubMed Google Scholar,5Thakar CV Liangos O Yared J-P et al.Predicting acute renal failure after cardiac surgery: Validation and redefinition of a risk-stratification algorithm.Hemodial Int. 2003; 7: 143-147Crossref PubMed Google Scholar Societal burden of this complication is also significant because patients needing postoperative renal replacement therapy comprise <2% of cardiac surgery patients yet use 12% of intensive care unit resources.6Mehta RH Grab JD O'Brien SM et al.Bedside tool for predicting the risk of postoperative dialysis in patients undergoing cardiac surgery.Circulation. 2006; 114: 2208-2216Crossref PubMed Scopus (439) Google Scholar Accurate prediction of AKI provides an opportunity to develop early diagnosis and treatment strategies. Identifying the ‘at-risk’ patient population may help implement preventive or therapeutic strategies. Many groups have attempted to develop a risk score for predicting postoperative AKI, but a comprehensive and universally accepted tool is still not available. The evolution of a risk stratification algorithm for postoperative renal dysfunction began 25 years ago with more than 43,000 predominantly White male patients from the Department of Veterans Affairs medical centers.7Chertow GM Lazarus JM Christiansen CL et al.Preoperative renal risk stratification.Circulation. 1997; 95: 878-884Crossref PubMed Google Scholar It started with 10 clinical variables that were related to cardiac and renal function and likely contributed to the development of renal failure after cardiac surgery. They created a preoperative risk stratification algorithm that placed patients into low-, medium-, or high-risk categories for developing renal failure based on the severity of the pertinent factors and their interaction.7Chertow GM Lazarus JM Christiansen CL et al.Preoperative renal risk stratification.Circulation. 1997; 95: 878-884Crossref PubMed Google Scholar Subsequent risk models attempted to validate and expand existing algorithms for use in increasingly more diverse populations.8Fortescue EB Bates DW Chertow GM. Predicting acute renal failure after coronary bypass surgery: Cross-validation of two risk-stratification algorithms.Kidney Int. 2000; 57: 2594-2602Abstract Full Text Full Text PDF PubMed Scopus (155) Google Scholar, 9Thakar CV Liangos O Yared JP et al.ARF after open-heart surgery: Influence of gender and race.Am J Kidney Dis. 2003; 41: 742-751Abstract Full Text Full Text PDF PubMed Google Scholar, 10Elmistekawy E McDonald B Hudson C et al.Clinical impact of mild acute kidney injury after cardiac surgery.Ann Thorac Surg. 2014; 98: 815-822Abstract Full Text Full Text PDF PubMed Google Scholar Continuing efforts focused on developing risk stratification scores that could be used readily to anticipate renal injury in clinical settings using bedside risk prediction tools. The pioneers in this effort developed the Cleveland Risk Score (CRS),11Thakar CV Arrigain S Worley S et al.A clinical score to predict acute renal failure after cardiac surgery.J Am Soc Nephrol. 2005; 16: 162-168Crossref PubMed Scopus (794) Google Scholar Simplified Predictive Index,6Mehta RH Grab JD O'Brien SM et al.Bedside tool for predicting the risk of postoperative dialysis in patients undergoing cardiac surgery.Circulation. 2006; 114: 2208-2216Crossref PubMed Scopus (439) Google Scholar and the Bedside Index.12Wijeysundera DN Karkouti K Dupuis J-Y et al.Derivation and validation of a simplified predictive index for renal replacement therapy after cardiac surgery.JAMA. 2007; 297: 1801-1809Crossref PubMed Scopus (338) Google Scholar These were soon followed by the Kidney Disease: Improving Global Outcomes-based model,13Birnie K Verheyden V Pagano D et al.Predictive models for kidney disease: Improving global outcomes (KDIGO) defined acute kidney injury in UK cardiac surgery.Crit Care. 2014; 18: 606Crossref PubMed Scopus (133) Google Scholar Ng Score,14Ng SY Sanagou M Wolfe R et al.Prediction of acute kidney injury within 30 days of cardiac surgery.J Thorac Cardiovasc Surg. 2014; 147: 1875-1883Abstract Full Text Full Text PDF PubMed Google Scholar and the CRATE score,15Jorge-Monjas P Bustamante-Munguira J Lorenzo M et al.Predicting cardiac surgery–associated acute kidney injury: The CRATE score.J Crit Care. 2016; 31: 130-138Crossref PubMed Google Scholar among others. Other studies validated previous risk score algorithms for new populations, including Indian,16Gangadharan S Sundaram KR Vasudevan S et al.Predictors of acute kidney injury in patients undergoing adult cardiac surgery.Ann Card Anaesth. 