Artificial Intelligence in Acute Kidney Injury: From Static to Dynamic Models
2021; Elsevier BV; Volume: 28; Issue: 1 Linguagem: Inglês
10.1053/j.ackd.2021.03.002
ISSN1548-5609
Autores Tópico(s)Sepsis Diagnosis and Treatment
ResumoArtificial intelligence (AI) is the development of computer systems that normally require human intelligence. In the field of acute kidney injury (AKI) AI has led to an evolution of risk prediction models. In the past, static prediction models were developed using baseline (eg, preoperative) data to evaluate AKI risk. Newer models which incorporated baseline as well as evolving data collected during a hospital admission have shown improved predicative abilities. In this review, we will summarize the advances made in AKI risk prediction over the last several years, including a shift toward more dynamic, real-time, electronic medical record-based models. In addition, we will be discussing the role of electronic AKI alerts and decision support tools. Recent studies have demonstrated improved patient outcomes through the use of these tools which monitor for nephrotoxin medication exposures as well as provide kidney focused care bundles for patients at high risk for severe AKI. Finally, we will briefly discuss the pitfalls and implications of implementing these scores, alerts, and support tools. Artificial intelligence (AI) is the development of computer systems that normally require human intelligence. In the field of acute kidney injury (AKI) AI has led to an evolution of risk prediction models. In the past, static prediction models were developed using baseline (eg, preoperative) data to evaluate AKI risk. Newer models which incorporated baseline as well as evolving data collected during a hospital admission have shown improved predicative abilities. In this review, we will summarize the advances made in AKI risk prediction over the last several years, including a shift toward more dynamic, real-time, electronic medical record-based models. In addition, we will be discussing the role of electronic AKI alerts and decision support tools. Recent studies have demonstrated improved patient outcomes through the use of these tools which monitor for nephrotoxin medication exposures as well as provide kidney focused care bundles for patients at high risk for severe AKI. Finally, we will briefly discuss the pitfalls and implications of implementing these scores, alerts, and support tools. Clinical Summary•Dynamic risk models that incorporate real-time variables (eg, intraoperative and postoperative factors) have improved accuracy compared with static/fix models.•Advanced artificial intelligence techniques (eg, machine-learning and neural networks) have helped develop highlight accurate acute kidney injury (AKI) risk prediction models; however, there is no data on their ability to improve outcomes after clinical implementation.•The Nephrotoxic Injury Negated by Just in time Action (NINJA) and its associated follow-up studies have demonstrated a clear reduction in pediatric AKI after the implementation of a clinical decision support tool around the administration of nephrotoxic medications.•Several clinical trials using serum creatinine and risk score–based trigger electronic alerts and kidney focused care bundles have demonstrated improved patient outcomes with earlier action in the setting of AKI and AKI risk. •Dynamic risk models that incorporate real-time variables (eg, intraoperative and postoperative factors) have improved accuracy compared with static/fix models.•Advanced artificial intelligence techniques (eg, machine-learning and neural networks) have helped develop highlight accurate acute kidney injury (AKI) risk prediction models; however, there is no data on their ability to improve outcomes after clinical implementation.•The Nephrotoxic Injury Negated by Just in time Action (NINJA) and its associated follow-up studies have demonstrated a clear reduction in pediatric AKI after the implementation of a clinical decision support tool around the administration of nephrotoxic medications.