Prediction of acute kidney injury using artificial intelligence: are we there yet?

2019 ◽  
Vol 35 (2) ◽  
pp. 204-205 ◽  
Author(s):  
Wim Van Biesen ◽  
Jill Vanmassenhove ◽  
Johan Decruyenaere
Keyword(s):  
Artificial Intelligence ◽  
Acute Kidney Injury ◽  
Kidney Injury

scholarly journals Cross-site transportability of an explainable artificial intelligence model for acute kidney injury prediction

2020 ◽  
Vol 11 (1) ◽  
Author(s):  
Xing Song ◽  
Alan S. L. Yu ◽  
John A. Kellum ◽  
Lemuel R. Waitman ◽  
Michael E. Matheny ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Acute Kidney Injury ◽  
Kidney Injury ◽  
Major Barrier ◽  
Injury Prediction ◽  
The Us ◽  
Performance Deterioration ◽  
Research Gap ◽  
Clinical Adoption ◽  
Cross Site

Abstract Artificial intelligence (AI) has demonstrated promise in predicting acute kidney injury (AKI), however, clinical adoption of these models requires interpretability and transportability. Non-interoperable data across hospitals is a major barrier to model transportability. Here, we leverage the US PCORnet platform to develop an AKI prediction model and assess its transportability across six independent health systems. Our work demonstrates that cross-site performance deterioration is likely and reveals heterogeneity of risk factors across populations to be the cause. Therefore, no matter how accurate an AI model is trained at the source hospital, whether it can be adopted at target hospitals is an unanswered question. To fill the research gap, we derive a method to predict the transportability of AI models which can accelerate the adaptation process of external AI models in hospitals.

7 Artificial Intelligence and Machine Learning as a Predictive Tool for Acute Kidney Injury in Trauma and Severely Burned Patients

2020 ◽  
Vol 41 (Supplement_1) ◽  
pp. S7-S8
Author(s):  
Stephanie M Falwell ◽  
Nam K Tran ◽  
Soman Sen ◽  
Tina L Palmieri ◽  
David G Greenhalgh ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Acute Kidney Injury ◽  
Kidney Injury ◽  
Support Vector ◽  
Trauma Patients ◽  
K Nearest Neighbor ◽  
Predictive Tool ◽  
Adult Age ◽  
Generalization Accuracy

Abstract Introduction Kidney injury doubles burn mortality—thus, early prediction of acute kidney injury (AKI) in the burn population could benefit from artificial intelligence (AI) and machine learning (ML). Our objective in this study was to build and assess the theoretical performances of such AI/ML algorithms and to develop generalizable models that could augment AKI recognition. Methods Two databases containing patients that received neutrophil gelatinase associated lipocalin (NGAL), creatinine, N-terminal pro-B-type natriuretic peptide (NT-proBNP) and urine output (UOP) measurements at admission were used to train, test, and generalize the AI/ML models. Models were first optimized in Cohort A for predicting AKI in Cohort B. Cohort A (n = 50) was based on a retrospective dataset of adult (age³18 years) burn patients, while Cohort B (n = 51) consisted of prospectively enrolled adult burned or non-burned trauma patients at risk for AKI. We employed a grid search and cross validation approach in building 68,100 unique ML models from five distinct ML approaches: logistic regression (LR), k-nearest neighbor (k-NN), support vector machine (SVM), random forest (RF), and deep neural networks (DNN) which enabled us to find the most accurate ML models. Results The best generalization accuracy (86%), sensitivity (91%), and specificity (85%) with NGAL alone was noted with LR, SVM and RF models. Generalizability prediction accuracy, sensitivity and specificity were respectively highest with the optimized DNN model (92%, 100%, and 90%) and the k-NN model (92%, 91%, and 93%) when tested with Cohort B using all four biomarkers. k-NN provided best generalization accuracy (84%) without NGAL using only NT-proBNP and creatinine, followed by DNN using creatinine only with an accuracy of 82%. AI/ML algorithms using results obtained at admission accelerated the average (SD) time to AKI prediction by 61.8 (32.5) hours. Conclusions NGAL is analytically superior to traditional AKI biomarkers such as creatinine and UOP. With machine learning, the AKI predictive capability of NGAL can be further enhanced and accelerated when combined with NT-proBNP, UOP, and creatinine. Applicability of Research to Practice Without NGAL, machine learning models continue to provide robust means in accelerating the prediction of AKI using both common and biomarkers of cardiorenal dysfunction.