2018; 21: 448-454Crossref PubMed Scopus (8) Google Scholar Chinese,17Jiang W Xu J Shen B et al.Validation of four prediction scores for cardiac surgery-associated acute kidney injury in Chinese patients.Braz J Cardiovasc Surg. 2017; 32: 481-486PubMed Google Scholar and Southeast Asian,18Nah CW Ti LK Liu W et al.A clinical score to predict acute kidney injury after cardiac surgery in a Southeast-Asian population.Interact Cardiovasc Thorac Surg. 2016; 23: 757-761Crossref PubMed Scopus (20) Google Scholar and attempted to validate their utility in less severe forms of AKI.19Wong B St Onge J Korkola S et al.Validating a scoring tool to predict acute kidney injury (AKI) following cardiac surgery.Can J Kidney Health Dis. 2015; 2: 37Crossref PubMed Scopus (20) Google Scholar, 20Huen SC Parikh CR. Predicting acute kidney injury after cardiac surgery: A systematic review.Ann Thorac Surg. 2012; 93: 337-347Abstract Full Text Full Text PDF PubMed Scopus (169) Google Scholar, 21Englberger L Suri RM Li Z et al.Validation of clinical scores predicting severe acute kidney injury after cardiac surgery.Am J Kidney Dis. 2010; 56: 623-631Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar Ranucci et al. presented a new model for predicting cardiac surgery–associated AKI (CSA-AKI). It was developed using data from 830 patients from a single institution, based on a “static” model, and then integrated with a “dynamic” model. The study aimed to develop a dynamic predictive model—one that could modify the preoperative risk stratification by superimposing modifiable intraoperative data onto the preoperative baseline assessment. The following 3 separate risk models were developed: (1) static risk model (SRM) using preoperative factors, (2) dynamic perfusion risk (DPR) founded in the perfusion-related variables, and (3) multifactorial dynamic perfusion index (MDPI), which is the combination of SRM and DPR. The SRM was created by merging age and hematocrit data into the CRS,11Thakar CV Arrigain S Worley S et al.A clinical score to predict acute renal failure after cardiac surgery.J Am Soc Nephrol. 2005; 16: 162-168Crossref PubMed Scopus (794) Google Scholar which is based on sex, congestive heart failure, type 1 diabetes, chronic obstructive pulmonary disease, left ventricular ejection fraction, preoperative use of an intra-aortic balloon pump, history of previous cardiac surgery, type of current cardiac surgery, and preoperative creatinine. The outcome parameter, AKI, was defined as a serum creatinine increase of at least 50% of the baseline value, occurring within the first 48 hours after surgery.22Ranucci M, Di Dedda U, Cotza M, et al. The multifactorial dynamic perfusion index: A predictive tool of cardiac surgery associated acute kidney injury [e-pub ahead of print]. Perfusion. https://doi.org/10.1177/02676591221137033. Accessed May 15, 2023.Google Scholar The DPR score was created with data collected during cardiopulmonary bypass (CPB) and narrowed down to 7 CPB-related variables (nadir pump flow indexed for body surface area, nadir hematocrit maintained for at least 10 minutes, nadir arterial oxygen delivery maintained for at least 10 minutes, time of exposure to a nadir arterial oxygen delivery below the critical value, peak blood lactates, positive packed red blood cell transfusion, and CPB duration). Specific CSA-AKI risk was calculated separately for each of the 7 predictors. The values of each factor-related risk were merged into a single logistic regression that represents the DPR. The CSA-AKI risk was determined by merging the SRM and the DPR into a single logistic regression equation defining the MDPI-CSA-AKI risk.22Ranucci M, Di Dedda U, Cotza M, et al. The multifactorial dynamic perfusion index: A predictive tool of cardiac surgery associated acute kidney injury [e-pub ahead of print]. Perfusion. https://doi.org/10.1177/02676591221137033. Accessed May 15, 2023.Google Scholar The predictive accuracy of risk stratification scores is determined by the area under the receiver operating characteristic curve (ROC). When evaluating the accuracy of the AKI score in 2 randomly selected patients—one with and one without AKI—the ROC indicates the probability that the patient with AKI will have a higher score than the patient without AKI. The higher the ROC, the greater the predictive accuracy of the examined risk score algorithm.23Hanley JA McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve.Radiology. 