Predictive modeling of blood pressure during hemodialysis: a comparison of linear model, random forest, support vector regression, XGBoost, LASSO regression and ensemble method

Schistosomiasis transmission in Zimbabwe: Modelling based on machine learning

Zimbabwe, located in Southern Africa, faces a significant public health challenge due to schistosomiasis. We investigated this issue with emphasis on risk prediction of schistosomiasis for the entire population. To this end, we reviewed available data on schistosomiasis in Zimbabwe from a literature search covering the 1980-2022 period considering the potential impact of 26 environmental and socioeconomic variables obtained from public sources. We studied the population requiring praziquantel with regard to whether or not mass drug administration (MDA) had been regularly applied. Three machine-learning algorithms were tested for their ability to predict the prevalence of schistosomiasis in Zimbabwe based on the mean absolute error (MAE), the root mean squared error (RMSE) and the coefficient of determination (R2). The findings revealed different roles of the 26 factors with respect to transmission and there were particular variations between Schistosoma haematobium and S. mansoni infections. We found that the top-five correlation factors, such as the past (rather than current) time, unsettled MDA implementation, constrained economy, high rainfall during the warmest season, and high annual precipitation were closely associated with higher S. haematobium prevalence, while lower elevation, high rainfall during the warmest season, steeper slope, past (rather than current) time, and higher minimum temperature in the coldest month were rather related to higher S. mansoni prevalence. The random forest (RF) algorithm was considered as the formal best model construction method, with MAE = 0.108; RMSE = 0.143; and R2 = 0.517 for S. haematobium, and with the corresponding figures for S. mansoni being 0.053; 0.082; and 0.458. Based on this optimal model, the current total schistosomiasis prevalence in Zimbabwe under MDA implementation was 19.8%, with that of S. haematobium at 13.8% and that of S. mansoni at 7.1%, requiring annual MDA based on a population of 3,003,928. Without MDA, the current total schistosomiasis prevalence would be 23.2%, that of S. haematobium 17.1% and that of S. mansoni prevalence at 7.4%, requiring annual MDA based on a population of 3,521,466. The study reveals that MDA alone is insufficient for schistosomiasis elimination, especially that due to S. mansoni. This study predicts a moderate prevalence of schistosomiasis in Zimbabwe, with its elimination requiring comprehensive control measures beyond the currently used strategies, including health education, snail control, population surveillance and environmental management.

Investigating the impact of climate variables on the organic honey yield in Turkey using XGBoost machine learning

BACKGROUND

The Turkish organic honey industry, a major player in the global market, faces challenges due to climate fluctuations. Understanding the influence of climate factors on honey production is vital for sustainable farming and economic stability.

METHOD

This study uses a machine learning approach with the XGBoost algorithm to analyze temperature, wind, humidity, precipitation and surface pressure over a 20-year period from 2004 to 2023.

RESULTS

The results show that these factors significantly impact organic honey production, with temperature, wind, humidity, precipitation and surface pressure having effects of 41.20%, 26.50%, 12.47%, 11.42% and 8.41%, respectively. Sensitivity analysis reveals the model's sensitivity to even minor fluctuations in these variables.

CONCLUSION

The results of this research underscore the necessity of integrating climate change mitigation and adaptation measures into agricultural policies and beekeeping practices. This study showcases the practical application of machine learning in deciphering the intricate relationship between climate change and the production of crops, emphasizing the importance of data-driven decision-making to guarantee long-term sustainability and financial stability in the sector. © 2024 The Author(s). Journal of the Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Long-Term Statistical Process Monitoring of an Ultrafiltration Water Treatment Process

As water treatment technology has improved, the amount of available process data has substantially increased, making real-time, data-driven fault detection a reality. One shortcoming of the fault detection literature is that methods are usually evaluated by comparing their performance on hand-picked, short-term case studies, which yields no insight into long-term performance. In this work, we first evaluate multiple statistical and machine learning approaches for detrending process data. Then, we evaluate the performance of a PCA-based fault detection approach, applied to the detrended data, to monitor influent water quality, filtrate quality, and membrane fouling of an ultrafiltration membrane system for indirect potable reuse. Based on two short case studies, the adaptive lasso detrending method is selected, and the performance of the multivariate approach is evaluated over more than a year. The method is tested for different sets of three critical tuning parameters, and we find that for long-term, autonomous monitoring to be successful, these parameters should be carefully evaluated. However, in comparison with industry standards of simpler, univariate monitoring or daily pressure decay tests, multivariate monitoring produces substantial benefits in long-term testing.

