Artificial intelligence: a new approach for prescription and monitoring of hemodialysis therapy

Precision Dialysis: Leveraging Big Data and Artificial Intelligence

The long-term mortality of patients with kidney failure remains unacceptably high. There are a multitude of reasons for the unfavorable status quo of dialysis care, such as the inadequate and suboptimal pattern of uremic toxin removal resulting in a metabolic and hemodynamic "roller coaster" induced by thrice-weekly in-center hemodialysis. Innovation in dialysis delivery systems is needed to build an adaptive and self-improving process to change the status quo of dialysis care with the aim of transforming it from being reactive to being proactive. The introduction of more physiologic and smart dialysis systems using artificial intelligence (AI) incorporating real-time data into the process of dialysis delivery is a realistic target. This would enable machine learning from both individual and collective patient treatment data. This has the potential to shift the paradigm from the practice of population-driven, evidence-based data to precision medicine. In this review, we describe the different components of an AI system, discuss the studied applications of AI in the field of dialysis, and outline parameters that can be used for future smart, adaptive dialysis delivery systems. The desired output is precision dialysis; a self-improving process that has the ability to prognosticate and develop instant and individualized predictive models.

Prediction the prognosis of the poisoned patients undergoing hemodialysis using machine learning algorithms

BACKGROUND

Hemodialysis is a life-saving treatment used to eliminate toxins and metabolites from the body during poisoning. Despite its effectiveness, there needs to be more research on this method precisely, with most studies focusing on specific poisoning. This study aims to bridge the existing knowledge gap by developing a machine-learning prediction model for forecasting the prognosis of the poisoned patient undergoing hemodialysis.

METHODS

Using a registry database from 2016 to 2022, this study conducted a retrospective cohort study at Loghman Hakim Hospital. First, the relief feature selection algorithm was used to identify the most important variables influencing the prognosis of poisoned patients undergoing hemodialysis. Second, four machine learning algorithms, including extreme gradient boosting (XGBoost), histgradient boosting (HGB), k-nearest neighbors (KNN), and adaptive boosting (AdaBoost), were trained to construct predictive models for predicting the prognosis of poisoned patients undergoing hemodialysis. Finally, the performance of paired feature selection and machine learning (ML) algorithm were evaluated to select the best models using five evaluation metrics including accuracy, sensitivity, specificity the area under the curve (AUC), and f1-score.

RESULT

The study comprised 980 patients in total. The experimental results showed that ten variables had a significant influence on prognosis outcomes including age, intubation, acidity (PH), previous medical history, bicarbonate (HCO3), Glasgow coma scale (GCS), intensive care unit (ICU) admission, acute kidney injury, and potassium. Out of the four models evaluated, the HGB classifier stood out with superior results on the test dataset. It achieved an impressive mean classification accuracy of 94.8%, a mean specificity of 93.5 a mean sensitivity of 94%, a mean F-score of 89.2%, and a mean receiver operating characteristic (ROC) of 92%.

CONCLUSION

ML-based predictive models can predict the prognosis of poisoned patients undergoing hemodialysis with high performance. The developed ML models demonstrate valuable potential for providing frontline clinicians with data-driven, evidence-based tools to guide time-sensitive prognosis evaluations and care decisions for poisoned patients in need of hemodialysis. Further large-scale multi-center studies are warranted to validate the efficacy of these models across diverse populations.

Digital health and acute kidney injury: consensus report of the 27th Acute Disease Quality Initiative workgroup

Acute kidney injury (AKI), which is a common complication of acute illnesses, affects the health of individuals in community, acute care and post-acute care settings. Although the recognition, prevention and management of AKI has advanced over the past decades, its incidence and related morbidity, mortality and health care burden remain overwhelming. The rapid growth of digital technologies has provided a new platform to improve patient care, and reports show demonstrable benefits in care processes and, in some instances, in patient outcomes. However, despite great progress, the potential benefits of using digital technology to manage AKI has not yet been fully explored or implemented in clinical practice. Digital health studies in AKI have shown variable evidence of benefits, and the digital divide means that access to digital technologies is not equitable. Upstream research and development costs, limited stakeholder participation and acceptance, and poor scalability of digital health solutions have hindered their widespread implementation and use. Here, we provide recommendations from the Acute Disease Quality Initiative consensus meeting, which involved experts in adult and paediatric nephrology, critical care, pharmacy and data science, at which the use of digital health for risk prediction, prevention, identification and management of AKI and its consequences was discussed.