1982; 143: 29-36Crossref PubMed Google Scholar,24Grunkemeier GL Jin R Receiver operating characteristic curve analysis of clinical risk models.Ann Thorac Surg. 2001; 72: 323-326Abstract Full Text Full Text PDF PubMed Scopus (84) Google Scholar Ranucci et al. tested the discrimination of MDPI using ROC analysis, and compared it to the other 3 existing CSA-AKI predictive models. The CRS had an area under the curve (AUC) of 0.580 (95% CI 0.545-0.614), the simplified predictive score had an AUC of 0.631 (95% CI 0.597-0.664), and the bedside tool had an AUC of 0.587 (95% CI 0.552-0.621). The presented static component (SRM model) achieved an AUC ROC of 0.696 (95% CI 0.663-0.727), the dynamic component (DPR model) showed an AUC ROC of 0.723 (95% CI 0.691-0.753), and the combined static and dynamic model (MDPI model) showed an AUC ROC of 0.769 (95% CI 0.739-0.798). The difference in AUC between the MDPI and the other scores yielded p < 0.001.22Ranucci M, Di Dedda U, Cotza M, et al. The multifactorial dynamic perfusion index: A predictive tool of cardiac surgery associated acute kidney injury [e-pub ahead of print]. Perfusion. https://doi.org/10.1177/02676591221137033. Accessed May 15, 2023.Google Scholar Hence, this dynamic MDPI model includes modifiable intraoperative factors, which, if corrected, may change the overall risk of CSA-AKI, and it demonstrates high predictive accuracy. The novelty of the approach characterized in this paper is the inclusion of static nonmodifiable data and dynamic intraoperative modifiable parameters into a single algorithm. The inclusion of the various equations from this model and the possibility for integration with existing intraoperative data monitoring systems allow the possibility of following the changes of the risk scores calculated using the DPR and the MDPI in real-time. Post-CPB dynamic factors, such as mean arterial pressure, hematocrit, or the use of inotropes and vasopressors, are not included in the MDPI model. They could be incorporated into future models to improve further completeness and inclusion of dynamic determinants of renal injury in the postoperative period. This approach may lend itself to developing the optimal predictive index; it is validated thoroughly, generalizable to all patient populations, has high predictive accuracy, is dynamically responsive to perioperative events, and is scored and interpreted easily. However, the complexity, large quantity of data input, and the need for ongoing risk score recalculation based on this data, will present challenges for real-time point-of-care use. Advances in AI and machine learning may present an answer to this logistical challenge, allowing for their application toward complex bedside scoring tools during perioperative care optimization. AI differs from traditional computer programming because computers are programmed with explicit instructions to perform calculations after incorporating large amounts of data. Machine learning, a subfield of AI, is the use and development of computer systems that are able to learn and adapt without following explicit instructions by using algorithms and statistical models to analyze and draw inferences from patterns in data.25Brown S. Machine learning, explained. Ideas Made to Matter. Available at:https://mitsloan.mit.edu/ideas-made-to-matter/machine-learning-explained. Accessed June 28, 2023.Google Scholar The application of AI into medicine has already begun. Studies have looked at the utilization of AI in the operating room, including one investigating the ability to predict events such as hypotension and sepsis.26Hashimoto DA Witkowski E Gao L et al.Artificial intelligence in anesthesiology: Current techniques, clinical applications, and limitations.Anesthesiology. 