A back propagation neural network based respiratory motion modelling method

BACKGROUND

This study presents the development of a backpropagation neural network-based respiratory motion modelling method (BP-RMM) for precisely tracking arbitrary points within lung tissue throughout free respiration, encompassing deep inspiration and expiration phases.

METHODS

Internal and external respiratory data from four-dimensional computed tomography (4DCT) are processed using various artificial intelligence algorithms. Data augmentation through polynomial interpolation is employed to enhance dataset robustness. A BP neural network is then constructed to comprehensively track lung tissue movement.

RESULTS

The BP-RMM demonstrates promising accuracy. In cases from the public 4DCT dataset, the average target registration error (TRE) between authentic deep respiration phases and those forecasted by BP-RMM for 75 marked points is 1.819 mm. Notably, TRE for normal respiration phases is significantly lower, with a minimum error of 0.511 mm.

CONCLUSIONS

The proposed method is validated for its high accuracy and robustness, establishing it as a promising tool for surgical navigation within the lung.

Predictive modeling of plastic pyrolysis process for the evaluation of activation energy: Explainable artificial intelligence based comprehensive insights

Pyrolysis, a thermochemical conversion approach of transforming plastic waste to energy has tremendous potential to manage the exponentially increasing plastic waste. However, understanding the process kinetics is fundamental to engineering a sustainable process. Conventional analysis techniques do not provide insights into the influence of characteristics of feedstock on the process kinetics. Present study exemplifies the efficacy of using machine learning for predictive modeling of pyrolysis of waste plastics to understand the complexities of the interrelations of predictor variables and their influence on activation energy. The activation energy for pyrolysis of waste plastics was evaluated using machine learning models namely Random Forest, XGBoost, CatBoost, and AdaBoost regression models. Feature selection based on the multicollinearity of data and hyperparameter tuning of the models utilizing RandomizedSearchCV was conducted. Random forest model outperformed the other models with coefficient of determination (R2) value of 0.941, root mean square error (RMSE) value of 14.69 and mean absolute error (MAE) value of 8.66 for the testing dataset. The explainable artificial intelligence-based feature importance plot and the summary plot of the shapely additive explanations projected fixed carbon content, ash content, conversion value, and carbon content as significant parameters of the model in the order; fixed carbon > carbon > ash content > degree of conversion. Present study highlighted the potential of machine learning as a powerful tool to understand the influence of the characteristics of plastic waste and the degree of conversion on the activation energy of a process that is essential for designing the large-scale operations and future scale-up of the process.

Recent advances in microplastic removal from drinking water by coagulation: Removal mechanisms and influencing factors

Microplastics (MPs) are emerging contaminants that are widely detected in drinking water and pose a potential risk to humans. Therefore, the MP removal from drinking water is a critical challenge. Recent studies have shown that MPs can be removed by coagulation. However, the coagulation removal of MPs from drinking water remains inadequately understood. Herein, the efficiency, mechanisms, and influencing factors of coagulation for removing MPs from drinking water are critically reviewed. First, the efficiency of MP removal by coagulation in drinking water treatment plants (DWTPs) and laboratories was comprehensively summarized, which indicated that coagulation plays an important role in MP removal from drinking water. The difference in removal effectiveness between the DWTPs and laboratory was mainly due to variations in treatment conditions and limitations of the detection techniques. Several dominant coagulation mechanisms for removing MPs and their research methods are thoroughly discussed. Charge neutralization is more relevant for small-sized MPs, whereas large-sized MPs are more dependent on adsorption bridging and sweeping. Furthermore, the factors influencing the efficiency of MP removal were jointly analyzed using meta-analysis and a random forest model. The meta-analysis was used to quantify the individual effects of each factor on coagulation removal efficiency by performing subgroup analysis. The random forest model quantified the relative importance of the influencing factors on removal efficiency, the results of which were ordered as follows: MPs shape > Coagulant type > Coagulant dosage > MPs concentration > MPs size > MPs type > pH. Finally, knowledge gaps and potential future directions are proposed. This review assists in the understanding of the coagulation removal of MPs, and provides novel insight into the challenges posed by MPs in drinking water.

