Machine learning approach in mortality rate prediction for hemodialysis patients

Analyzing Monthly Blood Test Data to Forecast 30-Day Hospital Readmissions among Maintenance Hemodialysis Patients

Background: The increase in the global population of hemodialysis patients is linked to aging demographics and the prevalence of conditions such as arterial hypertension and diabetes mellitus. While previous research in hemodialysis has mainly focused on mortality predictions, there is a gap in studies targeting short-term hospitalization predictions using detailed, monthly blood test data. Methods: This study employs advanced data preprocessing and machine learning techniques to predict hospitalizations within a 30-day period among hemodialysis patients. Initial steps include employing K-Nearest Neighbor (KNN) imputation to address missing data and using the Synthesized Minority Oversampling Technique (SMOTE) to ensure data balance. The study then applies a Support Vector Machine (SVM) algorithm for the predictive analysis, with an additional enhancement through ensemble learning techniques, in order to improve prediction accuracy. Results: The application of SVM in predicting hospitalizations within a 30-day period among hemodialysis patients resulted in an impressive accuracy rate of 93%. This accuracy rate further improved to 96% upon incorporating ensemble learning methods, demonstrating the efficacy of the chosen machine learning approach in this context. Conclusions: This study highlights the potential of utilizing machine learning to predict hospital readmissions within a 30-day period among hemodialysis patients based on monthly blood test data. It represents a significant leap towards precision medicine and personalized healthcare for this patient group, suggesting a paradigm shift in patient care through the proactive identification of hospitalization risks.

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.

Mortality prediction with adaptive feature importance recalibration for peritoneal dialysis patients

The study aims to develop AICare, an interpretable mortality prediction model, using electronic medical records (EMR) from follow-up visits for end-stage renal disease (ESRD) patients. AICare includes a multichannel feature extraction module and an adaptive feature importance recalibration module. It integrates dynamic records and static features to perform personalized health context representation learning. The dataset encompasses 13,091 visits and demographic data of 656 peritoneal dialysis (PD) patients spanning 12 years. An additional public dataset of 4,789 visits from 1,363 hemodialysis (HD) patients is also considered. AICare outperforms traditional deep learning models in mortality prediction while retaining interpretability. It uncovers mortality-feature relationships and variations in feature importance and provides reference values. An AI-doctor interaction system is developed for visualizing patients' health trajectories and risk indicators.

Machine learning approaches for the mortality risk assessment of patients undergoing hemodialysis

Introduction:

Mortality is a major primary endpoint for long-term hemodialysis (HD) patients. The clinical status of HD patients generally relies on longitudinal clinical observations such as monthly laboratory examinations and physical examinations.

Methods:

A total of 829 HD patients who met the inclusion criteria were analyzed. All patients were tracked from January 2009 to December 2013. Taken together, this study performed full-adjusted-Cox proportional hazards (CoxPH), stepwise-CoxPH, random survival forest (RSF)-CoxPH, and whale optimization algorithm (WOA)-CoxPH model for the all-cause mortality risk assessment in HD patients. The model performance between proposed selections of CoxPH models were evaluated using concordance index.

Results:

The WOA-CoxPH model obtained the highest concordance index compared with RSF-CoxPH and typical selection CoxPH model. The eight significant parameters obtained from the WOA-CoxPH model, including age, diabetes mellitus (DM), hemoglobin (Hb), albumin, creatinine (Cr), potassium (K), Kt/V, and cardiothoracic ratio, have also showed significant survival difference between low- and high-risk characteristics in single-factor analysis. By integrating the risk characteristics of each single factor, patients who obtained seven or more risk characteristics of eight selected parameters were dichotomized as high-risk subgroup, and remaining is considered as low-risk subgroup. The integrated low- and high-risk subgroup showed greater discrepancy compared with each single risk factor selected by WOA-CoxPH model.

Conclusion:

The study findings revealed WOA-CoxPH model could provide better risk assessment performance compared with RSF-CoxPH and typical selection CoxPH model in the HD patients. In summary, patients who had seven or more risk characteristics of eight selected parameters were at potentially increased risk of all-cause mortality in HD population.

Clinical risk assessment of chronic kidney disease patients using genetic programming

Chronic kidney disease (CKD) is one of the serious health concerns in the twenty-first century. CKD impacts over 37 million Americans. By applying machine learning (ML) techniques to clinical data, CKD can be diagnosed early. This early detection of CKD can prevent numerous loss of life. In this work, clinical data set of 400 patients, available on the UCI repository, are taken. Unfortunately, this data set doesn't have an equal distribution of CKD and Non-CKD samples. This imbalanced nature of data highly influences the learning capabilities of classifiers. Genetic Programming (GP) is an ML technique based on the evolution of species. GP with standard fitness function, also impacted by this imbalanced nature of data. A new Euclidean distance-based fitness function in GP is proposed to handle this imbalanced nature of the data set. To compare the robustness of the proposed work, other classification techniques, K-nearest neighborhood (KNN), KNN with particle swarm optimization (PSO), and GP with the standard fitness function, is also applied. For ten-fold cross-validation, the KNN shows an accuracy of 83.54% with an AUC value of 0.69, the PSO-KNN shows an accuracy of 96.79% with an AUC value of 0.94, and the GP, with the newly proposed fitness function, supersedes KNN and PSO-KNN and shows the accuracy of 99.33% with an AUC value of 0.99.