•Several clinical trials using serum creatinine and risk score–based trigger electronic alerts and kidney focused care bundles have demonstrated improved patient outcomes with earlier action in the setting of AKI and AKI risk. Artificial intelligence (AI), a term first coined by John McCarthy in 1956, was initially defined as "the science and engineering of making intelligent machines".1Hamet P. Tremblay J. Artificial intelligence in medicine.Metabolism. 2017; 69S: S36-S40Abstract Full Text Full Text PDF PubMed Scopus (647) Google Scholar The field of medicine, which has been dependent on advancing technologies, has unsurprisingly adapted the use of AI in various clinical aspects including but not limited to drug development, health monitoring, medical data management, and disease diagnosis.2Amisha, Malik P. Pathania M. Rathaur V.K. Overview of artificial intelligence in medicine.J Fam Med Prim Care. 2019; 8: 2328-2331Crossref PubMed Google Scholar AI in medicine is often split into 2 categories: physical and virtual.1Hamet P. Tremblay J. Artificial intelligence in medicine.Metabolism. 2017; 69S: S36-S40Abstract Full Text Full Text PDF PubMed Scopus (647) Google Scholar Physical AI refers to instruments that directly assist in patient care with examples including robot-assisted surgeries and advanced neural prostheses to aid physically disabled patients. Virtual AI, or machine learning, uses computer algorithms to analyze large quantities of data, or "big data", to discover patterns that aid in medical decision-making.3Deo R.C. Machine learning in medicine.Circulation. 2015; 132: 1920-1930Crossref PubMed Scopus (1251) Google Scholar,4Shameer K. Johnson K.W. Glicksberg B.S. Dudley J.T. Sengupta P.P. Machine learning in cardiovascular medicine: are we there yet?.Heart. 2018; 104: 1156-1164Crossref PubMed Scopus (233) Google Scholar Across the medical spectrum, the processing of big data contained within the electronic health records has led to the creation of risk prediction models that have anticipated outcomes ranging from postoperative inpatient mortality to the development of myocardial infarctions to the need for ICU transfer.5Hodgson L.E. Selby N. Huang T.M. Forni L.G. The role of risk prediction models in prevention and management of AKI.Semin Nephrol. 2019; 39: 421-430Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar,6Churpek M.M. Yuen T.C. Winslow C. et al.Multicenter development and validation of a risk stratification tool for ward patients.Am J Respir Crit Care Med. 2014; 190: 649-655Crossref PubMed Scopus (149) Google Scholar At the Acute Disease Quality Initiative consensus conference in 2015, acute kidney injury (AKI) was recognized as an ideal disease state to apply machine learning and big data.7Bagshaw S.M. Goldstein S.L. Ronco C. Kellum J.A. Group A.C. Acute kidney injury in the era of big data: the 15(th) consensus conference of the acute dialysis quality initiative (ADQI).Can J Kidney Health Dis. 2016; 3 (eCollection 2016): 5https://doi.org/10.1186/s40697-016-0103-zCrossref PubMed Scopus (31) Google Scholar This is because AKI not only affects a large portion of hospitalized patients but is associated with high morbidity and mortality making disease prediction extremely valuable to patient outcomes.5Hodgson L.E. Selby N. Huang T.M. Forni L.G. The role of risk prediction models in prevention and management of AKI.Semin Nephrol. 2019; 39: 421-430Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar,8Hoste E.A. Bagshaw S.M. Bellomo R. et al.Epidemiology of acute kidney injury in critically ill patients: the multinational AKI-EPI study.Intensive Care Med. 2015; 41: 1411-1423Crossref PubMed Scopus (1343) Google Scholar In addition, with serial monitoring of renal function, there exists predisease data sets to allow for prediction and temporal detection of AKI.9Sutherland S.M. Chawla L.S. Kane-Gill S.L. et al.Utilizing electronic health records to predict acute kidney injury risk and outcomes: workgroup statements from the 15(th) ADQI Consensus Conference.Can J Kidney Health Dis. 2016; 3 (eCollection 2016): 11https://doi.org/10.1186/s40697-016-0099-4Crossref PubMed Scopus (66) Google Scholar Finally, given that standard definitions to diagnose AKI have been widely recognized through various criteria including RIFLE, AKIN, and KDIGO, AKI prediction models can be universally applied.10Thomas M.E. Blaine C. Dawnay A. et al.The definition of acute kidney injury and its use in practice.Kidney Int. 2015; 87: 62-73Abstract Full Text Full Text PDF PubMed Scopus (419) Google Scholar,11Group K.A.W. KDIGO clinical practice guideline for acute kidney injury.Kidney Int Suppl. 