Developing a machine learning model for bleeding prediction in patients with cancer-associated thrombosis receiving anticoagulation therapy

BACKGROUND

Only 1 conventional score is available for assessing bleeding risk in patients with cancer-associated thrombosis (CAT): the CAT-BLEED score.

OBJECTIVES

Our aim was to develop a machine learning-based risk assessment model for predicting bleeding in CAT and to evaluate its predictive performance in comparison to that of the CAT-BLEED score.

METHODS

We collected 488 attributes (clinical data, biochemistry, and International Classification of Diseases, 10th Revision, diagnosis) in 1080 unique patients with CAT. We compared CAT-BLEED score, Ridge and Lasso logistic regression, random forest, and Extreme Gradient Boosting (XGBoost) algorithms for predicting major bleeding or clinically relevant nonmajor bleeding occurring 1 to 90 days, 1 to 365 days, and 90 to 455 days after venous thromboembolism (VTE).

RESULTS

The predictive performances of Lasso logistic regression, random forest, and XGBoost were higher than that of the CAT-BLEED score in the prediction of bleeding occurring 1 to 90 days and 1 to 365 days after VTE. For predicting major bleeding or clinically relevant nonmajor bleeding 1 to 90 days after VTE, the CAT-BLEED score achieved a mean area under the receiver operating characteristic curve (AUROC) of 0.48 ± 0.13, while Lasso logistic regression and XGBoost both achieved AUROCs of 0.64 ± 0.12. For predicting bleeding 1 to 365 days after VTE, the CAT-BLEED score achieved a mean AUROC of 0.47 ± 0.08, while Lasso logistic regression and XGBoost achieved AUROCs of 0.64 ± 0.08 and 0.59 ± 0.08, respectively.

CONCLUSION

This is the first machine learning-based risk model for bleeding prediction in patients with CAT receiving anticoagulation therapy. Its predictive performance was higher than that of the conventional CAT-BLEED score. With further development, this novel algorithm might enable clinicians to perform personalized anticoagulation strategies with improved clinical outcomes.

Taslim Uddin Raju S, Hossain MI, Dipto SA, M. Tanvir Uddin S, Sijan A, Chowdhury MAS, Ahmad A, Mahamudul Hasan Khan M. A continuous cuffless blood pressure measurement from optimal PPG characteristic features using machine learning algorithms

Background and objective

Hypertension is a potentially dangerous health condition that can be detected by measuring blood pressure (BP). Blood pressure monitoring and measurement are essential for preventing and treating cardiovascular diseases. Cuff-based devices, on the other hand, are uncomfortable and prevent continuous BP measurement.

Methods

In this study, a new non-invasive and cuff-less method for estimating Systolic Blood Pressure (SBP), Mean Arterial Pressure (MAP), and Diastolic Blood Pressure (DBP) has been proposed using characteristic features of photoplethysmogram (PPG) signals and nonlinear regression algorithms. PPG signals were collected from 219 participants, which were then subjected to preprocessing and feature extraction steps. Analyzing PPG and its derivative signals, a total of 46 time, frequency, and time-frequency domain features were extracted. In addition, the age and gender of each subject were also included as features. Further, correlation-based feature selection (CFS) and Relief F feature selection (ReliefF) techniques were used to select the relevant features and reduce the possibility of over-fitting the models. Finally, support vector regression (SVR), K-nearest neighbour regression (KNR), decision tree regression (DTR), and random forest regression (RFR) were established to develop the BP estimation model. Regression models were trained and evaluated on all features as well as selected features. The best regression models for SBP, MAP, and DBP estimations were selected separately.