scholarly journals Artificial Intelligence Pulse Coupled Neural Network Algorithm in the Diagnosis and Treatment of Severe Sepsis Complicated with Acute Kidney Injury under Ultrasound Image

2021 ◽  
Vol 2021 ◽  
pp. 1-8
Author(s):  
Fu Ying ◽  
Shuhua Chen ◽  
Guojun Pan ◽  
Zemin He
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Acute Kidney Injury ◽  
Severe Sepsis ◽  
Ultrasound Image ◽  
Kidney Injury ◽  
Ultrasound Images ◽  
Image Information ◽  
Ultrasonic Image ◽  
Pulse Coupled Neural Network

The objective of this study was to explore the diagnosis of severe sepsis complicated with acute kidney injury (AKI) by ultrasonic image information based on the artificial intelligence pulse coupled neural network (PCNN) algorithm. In this study, an algorithm of ultrasonic image information enhancement based on the artificial intelligence PCNN was constructed and compared with the histogram equalization algorithm and linear transformation algorithm. After that, it was applied to the ultrasonic image diagnosis of 20 cases of severe sepsis combined with AKI in hospital. The condition of each patient was diagnosed by ultrasound image performance, change of renal resistance index (RRI), ultrasound score, and receiver operator characteristic curve (ROC) analysis. It was found that the histogram distribution of this algorithm was relatively uniform, and the information of each gray level was obviously retained and enhanced, which had the best effect in this algorithm; there was a marked individual difference in the values of RRI. Overall, the values of RRI showed a slight upward trend after admission to the intensive care unit (ICU). The RRI was taken as the dependent variable, time as the fixed-effect model, and patients as the random effect; the parameter value of time was between 0.012 and 0.015, p = 0.000 < 0.05 . Besides, there was no huge difference in the ultrasonic score among different time measurements (t = 1.348 and p = 0.128 > 0.05 ). The area under the ROC curve of the RRI for the diagnosis of AKI at the 2nd day, 4th day, and 6th day was 0.758, 0.841, and 0.856, respectively, which was all greater than 0.5 ( p < 0.05 ). In conclusion, the proposed algorithm in this study could significantly enhance the amount of information in ultrasound images. In addition, the change of RRI values measured by ultrasound images based on the artificial intelligence PCNN was associated with AKI.

scholarly journals A clinical, proteomics, and artificial intelligence-driven model to predict acute kidney injury in patients undergoing coronary angiography

2019 ◽  
Vol 42 (2) ◽  
pp. 292-298 ◽  
Author(s):  
Nasrien E. Ibrahim ◽  
Cian P. McCarthy ◽  
Shreya Shrestha ◽  
Hanna K. Gaggin ◽  
Renata Mukai ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Acute Kidney Injury ◽  
Coronary Angiography ◽  
Kidney Injury ◽  
Clinical Proteomics

scholarly journals The potential for artificial intelligence to predict clinical outcomes in patients who have acquired acute kidney injury during the perioperative period

2021 ◽  
Vol 10 (1) ◽  
Author(s):  
Barry J. Kelly ◽  
Julio Chevarria ◽  
Barry O’Sullivan ◽  
George Shorten
Keyword(s):  
Artificial Intelligence ◽  
Acute Kidney Injury ◽  
Kidney Injury ◽  
Perioperative Period ◽  
Medical Problem ◽  
Economic Consequences ◽  
Term Survival ◽  
Hospitalised Patients ◽  
Postoperative Aki

AbstractAcute kidney injury (AKI) is a common medical problem in hospitalised patients worldwide that may result in negative physiological, social and economic consequences. Amongst patients admitted to ICU with AKI, over 40% have had either elective or emergency surgery prior to admission. Predicting outcomes after AKI is difficult and the decision on whom to initiate RRT with a goal of renal recovery or predict a long-term survival benefit still poses a challenge for acute care physicians. With the increasing use of electronic healthcare records, artificial intelligence may allow postoperative AKI prognostication and aid clinical management. Patients will benefit if the data can be readily accessed andregulatory, ethical and human factors challenges can be overcome.

scholarly journals Artificial Intelligence Imaging to Observe the Protective Effect of Hydrogen Sulfide on Acute Kidney Injury Caused by Urinary Sepsis