2012; 2(1): 1-138Google Scholar Our current markers for AKI including serum creatinine (SCr) and urine output are indicators of kidney function but not of kidney damage causing potential delays in detection. In theory, if AI can predict AKI early enough, preventative interventions can be applied in a timely fashion to avoid severe disease progression. Numerous traditional risk assessment models have been created to predict AKI in various patient populations including those undergoing general surgery, those admitted to intensive care units, and those receiving iodinated contrast.12Kheterpal S. Tremper K.K. Heung M. et al.Development and validation of an acute kidney injury risk index for patients undergoing general surgery: results from a national data set.Anesthesiology. 2009; 110: 505-515Crossref PubMed Scopus (406) Google Scholar, 13Mehta R.L. Pascual M.T. Gruta C.G. Zhuang S. Chertow G.M. Refining predictive models in critically ill patients with acute renal failure.J Am Soc Nephrol. 2002; 13: 1350-1357Crossref PubMed Scopus (294) Google Scholar, 14Silver S.A. Shah P.M. Chertow G.M. Harel S. Wald R. Harel Z. Risk prediction models for contrast induced nephropathy: systematic review.BMJ. 2015; 351: h4395Crossref PubMed Scopus (124) Google Scholar This is not meant to diminish the importance of traditional risk scores as when well-constructed and externally validated they may be of use. Recently Bell and colleagues have published a much simpler 4-variable model that was developed from population level data (including outpatient data) in 273,450 patients from the Tayside region of Scotland and then externally validated in data from Kent, UK, and Alberta, Canada.15Bell S. James M.T. Farmer C.K.T. Tan Z. de Souza N. Witham M.D. Development and external validation of an acute kidney injury risk score for use in the general population.Clin Kidney J. 2020; 13: 402-412Crossref PubMed Google Scholar Their unique 4-variable model (age, baseline GFR, presence of diabetes mellitus, and presence of congestive heart failure [Table 1]) provided a C-statistic (95% CI) of 0.80 (0.80-0.81) for the development of KDIGO serum creatinine–based AKI. The score performed slightly worse in the Kent cohort (n = 219,091), 0.71 (0.70-0.72), but a little better in the larger Canadian cohort (n = 1,173,607), 0.76 (0.75-0.76).15Bell S. James M.T. Farmer C.K.T. Tan Z. de Souza N. Witham M.D. Development and external validation of an acute kidney injury risk score for use in the general population.Clin Kidney J. 2020; 13: 402-412Crossref PubMed Google Scholar This simple score, which combined outpatient data with demographics and comorbidities, remains one of the few models to be externally validated in a large-scale cohort.Table 1Bell and Colleagues—AKI Risk Score15Bell S. James M.T. Farmer C.K.T. Tan Z. de Souza N. Witham M.D. Development and external validation of an acute kidney injury risk score for use in the general population.Clin Kidney J. 2020; 13: 402-412Crossref PubMed Google ScholarVariablePointAge 20-290 30-492 50-594 60-696 70-798 >809eGFR >60 mL/min0 45-59 mL/min5 30-44 mL/min8 < 30 mL/min16Diabetes mellitus3Heart failure4Abbreviations: AKI, acute kidney injury; eGFR, estimated glomerular filtration rate.A score of ≤4 provides a positive predictive value (PPV) of 3.6% and negative predictive value (NPV) of 99.5% for any AKI. While ≤14 provides a PPV of 15.7% and NPV of 98.7% for any AKI.Data from Bell et al.15Bell S. James M.T. Farmer C.K.T. Tan Z. de Souza N. Witham M.D. Development and external validation of an acute kidney injury risk score for use in the general population.Clin Kidney J. 2020; 13: 402-412Crossref PubMed Google Scholar Open table in a new tab Abbreviations: AKI, acute kidney injury; eGFR, estimated glomerular filtration rate. A score of ≤4 provides a positive predictive value (PPV) of 3.6% and negative predictive value (NPV) of 99.5% for any AKI. While ≤14 provides a PPV of 15.7% and NPV of 98.7% for any AKI. Data from Bell et al.15Bell S. James M.T. Farmer C.K.T. Tan Z. de Souza N. Witham M.D. Development and external validation of an acute kidney injury risk score for use in the general population.Clin Kidney J. 2020; 13: 402-412Crossref PubMed Google Scholar Cardiac surgery–associated AKI has been a focus for traditional risk score studies due to high rates of morbidity and mortality.16Mangano C.M. Diamondstone L.S. Ramsay J.G. Aggarwal A. Herskowitz A. Mangano D.T. 