Predict, diagnose, and treat chronic kidney disease with machine learning: a systematic literature review

OBJECTIVES

In this systematic review we aimed at assessing how artificial intelligence (AI), including machine learning (ML) techniques have been deployed to predict, diagnose, and treat chronic kidney disease (CKD). We systematically reviewed the available evidence on these innovative techniques to improve CKD diagnosis and patient management.

METHODS

We included English language studies retrieved from PubMed. The review is therefore to be classified as a "rapid review", since it includes one database only, and has language restrictions; the novelty and importance of the issue make missing relevant papers unlikely. We extracted 16 variables, including: main aim, studied population, data source, sample size, problem type (regression, classification), predictors used, and performance metrics. We followed the Preferred Reporting Items for Systematic Reviews (PRISMA) approach; all main steps were done in duplicate.

RESULTS

From a total of 648 studies initially retrieved, 68 articles met the inclusion criteria. Models, as reported by authors, performed well, but the reported metrics were not homogeneous across articles and therefore direct comparison was not feasible. The most common aim was prediction of prognosis, followed by diagnosis of CKD. Algorithm generalizability, and testing on diverse populations was rarely taken into account. Furthermore, the clinical evaluation and validation of the models/algorithms was perused; only a fraction of the included studies, 6 out of 68, were performed in a clinical context.

CONCLUSIONS

Machine learning is a promising tool for the prediction of risk, diagnosis, and therapy management for CKD patients. Nonetheless, future work is needed to address the interpretability, generalizability, and fairness of the models to ensure the safe application of such technologies in routine clinical practice.

Big Data in Nephrology

A huge array of data in nephrology is collected through patient registries, large epidemiological studies, electronic health records, administrative claims, clinical trial repositories, mobile health devices and molecular databases. Application of these big data, particularly using machine-learning algorithms, provides a unique opportunity to obtain novel insights into kidney diseases, facilitate personalized medicine and improve patient care. Efforts to make large volumes of data freely accessible to the scientific community, increased awareness of the importance of data sharing and the availability of advanced computing algorithms will facilitate the use of big data in nephrology. However, challenges exist in accessing, harmonizing and integrating datasets in different formats from disparate sources, improving data quality and ensuring that data are secure and the rights and privacy of patients and research participants are protected. In addition, the optimism for data-driven breakthroughs in medicine is tempered by scepticism about the accuracy of calibration and prediction from in silico techniques. Machine-learning algorithms designed to study kidney health and diseases must be able to handle the nuances of this specialty, must adapt as medical practice continually evolves, and must have global and prospective applicability for external and future datasets.

Machine Learning Analysis of Time-Dependent Features for Predicting Adverse Events During Hemodialysis Therapy: Model Development and Validation Study

BACKGROUND

Hemodialysis (HD) therapy is an indispensable tool used in critical care management. Patients undergoing HD are at risk for intradialytic adverse events, ranging from muscle cramps to cardiac arrest. So far, there is no effective HD device-integrated algorithm to assist medical staff in response to these adverse events a step earlier during HD.

OBJECTIVE

We aimed to develop machine learning algorithms to predict intradialytic adverse events in an unbiased manner.

METHODS

Three-month dialysis and physiological time-series data were collected from all patients who underwent maintenance HD therapy at a tertiary care referral center. Dialysis data were collected automatically by HD devices, and physiological data were recorded by medical staff. Intradialytic adverse events were documented by medical staff according to patient complaints. Features extracted from the time series data sets by linear and differential analyses were used for machine learning to predict adverse events during HD.