2020; 132: 379-394Crossref PubMed Scopus (163) Google Scholar A few studies are investigating the integration of machine learning to determine the risk of CSA-AKI.27Petrosyan Y Mesana TG Sun LY. Prediction of acute kidney injury risk after cardiac surgery: Using a hybrid machine learning algorithm.BMC Med Inform Decis Mak. 2022; 22: 137Crossref PubMed Scopus (3) Google Scholar, 28Ejmalian A Aghaei A Nabavi S et al.Prediction of acute kidney injury after cardiac surgery using interpretable machine learning.Anesth Pain Med. 2022; 12e127140Crossref PubMed Scopus (1) Google Scholar, 29Tseng PY Chen YT Wang CH et al.Prediction of the development of acute kidney injury following cardiac surgery by machine learning.Crit Care. 2020; 24: 478Crossref PubMed Scopus (140) Google Scholar The results indicate potential utility; however, the various machine learning programs available currently analyze the different variables dissimilarly. This leads to incongruous prediction results from the various machine-learning programs. Previous risk models were developed using statistical assumptions; for example, linear relationships and statistically significant variables. However, not every relationship is linear, and some variables may have a causal effect but not be statistically significant. Including only significant variables in the calculation will decrease the model's predictive power. The incorporation of dynamic clinical data into the predictive algorithms is novel and challenging in itself. Most of the information used to develop predictive models has been with retrospective data only. One study used machine learning and incorporated new live data into the predictive models to determine performance.30Sun H Depraetere K Meesseman L et al.Machine learning-based prediction models for different clinical risks in different hospitals: Evaluation of live performance.J Med Internet Res. 2022; 24: e34295Crossref PubMed Scopus (5) Google Scholar The prediction models performed slightly worse when using the live data than the retrospective data. Advances in the evolution of predictive models incorporating live data will be necessary to effectively integrate this input type and precisely calculate and predict outcomes. Efficient analysis of dynamic data to obtain real-time adjustment in risk scores may be precisely what is necessary for the prevention of CSA-AKI. The difficulty in initiating effective early treatments is the inability to anticipate renal dysfunction. Significant delay in diagnosis of renal injury due to the use of surrogate markers that present late after the sustained trauma may result in lower treatment success. Furthermore, evidence suggests that therapeutic interventions to prevent CSA-AKI should be performed early, possibly within 24-to-48 hours after the inciting renal injury, to be successful.31Star RA. Treatment of acute renal failure.Kidney Int. 1998; 54: 1817-1831Abstract Full Text Full Text PDF PubMed Scopus (697) Google Scholar,32Bonventre JV Weinberg JM. Recent advances in the pathophysiology of ischemic acute renal failure.J Am Soc Nephrol. 2003; 14: 2199-2210Crossref PubMed Scopus (654) Google Scholar The ability to detect events contributing to injury in real-time is optimal for patient care, thus allowing for timely therapeutic and preventive interventions. Such detection and care models are very complex and may benefit from using AI technology. Once machine learning programming has been validated, it can perhaps be applied in real-time at the bedside using live data. This research prompts discussion about the various channels to specifically use machine learning. It is conceivable that clinicians could use machine learning to effectively change outcomes. One could envision receiving a notification regarding the risk of renal failure based on ongoing analysis of dynamic parameters contributing to CSA-AKI superimposed on the static preoperative risk. Furthermore, recommendations as to which parameters may need to be adjusted or interventions to be initiated to mitigate any dynamic risk increase could be generated in real-time. This type of application of AI and machine learning could change how clinicians practice medicine and is not as improbable as you may think. This type of technology is already prevalent in the modern world, from shopping ideas on websites, movie suggestions on streaming platforms, and predictive text functions on your smartphone. Why not bedside patient care? None." @default.
• W4383199473 created "2023-07-06" @default.
• W4383199473 creator A5025134837 @default.
• W4383199473 creator A5049907709 @default.