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 Scopus (929) Google Scholar All seven of the cardiac surgery–associated AKI risk prediction models reviewed by Huen and colleagues used variables determined in the preoperative setting with prior renal insufficiency, prior heart surgery, age, and diagnosis of diabetes being the most frequently used.17Huen S.C. Parikh C.R. Predicting acute kidney injury after cardiac surgery: a systematic review.Ann Thorac Surg. 2012; 93: 337-347Abstract Full Text Full Text PDF PubMed Scopus (155) Google Scholar Of the seven models studied, only 2 used any variables beyond the preoperative period including increased operative time, inotrope requirement, postoperative volume overload, and low cardiac output.18Aronson S. Fontes M.L. Miao Y. Mangano D.T. Group IotMSoPIR, Foundation IRaE Risk index for perioperative renal dysfunction/failure: critical dependence on pulse pressure hypertension.Circulation. 2007; 115: 733-742Crossref PubMed Scopus (180) Google Scholar,19Palomba H. de Castro I. Neto A.L. Lage S. Yu L. Acute kidney injury prediction following elective cardiac surgery: AKICS Score.Kidney Int. 2007; 72: 624-631Abstract Full Text Full Text PDF PubMed Scopus (248) Google Scholar Only four of the 7 models studied had external validation with areas under the receiver operator characteristic curve (AUC) ranging from 0.66 to 0.86. However, of these models, all 4 predicted AKI requiring dialysis that merely occurs in roughly 1% of patients undergoing cardiac surgery.20Lenihan C.R. Montez-Rath M.E. Mora Mangano C.T. Chertow G.M. Winkelmayer W.C. Trends in acute kidney injury, associated use of dialysis, and mortality after cardiac surgery, 1999 to 2008.Ann Thorac Surg. 2013; 95: 20-28Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar These initial cardiac surgery–associated AKI prediction models have significant limitations not only for their narrow definition of AKI but also for homogenous patient population, limited external validation, but perhaps most of all for their static view of risk. As patients progress through their perioperative course, their risk changes based not only on any changes in urine output or SCr but also due to their hemodynamics, administered medications, postoperative complications, as well as other interventions. As such, others have attempted to use acute risk factors to predict postoperative AKI risk knowing that intraoperative or early postoperative factors seem to impact the incidence and severity of AKI in the postoperative period.21Koyner J.L. Garg A.X. Coca S.G. et al.Biomarkers predict progression of acute kidney injury after cardiac surgery.J Am Soc Nephrol. 2012; 23: 905-914Crossref PubMed Scopus (199) Google Scholar, 22Parikh C.R. Thiessen-Philbrook H. Garg A.X. et al.Performance of kidney injury molecule-1 and liver fatty acid-binding protein and combined biomarkers of AKI after cardiac surgery.Clin J Am Soc Nephrol. 2013; 8: 1079-1088Crossref PubMed Scopus (167) Google Scholar, 23Haines R.W. Fowler A.J. Kirwan C.J. Prowle J.R. The incidence and associations of acute kidney injury in trauma patients admitted to critical care: a systematic review and meta-analysis.J Trauma Acute Care Surg. 2019; 86: 141-147Crossref PubMed Scopus (31) Google Scholar Recently, Mathis and colleagues sought to determine the links between intraoperative hypotension and AKI and preoperative risk. Analyzing 138,021 cases of major noncardiac surgery from 8 hospitals from 2008 to 2015, they defined hypotension as the lowest mean arterial pressure range achieved for more than 10 minutes with ranges being defined in absolute and relative decreases. They defined risk based on the patient's preoperative kidney function, American Society of Anesthesiology Physical Status score, anemia, and anticipated duration of anesthesia, and determined that in patients with the highest risk—mild hypotension (MAPs of 55-59) was associated with a 1.34 (1.16-1.56) increased adjusted odds of developing AKI.24Mathis M.R. Naik B.I. Freundlich R.E. et al.Preoperative risk and the association between hypotension and postoperative acute kidney injury.Anesthesiology. 