Results

The SVR model, along with the ReliefF-based feature selection algorithm, outperforms other algorithms in estimating the SBP, MAP, and DBP with the mean absolute error of 2.49, 1.62 and 1.43 mmHg, respectively. The proposed method meets the Advancement of Medical Instrumentation standard for BP estimations. Based on the British Hypertension Society standard, the results also fall within Grade A for SBP, MAP, and DBP.

Conclusion

The findings show that the method can be used to estimate blood pressure non-invasively, without using a cuff or calibration, and only by utilizing the PPG signal characteristic features.

Assessing the risk of concurrent mycoplasma pneumoniae pneumonia in children with tracheobronchial tuberculosis: retrospective study

Objective

This study aimed to create a predictive model based on machine learning to identify the risk for tracheobronchial tuberculosis (TBTB) occurring alongside Mycoplasma pneumoniae pneumonia in pediatric patients.

Methods

Clinical data from 212 pediatric patients were examined in this retrospective analysis. This cohort included 42 individuals diagnosed with TBTB and Mycoplasma pneumoniae pneumonia (combined group) and 170 patients diagnosed with lobar pneumonia alone (pneumonia group). Three predictive models, namely XGBoost, decision tree, and logistic regression, were constructed, and their performances were assessed using the receiver's operating characteristic (ROC) curve, precision-recall curve (PR), and decision curve analysis (DCA). The dataset was divided into a 7:3 ratio to test the first and second groups, utilizing them to validate the XGBoost model and to construct the nomogram model.

Results

The XGBoost highlighted eight significant signatures, while the decision tree and logistic regression models identified six and five signatures, respectively. The ROC analysis revealed an area under the curve (AUC) of 0.996 for XGBoost, significantly outperforming the other models (p < 0.05). Similarly, the PR curve demonstrated the superior predictive capability of XGBoost. DCA further confirmed that XGBoost offered the highest AIC (43.226), the highest average net benefit (0.764), and the best model fit. Validation efforts confirmed the robustness of the findings, with the validation groups 1 and 2 showing ROC and PR curves with AUC of 0.997, indicating a high net benefit. The nomogram model was shown to possess significant clinical value.

Conclusion

Compared to machine learning approaches, the XGBoost model demonstrated superior predictive efficacy in identifying pediatric patients at risk of concurrent TBTB and Mycoplasma pneumoniae pneumonia. The model's identification of critical signatures provides valuable insights into the pathogenesis of these conditions.

Machine learning for layer-by-layer nanofiltration membrane performance prediction and polymer candidate exploration

In this study, machine learning-based models were established for layer-by-layer (LBL) nanofiltration (NF) membrane performance prediction and polymer candidate exploration. Four different models, i.e., linear, random forest (RF), boosted tree (BT), and eXtreme Gradient Boosting (XGBoost), were formed, and membrane performance prediction was determined in terms of membrane permeability and selectivity. The XGBoost exhibited optimal prediction accuracy for membrane permeability (coefficient of determination (R2): 0.99) and membrane selectivity (R2: 0.80). The Shapley Additive exPlanation (SHAP) method was utilized to evaluate the effects of different LBL NF membrane fabrication conditions on membrane performances. The SHAP method was also used to identify the relationships between polymer structure and membrane performance. Polymers were represented by Morgan fingerprint, which is an effective description approach for developing modeling. Based on the SHAP value results, two reference Morgan fingerprints were constructed containing atomic groups with positive contributions to membrane permeability and selectivity. According to the reference Morgan fingerprint, 204 potential polymers were explored from the largest polymer database (PoLyInfo). By calculating the similarities between each potential polymer and both reference Morgan fingerprints, 23 polymer candidates were selected and could be further used for LBL NF membrane fabrication with the potential for providing good membrane performance. Overall, this work provided new ways both for LBL NF membrane performance prediction and high-performance polymer candidate exploration. The source code for the models and algorithms used in this study is publicly available to facilitate replication and further research. https://github.com/wangliwfsd/LLNMPP/.