RESULTS

Time series dialysis data were collected during the 4-hour HD session in 108 patients who underwent maintenance HD therapy. There were a total of 4221 HD sessions, 406 of which involved at least one intradialytic adverse event. Models were built by classification algorithms and evaluated by four-fold cross-validation. The developed algorithm predicted overall intradialytic adverse events, with an area under the curve (AUC) of 0.83, sensitivity of 0.53, and specificity of 0.96. The algorithm also predicted muscle cramps, with an AUC of 0.85, and blood pressure elevation, with an AUC of 0.93. In addition, the model built based on ultrafiltration-unrelated features predicted all types of adverse events, with an AUC of 0.81, indicating that ultrafiltration-unrelated factors also contribute to the onset of adverse events.

CONCLUSIONS

Our results demonstrated that algorithms combining linear and differential analyses with two-class classification machine learning can predict intradialytic adverse events in quasi-real time with high AUCs. Such a methodology implemented with local cloud computation and real-time optimization by personalized HD data could warn clinicians to take timely actions in advance.

Dialysis adequacy predictions using a machine learning method

Dialysis adequacy is an important survival indicator in patients with chronic hemodialysis. However, there are inconveniences and disadvantages to measuring dialysis adequacy by blood samples. This study used machine learning models to predict dialysis adequacy in chronic hemodialysis patients using repeatedly measured data during hemodialysis. This study included 1333 hemodialysis sessions corresponding to the monthly examination dates of 61 patients. Patient demographics and clinical parameters were continuously measured from the hemodialysis machine; 240 measurements were collected from each hemodialysis session. Machine learning models (random forest and extreme gradient boosting [XGBoost]) and deep learning models (convolutional neural network and gated recurrent unit) were compared with multivariable linear regression models. The mean absolute percentage error (MAPE), root mean square error (RMSE), and Spearman's rank correlation coefficient (Corr) for each model using fivefold cross-validation were calculated as performance measurements. The XGBoost model had the best performance among all methods (MAPE = 2.500; RMSE = 2.906; Corr = 0.873). The deep learning models with convolutional neural network (MAPE = 2.835; RMSE = 3.125; Corr = 0.833) and gated recurrent unit (MAPE = 2.974; RMSE = 3.230; Corr = 0.824) had similar performances. The linear regression models had the lowest performance (MAPE = 3.284; RMSE = 3.586; Corr = 0.770) compared with other models. Machine learning methods can accurately infer hemodialysis adequacy using continuously measured data from hemodialysis machines.

Machine learning in nephrology: scratching the surface

Machine learning shows enormous potential in facilitating decision-making regarding kidney diseases. With the development of data preservation and processing, as well as the advancement of machine learning algorithms, machine learning is expected to make remarkable breakthroughs in nephrology. Machine learning models have yielded many preliminaries to moderate and several excellent achievements in the fields, including analysis of renal pathological images, diagnosis and prognosis of chronic kidney diseases and acute kidney injury, as well as management of dialysis treatments. However, it is just scratching the surface of the field; at the same time, machine learning and its applications in renal diseases are facing a number of challenges. In this review, we discuss the application status, challenges and future prospects of machine learning in nephrology to help people further understand and improve the capacity for prediction, detection, and care quality in kidney diseases.

Using Artificial Intelligence Resources in Dialysis and Kidney Transplant Patients: A Literature Review

BACKGROUND

The purpose of this review is to depict current research and impact of artificial intelligence/machine learning (AI/ML) algorithms on dialysis and kidney transplantation. Published studies were presented from two points of view: What medical aspects were covered? What AI/ML algorithms have been used?

METHODS

We searched four electronic databases or studies that used AI/ML in hemodialysis (HD), peritoneal dialysis (PD), and kidney transplantation (KT). Sixty-nine studies were split into three categories: AI/ML and HD, PD, and KT, respectively. We identified 43 trials in the first group, 8 in the second, and 18 in the third. Then, studies were classified according to the type of algorithm.

RESULTS

AI and HD trials covered: (a) dialysis service management, (b) dialysis procedure, (c) anemia management, (d) hormonal/dietary issues, and (e) arteriovenous fistula assessment. PD studies were divided into (a) peritoneal technique issues, (b) infections, and (c) cardiovascular event prediction. AI in transplantation studies were allocated into (a) management systems (ML used as pretransplant organ-matching tools), (b) predicting graft rejection, (c) tacrolimus therapy modulation, and (d) dietary issues.