2020; 132: 461-475Crossref PubMed Scopus (63) Google Scholar Dividing their cohort into derivation and validation groups, they were able to demonstrate and replicate that relative hypotension (percent decrease) was not as predictive of impending AKI compared with absolute hypotension. Regardless, their use of big data to tease out the complicated interplay of preoperative risk, intraoperative hypotension, and development of AKI is part of a larger movement in incorporate real-time/intraoperative data into assessing AKI risk.24Mathis M.R. Naik B.I. Freundlich R.E. et al.Preoperative risk and the association between hypotension and postoperative acute kidney injury.Anesthesiology. 2020; 132: 461-475Crossref PubMed Scopus (63) Google Scholar Lei and colleagues used data from 42,615 patients undergoing noncardiac surgery at 4 tertiary academic hospitals in the United States to develop several models to predict AKI.25Lei V.J. Luong T. Shan E. et al.Risk stratification for postoperative acute kidney injury in major Noncardiac surgery using preoperative and intraoperative data.JAMA Netw Open. 2019; 2: e1916921Crossref PubMed Scopus (27) Google Scholar Using logistic regression, gradient boosting machine (GBM) learning, and a random forest approach, they developed several models and assessed their performance through the addition of prehospitalization, preoperative and intraoperative variables. The inclusion of these successive time-based variables did lead to an increase in model performance. For example the GBM models predicted KDIGO AKI within the first 7 postoperative days with AUCs (95% CI) of 0.712 (0.69-0.73) for the prehospital data, 0.804 (0.79-0.82) for the inclusion of the preoperative variable and maxing out at 0.817 (0.80-0.83) for all the data.25Lei V.J. Luong T. Shan E. et al.Risk stratification for postoperative acute kidney injury in major Noncardiac surgery using preoperative and intraoperative data.JAMA Netw Open. 2019; 2: e1916921Crossref PubMed Scopus (27) Google Scholar The authors do not provide information about which features gained or loss importance in these progressive models. While the clinical significance of an increase in AUC of 0.013 between these last 2 modes remains unclear, it demonstrates that the inclusion of readily available intraoperative data (eg, intraoperative vitals, medications, blood products, etc.) improves AKI risk prediction. Separately, Bihorac and colleagues have used single-center retrospective data in concert with machine learning and advanced analytic techniques to develop perioperative models to predict AKI in the first 3 and 7 postoperative days.26Adhikari L. Ozrazgat-Baslanti T. Ruppert M. et al.Improved predictive models for acute kidney injury with IDEA: intraoperative Data Embedded Analytics.PLoS One. 2019; 14: e0214904Crossref PubMed Scopus (29) Google Scholar Using random forest classifiers, they used preoperative and intraoperative variables to predict outcomes in 2911 adults. They demonstrated that inclusion of intraoperative data improved the AUC from 0.84 to 0.86 with an improvement in accuracy from 0.76 to 0.78. In addition, including these data led to improvement of the net reclassification for patients (predominantly the false negatives) by 8% at 3 days and 7% at 7 days.26Adhikari L. Ozrazgat-Baslanti T. Ruppert M. et al.Improved predictive models for acute kidney injury with IDEA: intraoperative Data Embedded Analytics.PLoS One. 2019; 14: e0214904Crossref PubMed Scopus (29) Google Scholar Building on these aforementioned static models that only attempted to forecast AKI risk based on preoperative or risk factors at the time of ICU admission, there has been an increasing number of published models that have attempted to add additional information to their models in the hopes of improving accuracy. Malhotra and colleagues conducted a multicenter prospective cohort study to develop and validate a risk score for predicting AKI within the first 7 days of ICU admission.27Malhotra R. Kashani K.B. Macedo E. et al.A risk prediction score for acute kidney injury in the intensive care unit.Nephrol Dial Transplant. 