• W4383199473 date "2023-10-01" @default.
• W4383199473 modified "2023-10-16" @default.
• W4383199473 title "I, Robot: Healthcare Decisions Made With Artificial Intelligence" @default.
• W4383199473 cites W1552974878 @default.
• W4383199473 cites W1967237457 @default.
• W4383199473 cites W1974190701 @default.
• W4383199473 cites W1986347387 @default.
• W4383199473 cites W1990622808 @default.
• W4383199473 cites W2000228088 @default.
• W4383199473 cites W2004191596 @default.
• W4383199473 cites W2009881682 @default.
• W4383199473 cites W2051681437 @default.
• W4383199473 cites W2055675692 @default.
• W4383199473 cites W2057571412 @default.
• W4383199473 cites W2071273488 @default.
• W4383199473 cites W2071630622 @default.
• W4383199473 cites W2095112808 @default.
• W4383199473 cites W2098834904 @default.
• W4383199473 cites W2101015806 @default.
• W4383199473 cites W2105133745 @default.
• W4383199473 cites W2129611051 @default.
• W4383199473 cites W2141413316 @default.
• W4383199473 cites W2157825442 @default.
• W4383199473 cites W2193019744 @default.
• W4383199473 cites W2411969659 @default.
• W4383199473 cites W2474248963 @default.
• W4383199473 cites W2897826231 @default.
• W4383199473 cites W2972320421 @default.
• W4383199473 cites W3046305373 @default.
• W4383199473 cites W4225394392 @default.
• W4383199473 cites W4280515499 @default.
• W4383199473 cites W4297494746 @default.
• W4383199473 cites W4307500733 @default.
• W4383199473 doi "https://doi.org/10.1053/j.jvca.2023.06.045" @default.
• W4383199473 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37500370" @default.
• W4383199473 hasPublicationYear "2023" @default.
• W4383199473 type Work @default.
• W4383199473 citedByCount "0" @default.
• W4383199473 crossrefType "journal-article" @default.
• W4383199473 hasAuthorship W4383199473A5025134837 @default.
• W4383199473 hasAuthorship W4383199473A5049907709 @default.
• W4383199473 hasConcept C154945302 @default.
• W4383199473 hasConcept C160735492 @default.
• W4383199473 hasConcept C162324750 @default.
• W4383199473 hasConcept C41008148 @default.
• W4383199473 hasConcept C50522688 @default.
• W4383199473 hasConcept C71924100 @default.
• W4383199473 hasConcept C90509273 @default.
• W4383199473 hasConceptScore W4383199473C154945302 @default.
• W4383199473 hasConceptScore W4383199473C160735492 @default.
• W4383199473 hasConceptScore W4383199473C162324750 @default.
• W4383199473 hasConceptScore W4383199473C41008148 @default.
• W4383199473 hasConceptScore W4383199473C50522688 @default.
• W4383199473 hasConceptScore W4383199473C71924100 @default.
• W4383199473 hasConceptScore W4383199473C90509273 @default.
• W4383199473 hasIssue "10" @default.
• W4383199473 hasLocation W43831994731 @default.
• W4383199473 hasLocation W43831994732 @default.
• W4383199473 hasOpenAccess W4383199473 @default.
• W4383199473 hasPrimaryLocation W43831994731 @default.
• W4383199473 hasRelatedWork W1506200166 @default.
• W4383199473 hasRelatedWork W1995515455 @default.
• W4383199473 hasRelatedWork W2048182022 @default.
• W4383199473 hasRelatedWork W2080531066 @default.
• W4383199473 hasRelatedWork W2604872355 @default.
• W4383199473 hasRelatedWork W2748952813 @default.
• W4383199473 hasRelatedWork W2899084033 @default.
• W4383199473 hasRelatedWork W3031052312 @default.
• W4383199473 hasRelatedWork W3032375762 @default.
• W4383199473 hasRelatedWork W3108674512 @default.
• W4383199473 hasVolume "37" @default.
• W4383199473 isParatext "false" @default.
• W4383199473 isRetracted "false" @default.
• W4383199473 workType "article" @default.