2021 ◽  
Vol 2021 ◽  
pp. 1-13
Author(s):  
Ziying Yu ◽  
Xiaoli Zhang
Keyword(s):  
Artificial Intelligence ◽  
Acute Kidney Injury ◽  
Hydrogen Sulfide ◽  
Protective Effect ◽  
Kidney Injury ◽  
Renal Tissue ◽  
Diagnosis And Treatment ◽  
Sepsis Group ◽  
Short Period ◽  
Urinary Sepsis

With the development of medical technology products and the rapid development of computer technology, medical AI has become a hotbed in scientific research and clinical practice. Some medical AI-assisted diagnosis has been applied to the clinic to assist doctors in formulating treatment plans. The traditional method of clinical diagnosis and treatment is that the physician makes an intentional diagnosis and then performs ancillary tests. The clinician performs diagnosis and treatment by identifying clinical symptoms and analyzing auxiliary examination results. Modern medical AI is based on big data collection and analyzes the test results through artificial intelligence and computer algorithms. It can output diagnostic results with high sensitivity and specificity for clinical tests. Acute kidney injury (AKI) is a common clinical emergency. The main clinical features are elevated blood creatinine, decreased urine output, and sharp decline in renal function within a short period of time, and it is a hot spot worldwide. In this experiment, a rabbit sepsis model was replicated by inoculating E. coli bacteria into the rabbit’s unilateral ureteral lumen and ligation. NaHS was used as an exogenous hydrogen sulfide donor to observe the effects of hydrogen sulfide on UTIs. The protective effect of oxidative stress and inflammatory response in acute kidney injury with hyperemia. In the experiment, the production of endogenous hydrogen sulfide was decreased in the Sepsis group, and the renal CSE activity was decreased, while the content of endogenous hydrogen sulfide in the NaHS group was higher than that of the Sepsis group, and the CSE activity of renal tissue was increased. It can be seen that the plasma hydrogen sulfide and renal tissue SCE levels in septic acute kidney injury increased after NaHS intervention, and the renal tissue damage was reduced, suggesting that hydrogen sulfide is mainly generated endogenously through the action of CSE, which causes damage to the kidneys. The expressions of iNOS and HO-1 in renal tissues of urinary sepsis are increased. H2S can play a certain protective effect on acute kidney injury in urinary sepsis by down-regulating iNOS and up-regulating the expression of HO-1.

scholarly journals Artificial intelligence to guide management of acute kidney injury in the ICU

2020 ◽  
Vol 26 (6) ◽  
pp. 563-573
Author(s):  
Greet De Vlieger ◽  
Kianoush Kashani ◽  
Geert Meyfroidt
Keyword(s):  
Artificial Intelligence ◽  
Acute Kidney Injury ◽  
Kidney Injury

scholarly journals Artificial Intelligence for Risk Prediction of Rehospitalization with Acute Kidney Injury in Sepsis Survivors

2022 ◽  
Vol 12 (1) ◽  
pp. 43
Author(s):  
Shuo-Ming Ou ◽  
Kuo-Hua Lee ◽  
Ming-Tsun Tsai ◽  
Wei-Cheng Tseng ◽  
Yuan-Chia Chu ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Acute Kidney Injury ◽  
Kidney Injury ◽  
Gradient Boosting ◽  
Light Gradient ◽  
Cell Counts ◽  
Tree Classifier ◽  
Extreme Gradient Boosting ◽  
Sepsis Survivors

Sepsis survivors have a higher risk of long-term complications. Acute kidney injury (AKI) may still be common among sepsis survivors after discharge from sepsis. Therefore, our study utilized an artificial-intelligence-based machine learning approach to predict future risks of rehospitalization with AKI between 1 January 2008 and 31 December 2018. We included a total of 23,761 patients aged ≥ 20 years who were admitted due to sepsis and survived to discharge. We adopted a machine learning method by using models based on logistic regression, random forest, extra tree classifier, gradient boosting decision tree (GBDT), extreme gradient boosting, and light gradient boosting machine (LGBM). The LGBM model exhibited the highest area under the receiver operating characteristic curves (AUCs) of 0.816 to predict rehospitalization with AKI in sepsis survivors and followed by the GBDT model with AUCs of 0.813. The top five most important features in the LGBM model were C-reactive protein, white blood cell counts, use of inotropes, blood urea nitrogen and use of diuretics. We established machine learning models for the prediction of the risk of rehospitalization with AKI in sepsis survivors, and the machine learning model may set the stage for the broader use of clinical features in healthcare.

Sign in / Sign up

Export Citation Format

Share Document