CONCLUSIONS

Although guidelines are reluctant to recommend AI implementation in daily practice, there is plenty of evidence that AI/ML algorithms can predict better than nephrologists: volumes, Kt/V, and hypotension or cardiovascular events during dialysis. Altogether, these trials report a robust impact of AI/ML on quality of life and survival in G5D/T patients. In the coming years, one would probably witness the emergence of AI/ML devices that facilitate the management of dialysis patients, thus increasing the quality of life and survival.

Artificial Intelligence in Nephrology: Core Concepts, Clinical Applications, and Perspectives

Artificial intelligence is playing an increasingly important role in many fields of medicine, assisting physicians in most steps of patient management. In nephrology, artificial intelligence can already be used to improve clinical care, hemodialysis prescriptions, and follow-up of transplant recipients. However, many nephrologists are still unfamiliar with the basic principles of medical artificial intelligence. This review seeks to provide an overview of medical artificial intelligence relevant to the practicing nephrologist, in all fields of nephrology. We define the core concepts of artificial intelligence and machine learning and cover the basics of the functioning of neural networks and deep learning. We also discuss the most recent clinical applications of artificial intelligence in nephrology and medicine; as an example, we describe how artificial intelligence can predict the occurrence of progressive immunoglobulin A nephropathy. Finally, we consider the future of artificial intelligence in clinical nephrology and its impact on medical practice, and conclude with a discussion of the ethical issues that the use of artificial intelligence raises in terms of clinical decision making, physician-patient relationship, patient privacy, and data collection.

Progress in the Development and Challenges for the Use of Artificial Kidneys and Wearable Dialysis Devices

Background

Renal transplantation is the treatment of choice for chronic kidney disease (CKD) patients, but the shortage of kidneys and the disabling medical conditions these patients suffer from make dialysis essential for most of them. Since dialysis drastically affects the patients' lifestyle, there are great expectations for the development of wearable artificial kidneys, although their use is currently impeded by major concerns about safety. On the other hand, dialysis patients with hemodynamic instability do not usually tolerate intermittent dialysis therapy because of their inability to adapt to a changing scenario of unforeseen events. Thus, the development of novel wearable dialysis devices and the improvement of clinical tolerance will need contributions from new branches of engineering such as artificial intelligence (AI) and machine learning (ML) for the real-time analysis of equipment alarms, dialysis parameters, and patient-related data with a real-time feedback response. These technologies are endowed with abilities normally associated with human intelligence such as learning, problem solving, human speech understanding, or planning and decision-making. Examples of common applications of AI are visual perception (computer vision), speech recognition, and language translation. In this review, we discuss recent progresses in the area of dialysis and challenges for the use of AI in the development of artificial kidneys.

Summary and Key Messages

Emerging technologies derived from AI, ML, electronics, and robotics will offer great opportunities for dialysis therapy, but much innovation is needed before we achieve a smart dialysis machine able to analyze and understand changes in patient homeostasis and to respond appropriately in real time. Great efforts are being made in the fields of tissue engineering and regenerative medicine to provide alternative cell-based approaches for the treatment of renal failure, including bioartificial renal systems and the implantation of bioengineered kidney constructs.

Recognition Patterns Construction of Coronary Heart Disease Patients with Qi Deficiency Syndrome Based on Artificial Neural Network

Coronary heart disease (CHD), called “thoracic obstruction” in TCM, is one of the most important types of heart disease for its high incidence and mortality. The methods of syndrome studies in TCM can not be completely in accordance with these of modern medicine because of the complexity itself. In this paper, we investigated the ability of Artificial Neural Networks (ANNs) to predict CHD patients with or without qi deficiency syndrome. Predictions with Multilayer Perceptron Neural Network (MPLNN, one type of the ANNS), we obtained recognition patterns made up of eight biological parameters. The accuracy of this recognition pattern was 82.2%, and the accuracy of validation pattern was 80.0%.