Predicting Prostate Surgery Outcomes from Standard Clinical Assessments of Lower Urinary Tract Symptoms To Derive Prognostic Symptom and Flowmetry Criteria

BACKGROUND

Assessment of male lower urinary tract symptoms (LUTS) needs to identify predictors of symptom outcomes when interventional treatment is planned.

OBJECTIVE

To develop a novel prediction model for prostate surgery outcomes and validate it using a separate patient cohort and derive thresholds for key clinical parameters.

DESIGN, SETTING, AND PARTICIPANTS

From the UPSTREAM trial of 820 men seeking treatment for LUTS, analysis of bladder diary (BD), International Prostate Symptom Score (IPSS), IPSS-quality of life, and uroflowmetry data was performed for 176 participants who underwent prostate surgery and provided complete data. For external validation, data from a retrospective database of surgery outcomes in a Japanese urology department (n = 227) were used.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS

Symptom improvement was defined as a reduction in total IPSS of ≥3 points. Multiple logistic regression, classification tree analysis, and random forest models were generated, including versions with and without BD data.

RESULTS AND LIMITATIONS

Multiple logistic regression without BD data identified age (p = 0.029), total IPSS (p = 0.0016), and maximum flow rate (Qmax; p = 0.066) as predictors of outcomes, with area under the receiver operating characteristic curve (AUC) of 77.1%. Classification tree analysis without BD data gave thresholds of IPSS <16 and Qmax ≥13 ml/s (AUC 75.0%). The random forest model, which included all clinical parameters except BD data, had an AUC of 94.7%. Internal validation using the bootstrap method showed reasonable AUCs (69.6-85.8%). Analyses using BD data marginally improved the model fits. External validation gave comparable AUCs for logistic regression, classification tree analysis, and random forest models (all without BD; 70.9%, 67.3%, and 68.5%, respectively). Limitations include the significant number of men with incomplete baseline data and limited assessments in the external validation cohort.

CONCLUSIONS

Outcomes of prostate surgery can be predicted preoperatively using age, total IPSS, and uroflowmetry data, with prognostic thresholds of 16 for IPSS and 13 ml/s for Qmax.

PATIENT SUMMARY

This study identified key preoperative factors that can predict outcomes of prostate surgery for bothersome urinary symptoms, including which patients are at risk of a poor outcome.

Leveraging machine learning to distinguish between bacterial and viral induced pharyngitis using hematological markers: a retrospective cohort study

Accurate differentiation between bacterial and viral-induced pharyngitis is recognized as essential for personalized treatment and judicious antibiotic use. From a cohort of 693 patients with pharyngitis, data from 197 individuals clearly diagnosed with bacterial or viral infections were meticulously analyzed in this study. By integrating detailed hematological insights with several machine learning algorithms, including Random Forest, Neural Networks, Decision Trees, Support Vector Machine, Naive Bayes, and Lasso Regression, for potential biomarkers were identified, with an emphasis being placed on the diagnostic significance of the Monocyte-to-Lymphocyte Ratio. Distinct inflammatory signatures associated with bacterial infections were spotlighted in this study. An innovation introduced in this research was the adaptation of the high-accuracy Lasso Regression model for the TI-84 calculator, with an AUC (95% CI) of 0.94 (0.925-0.955) being achieved. Using this adaptation, pivotal laboratory parameters can be input on-the-spot and infection probabilities can be computed subsequently. This methodology embodies an improvement in diagnostics, facilitating more effective distinction between bacterial and viral infections while fostering judicious antibiotic use.