2017; 32: 814-822Crossref PubMed Scopus (105) Google Scholar The model was developed in 573 patients from UCSD and validated in 144 patients from UCSD and 1300 patients from the Mayo Clinic. In total, 754 (37%) developed KDIGO-defined AKI. Their final model consisted of 10 independently weighted risk factors (Table 2) and provide an AUC of 0.81 (0.79-0.83) in the validation cohort (0.79 in the developmental cohort).27Malhotra R. Kashani K.B. Macedo E. et al.A risk prediction score for acute kidney injury in the intensive care unit.Nephrol Dial Transplant. 2017; 32: 814-822Crossref PubMed Scopus (105) Google Scholar Although this has not been validated by others, it serves as a new and fairly parsimonious model based on the presence of baseline and a few acute risk factors.Table 2Malhotra and Colleagues AKI Risk Prediction at ICU Admission27Malhotra R. Kashani K.B. Macedo E. et al.A risk prediction score for acute kidney injury in the intensive care unit.Nephrol Dial Transplant. 2017; 32: 814-822Crossref PubMed Scopus (105) Google ScholarRisk FactorsPointsChronicCKD2Chronic liver disease2Congestive heart failure2Hypertension2Atherosclerotic coronary disease2AcutepH ≤ 7.303Nephrotoxin exposure3Severe infection/sepsis2Mechanical ventilation2Anemia1Abbreviations: AKI, acute kidney injury; CKD, chronic kidney disease.A cutoff value of greater than 5 points provided a positive predictive value of 32% and a negative predictive value of 95%.27Malhotra R. Kashani K.B. Macedo E. et al.A risk prediction score for acute kidney injury in the intensive care unit.Nephrol Dial Transplant. 2017; 32: 814-822Crossref PubMed Scopus (105) Google ScholarReprinted with permission from Malhotra et al.27Malhotra R. Kashani K.B. Macedo E. et al.A risk prediction score for acute kidney injury in the intensive care unit.Nephrol Dial Transplant. 2017; 32: 814-822Crossref PubMed Scopus (105) Google Scholar Open table in a new tab Abbreviations: AKI, acute kidney injury; CKD, chronic kidney disease. A cutoff value of greater than 5 points provided a positive predictive value of 32% and a negative predictive value of 95%.27Malhotra R. Kashani K.B. Macedo E. et al.A risk prediction score for acute kidney injury in the intensive care unit.Nephrol Dial Transplant. 2017; 32: 814-822Crossref PubMed Scopus (105) Google Scholar Reprinted with permission from Malhotra et al.27Malhotra R. Kashani K.B. Macedo E. et al.A risk prediction score for acute kidney injury in the intensive care unit.Nephrol Dial Transplant. 2017; 32: 814-822Crossref PubMed Scopus (105) Google Scholar The use of progressive data to enhance AKI risk prediction is not isolated to surgical cases. Moving away from these more traditional models, Flechet and colleagues developed the AKI predictor, a random forest machine learning risk algorithm for critically ill adults. Using data from the Early vs Late Parenteral Nutrition in Critically Ill Adults (EPaNIC) multicenter database (n = 4490), they developed and validated an AKI risk score using clinical information available (1) before and (2) on ICU admission as well as (3) data from ICU day 1.28Flechet M. Güiza F. Schetz M. et al.AKIpredictor, an online prognostic calculator for acute kidney injury in adult critically ill patients: development, validation and comparison to serum neutrophil gelatinase-associated lipocalin.Intensive Care Med. 2017; 43: 764-773Crossref PubMed Scopus (84) Google Scholar The model was used to predict all 3 stages of AKI as well as just stage 2 or 3. Model performance for predicting any AKI outcome consistently improved with progressive data added to the model (eg, AUCs increasing from 0.77 (0.77-0.77) to 0.84 (0.83-0.84) for the prediction of stage 2 or 3 AKI. In addition, in their original validation cohort they demonstrated that their progressive model outperformed neutrophil gelatinase associated lipocalin for the development of stage 2 or 3 AKI (AUC 0.79 (0.79-0.79) compared with the aforementioned 0.84. Subsequently, these same authors further compared their machine learning algorithm's ability to predict AKI to the treating physicians' ability