Application of intelligent systems in asthma disease: designing a fuzzy rule-based system for evaluating level of asthma exacerbation

This paper discusses the capacities of artificial intelligence in the process of asthma diagnosing and asthma treatment. Developed intelligent systems for asthma disease have been classified in five categories including diagnosing, evaluating, management, communicative facilities, and prediction. Considering inputs, results, and methodologies of the systems show that by focusing on meticulous analysis of quality of life as an input variable and developing patient-based systems, under-diagnosing and asthma morbidity and mortality would decrease significantly. Regard to the importance of accurate evaluation in accurate prescription and expeditious treatment, the methodology of developing a fuzzy expert system for evaluating level of asthma exacerbation is presented in this paper too. The performance of this system has been tested in Asthma, Allergy, and Immunology Center of Iran using 25 asthmatic patients. Comparison between system's results and physicians' evaluations using Kappa coefficient (K) reinforces the value of K = 1. In addition this system assigns a degree in gradation (0-10) to every patient representing the slight differences between patients assigned to a specific category.

Applying an artificial neural network to predict total body water in hemodialysis patients

BACKGROUND

Estimating total body water (TBW) is crucial in determining dry weight and dialytic dose for hemodialysis patients. Several anthropometric equations have been used to predict TBW, but a more accurate method is needed. We developed an artificial neural network (ANN) to predict TBW in hemodialysis patients.

METHODS

Demographic data, anthropometric measurements, and multifrequency bioelectrical impedance analysis (MF-BIA) were investigated in 54 patients. TBW measured by MF-BIA (TBW-BIA) was the reference. The predictive value of TBW based on ANN and five anthropometric equations (58% of actual body weight, Watson formula, Hume formula, Chertow formula, and Lee formula) was evaluated.

RESULTS

Predictive TBW values derived from anthropometric equations were significantly higher than TBW-BIA (31.341 +/- 6.033 liters). The only non-significant difference was between TBW-ANN (31.468 +/- 5.301 liters) and TBW-BIA (p = 0.639). ANN had the strongest Pearson's correlation coefficient (0.911) and smallest root mean square error (2.480); its peak centered most closely to zero with the shortest tails in an empirical cumulative distribution plot when compared with the other five equations.

CONCLUSION

ANN could surpass traditional anthropometric equations and serve as a feasible alternative method of TBW estimation for chronic hemodialysis patients.

Comparison of Standard and Artificial Neural Network Estimators of Hemodialysis Adequacy

The National Kidney Foundation and the European Renal Association recommend routine measurement of hemodialysis (HD) dose and have set standards for adequacy of treatment. We compare the results of five methods for HD dose estimation, classifying each result as adequate or inadequate on the basis of equilibrated (eq) Urea Reduction Ratio (URR(eq)) > or = 65% or Kt/V(eq) > or = 1.2, to assess the accuracy of each method as a diagnostic tool. Data from 113 patients from two different dialysis units were analyzed. Equilibrated postdialysis blood urea was measured 60 min after each hemodialysis session to calculate URR(eq) and Kt/V(eq), considered as gold standard indexes (GSI). URR and Kt/V were estimated by using the Smye formula, an artificial neural network (ANN), modified URR, the second generation Kt/V Daugirdas formula, and standard indexes based on postdialysis urea, then compared to the GSI. For URR, best estimator was ANN (error rate: ER% = 12.70), followed by modified URR (ER% = 17.46%), the Smye (ER% = 22.22), and standard URR (ER% = 23.81). For Kt/V, the Daugirdas equation and the ANN were similar (ER% = 9.52 and 11.11). The single-pool Kt/V (Kt/V(sp)) > or = 1.4 (ERA recommended) produced an ER% = 7.94 and a false positive rate (FPR%) equal to that shown by the ANN (FPR% = 3.17). According to the current threshold limits for HD dose adequacy, the ANN was a reliable and accurate tool for URR monitoring, better than the Smye and the modified URR methods. The use of the ANN urea estimation yields accurate results when used to calculate Kt/V. The Kt/V(sp) with an adequacy threshold of 1.4 is a superior approach for HD adequacy monitoring, suggesting that the current adequacy limits should be reviewed for both URR and Kt/V.