Models to predict the surgical outcome of mini-ECIRS (endoscopic combined intrarenal surgery) for renal and/or ureteral stones

To establish a safer and more efficient treatment strategy with mini-endoscopic combined intrarenal surgery (ECIRS), the present study aimed to develop models to predict the outcomes of mini-ECIRS in patients with renal and/or ureteral stones. We retrospectively analysed consecutive patients with renal and/or ureteral stones who underwent mini-ECIRS at three Japanese tertiary institutions. Final treatment outcome was evaluated by CT imaging at 1 month postoperatively and stone free (SF) was defined as completely no residual stone or residual stone fragments ≤ 2 mm. Three prognostic models (multiple logistic regression, classification tree analysis, and machine learning-based random forest) were developed to predict surgical outcomes using preoperative clinical factors. Clinical data from 1432 ECIRS were pooled from a database registered at three institutions, and 996 single sessions of mini-ECIRS were analysed in this study. The overall SF rate was 62.3%. The multiple logistic regression model consisted of stone burden (P < 0.001), number of involved calyces (P < 0.001), nephrostomy prior to mini-ECIRS (P = 0.091), and ECOG-PS (P = 0.110), wherein the area under the curve (AUC) was 70.7%. The classification tree analysis consisted of the number of involved calyces with an AUC of 61.7%. The random forest model showed that the top predictive variable was the number of calyces involved, with an AUC of 91.9%. Internal validation revealed that the AUCs for the multiple logistic regression model, classification tree analysis and random forest models were 70.4, 69.6 and 85.9%, respectively. The number of involved calyces, and a smaller stone burden implied a SF outcome. The machine learning-based model showed remarkably high accuracy and may be a promising tool for physicians and patients to obtain proper consent, avoid inefficient surgery, and decide preoperatively on the most efficient treatment strategies, including staged mini-ECIRS.

Machine learning-based prediction models for parathyroid carcinoma using pre-surgery cognitive function and clinical features

Patients with parathyroid carcinoma (PC) are often diagnosed postoperatively, due to incomplete resection during the initial surgery, resulting in poor outcomes. The aim of our study was to investigate the pre-surgery indicators of PC and try to develop a predictive model for PC utilizing machine learning. Evaluation of pre-surgery neuropsychological function and confirmation of pathology were carried out in 133 patients with primary hyperparathyroidism in Beijing Chaoyang Hospital from December 2019 to January 2023. Patients were randomly divided into a training cohort (n = 93) and a validating cohort (n = 40). Analysis of the clinical dataset, two machine learning including the extreme gradient boosting (XGBoost) and the least absolute shrinkage and selection operator (LASSO) regression were utilized to develop the prediction model for PC. Logistic regression analysis was also conducted for comparison. Significant differences in elevated parathyroid hormone and decreased serum phosphorus in PC compared to (BP). The lower score of MMSE and MOCA was observed in PC and a cutoff of MMSE < 24 was the optimal threshold to stratify PC from BP (area under the curve AUC 0.699 vs 0.625). The predicted probability of PC by machine learning was similar to the observed probability in the test set, whereas the logistic model tended to overpredict the possibility of PC. The XGBoost model attained a higher AUC than the logistic algorithms and LASSO models. (0.835 vs 0.683 vs 0.607). Preoperative cognitive function may be a probable predictor for PC. The cognitive function-based prediction model based on the XGBoost algorithm outperformed LASSO and logistic regression, providing valuable preoperative assistance to surgeons in clinical decision-making for patients suspected PC.