to predict the same outcomes.29Flechet M. Falini S. Bonetti C. et al.Machine learning versus physicians' prediction of acute kidney injury in critically ill adults: a prospective evaluation of the AKIpredictor.Crit Care. 2019; 23: 282Crossref PubMed Scopus (34) Google Scholar In 252 patients from 5 ICUs from a single tertiary academic center, the AKIpredictor did not outperform physicians in its ability to predict stage 2-3 AKI but there was no statistical difference in their performance (AUC 0.75 (0.62-0.88) vs 0.80 (0.69-0.92), P = 0.25). Physicians tended to overestimate the risk of AKI, thus allowing the AKIpredictor to have a higher net benefit compared with physicians but in total this study was limited in that only 30 (12%) patients developed severe AKI.29Flechet M. Falini S. Bonetti C. et al.Machine learning versus physicians' prediction of acute kidney injury in critically ill adults: a prospective evaluation of the AKIpredictor.Crit Care. 2019; 23: 282Crossref PubMed Scopus (34) Google Scholar In addition to these aforementioned models, there have been several models developed and published in the last 5 years that have sought to use both advanced AI techniques (eg, neural networks and machine learning) as well as harness the power of big data and the electronic medical records to predict AKI across all hospitalized patients. Recently, Tomasev and colleagues published a seminal article on an AKI risk score using data from 703,782 adult patients in the United States Department of Veterans Affairs Healthcare System (172 inpatient and 1062 outpatient locations).30Tomašev N. Glorot X. Rae J.W. et al.A clinically applicable approach to continuous prediction of future acute kidney injury.Nature. 2019; 572: 116-119Crossref PubMed Scopus (394) Google Scholar They used a recurrent neural network to develop a highly accurate model that could detect KDIGO-defined AKI. Using the full capacity of the electronic medical record including 620,000 base features (variables), they attempted to predict AKI with a lead time of 48 hours interpreting data in 6-h intervals. Their final model (33% precision) provided a sensitivity of 55.8% and specificity of 82.7%, based on a 2:1 false to true alert ratio. While this may be a higher than desired false positive rate, it is important to note that 25% of the false positives were in patients who eventually developed AKI; however, not within the first 48 hours. It seemed like prior CKD may have impacted these results. The model performed much better at predicting severe dialysis requiring AKI, providing 84.3% correct prediction of kidney replacement therapy (KRT) within the next 30 days and over 90% accuracy at 90 days. However, this study was not without its limitations. Given that it used data from the Veteran's administration, less than 7% of all subjects were female, limiting its generalizability. In addition, they used a randomly selected group of patients to serve as their test set, which has been shown to provide optimistic estimates of accuracy as opposed to external validation.31Siontis G.C. Tzoulaki I. Castaldi P.J. Ioannidis J.P. External validation of new risk prediction models is infrequent and reveals worse prognostic discrimination.J Clin Epidemiol. 2015; 68: 25-34Abstract Full Text Full Text PDF PubMed Scopus (190) Google Scholar In addition, they used deep learning and included over 300 features making the model fairly complex and it is unclear how pairing the model down with fewer features would impact discrimination and accuracy. While this model should eventually be validated in a more generalized global population, it remains the state-of-the-art model for AI detection of AKI. This is not to say that there are no other AI models that have been developed and investigated. Koyner and Churpek have published a series of 3 articles using AI techniques to develop and validate AKI risk models. The first article, published in 2016, sought to predict ward based AKI in a cohort of 202,961 patients from 5 hospitals across Chicago.32Koyner J.L. Adhikari R. Edelson D.P. Churpek M.M. Development of a multicenter ward-based AKI prediction model.Clin J Am Soc Nephrol. 2016; 11: 1935-1943Crossref PubMed Scopus (63) Google Scholar Owing to the
Referência(s)