Usefulness of artificial neural networks to predict follow-up dietary protein intake in hemodialysis patients

BACKGROUND

Artificial neural networks (ANN) represent a promising alternative to classical statistical and mathematic methods to solve multidimensional nonlinear problems. The aim of the study was to verify, by comparing the performance of ANN with that of experienced nephrologists, whether ANN are useful tools in hemodialysis to predict the follow-up (=1 month after the observation used for the prediction) dietary protein intake (PCR), and whether their performance is influenced by the size of the population and by the data pool used to built the model.

METHODS

A combined retrospective and prospective observational study was performed in two Swiss dialysis units (84 chronic hemodialysis patients, 500 monthly clinical observations and biochemical test results). Using mathematical models based on linear regressions to evaluate the variables, ANN were built and then prospectively and interinstitutionally compared with the ability of six experienced nephrologists to predict the follow-up PCR.

RESULTS

ANN compared with nephrologists gave a more accurate correlation between estimated and calculated follow-up PCR (P < 0.001). The same superiority of ANN was also seen in the ability to detect a follow-up PCR <1.00 g/kg/day expressed as a percentage of correct predictions, sensitivity, specificity, and predictivity. The interinstitutional performance of the ANN is positively influenced by the size and the variability of the population used to build the mathematical model.

CONCLUSION

The use of ANN significantly improves the ability of the experienced nephrologist to estimate and to detect an unsatisfactory (<1.00 g/kg/day) follow-up PCR. The size of the population selected to build the ANN is critical for his performance.

Artificial neural networks improve the prediction of Kt/V, follow-up dietary protein intake and hypotension risk in haemodialysis patients

BACKGROUND

Artificial neural networks (ANN) represent a promising alternative to classical statistical and mathematical methods to solve multidimensional non-linear problems. The aim of the study was to compare the performance of ANN in predicting the dialysis quality (Kt/V), the follow-up dietary protein intake and the risk of intradialytic hypotension in haemodialysis patients with that predicted by experienced nephrologists.

METHODS

A combined retrospective and prospective observational study was performed in two Swiss dialysis units (80 chronic haemodialysis patients, 480 monthly clinical observations and biochemical test results). Using mathematical models based on linear and logistic regressions as background, ANN were built and then prospectively compared with the ability of six experienced nephrologists to predict the Kt/V and the follow-up protein catabolic rate (PCR) and to detect a Kt/V < 1.30, a follow-up PCR < 1.00 g/kg/day and the occurrence of hypotension.

RESULTS

ANN compared with nephrologists gave a more accurate correlation between estimated and calculated Kt/V and follow-up PCR (P<0.001). The same superiority of ANN was also seen in the ability to detect a Kt/V < 1.30, a follow-up PCR < 1.00 g/kg/day and the occurrence of hypotension expressed as a percentage of correct answers, sensitivity, specificity and predictivity.

CONCLUSIONS

The use of ANN significantly improves the ability of experienced nephrologists to estimate the Kt/V and the follow-up PCR and to detect a Kt/V < 1.30, a follow-up PCR < 1.00 g/kg/day and the occurrence of intradialytic hypotension.

Dialysate-side urea kinetics. Neural network predicts dialysis dose during dialysis

Determination of the adequacy of dialysis is a routine but crucial procedure in patient evaluation. The total dialysis dose, expressed as Kt/V, has been widely recognised to be a major determinant of morbidity and mortality in haemodialysed patients. Many different factors influence the correct determination of Kt/V, such as urea sequestration in different body compartments, access and cardiopulmonary recirculation. These factors are responsible for urea rebound after the end of the haemodialysis session, causing poor Kt/V estimation. There are many techniques that try to overcome this problem. Some of them use analysis of blood-side urea samples, and, in recent years, on-line urea monitors have become available to calculate haemodialysis dose from dialysate-side urea kinetics. All these methods require waiting until the end of the session to calculate the Kt/V dose. In this work, a neural network (NN) method is presented for early prediction of the Kt/V dose. Two different portions of the dialysate urea concentration-time profile (provided by an on-line urea monitor) were analysed: the entire curve A and the first half B, using an NN to predict the Kt/V and compare this with that provided by the monitor. The NN was able to predict Kt/V is the middle of the 4h session (B data) without a significant increase in the percentage error (B data: 6.69% +/- 2.46%; A data: 5.58% +/- 8.77%, mean +/- SD) compared with the monitor Kt/V.