Biodegradability analysis of Dioxins through in silico methods: Model construction and mechanism analysis

The biodegradation treatment of dioxins has long been of interest due to its good ecological and economic effects. In this study, the biodegradability of polychlorinated dibenzo-p-dioxins (PCDDs) were investigated by constructing machine learning and multiple linear regression models. The maximum chlorine atomic charge (qHirshfeldCl+), which characterizes the biodegradation ability of PCDDs, was used as the response value. The random forest model was used to rank the importance on the 1471 descriptors of PCDDs, and the BCUTp-1 h, QXZ, JGI4, ATSC8c, VE3_Dt, topoShape, and maxwHBa were screened as the important descriptors by Pearson's correlation coefficient method. A quantitative structure-activity relationship (QSAR) model was constructed to predict the biodegradability of PCDDs. In addition, the extreme gradient boosting (XGBoost) and random forest model were also constructed and proved the good predictability of QSAR model. The biodegradability of polychlorinated dibenzofurans (PCDFs) can also be predicted by the constructed three models from a certain level after adjusting some model parameters, which further proved the versatility of the models. Besides, the sensitivity analysis of the QSAR model and a 3D-QSAR model was developed to investigate the biodegradability mechanisms of PCDDs. Results showed that the descriptors BCUTp-1 h, JGI4, and maxwHBa were the key descriptors in the biodegradability effect by the sensitivity analysis of the QSAR model. Coupled with the results of PCDDs biodegradability 3D-QSAR model, BCUTp-1 h, JGI4, and maxwHBa were confirmed as the main descriptors that affect the biodegradability of dioxins. This study provides a novel theoretical perspective for the research of the biodegradation of both PCDDs and PCDFs dioxins.

Construction of an Early Alert System for Intradialytic Hypotension before Initiating Hemodialysis Based on Machine Learning

Introduction

Intradialytic hypotension (IDH) is prevalent and associated with high hospitalization and mortality rates. The purpose of this study was to explore the risk factors for IDH and use artificial intelligence to establish an early alert system before hemodialysis sessions to identify patients at high risk of IDH.

Materials and Methods

We obtained data on 314,534 hemodialysis sessions conducted at Sichuan Provincial People's Hospital from the renal disease treatment information system. IDH was defined as a systolic blood pressure drop ≥20 mm Hg, a mean arterial pressure drop ≥10 mm Hg during dialysis, or the occurrence of clinical hypotensive events requiring nursing intervention. After pre-processing, the data were randomly divided into training (80%) and testing (20%) sets. Four interpolation methods, three feature selection methods, and 18 machine learning algorithms were used to construct predictive models. The area under the receiver operating characteristic curve (AUC) was the main indicator for evaluating the performance of the models, while Shapley Additive ExPlanation was used to explain the contribution of each variable to the best predictive model.

Results

A total of 3,906 patients and 314,534 dialysis sessions were included, of which 142,237 cases showed IDH (incidence rate, 45.2%). Nineteen parameters were identified through artificial intelligence feature screening. They included age, pre-dialysis weight, dry weight, pre-dialysis blood pressure, heart rate, prescribed ultrafiltration, blood cell counts (neutrophil, lymphocyte, monocyte, eosinophil, lymphocyte, and platelet counts), hematocrit, serum calcium, creatinine, urea, glucose, and uric acid. Random forest, gradient boosting, and logistic regression were the three best models, and the AUCs were 0.812 (95% confidence interval [CI], 0.811-0.813), 0.748 (95% CI, 0.747-0.749), and 0.743 (95% CI, 0.742-0.744), respectively.

Conclusion

Our dialysis software-based artificial intelligence alert system can be used to predict IDH occurrence, enabling the initiation of relevant interventions.

Machine learning-based intradialytic hypotension prediction of patients undergoing hemodialysis: A multicenter retrospective study

BACKGROUND AND OBJECTIVE

Intradialytic hypotension (IDH) is closely associated with adverse clinical outcomes in HD-patients. An IDH predictor model is important for IDH risk screening and clinical decision-making. In this study, we used Machine learning (ML) to develop IDH model for risk prediction in HD patients.

METHODS

62,227 dialysis sessions were randomly partitioned into training data (70%), test data (20%), and validation data (10%). IDH-A model based on twenty-seven variables was constructed for risk prediction for the next HD treatment. IDH-B model based on ten variables from 64,870 dialysis sessions was developed for risk assessment before each HD treatment. Light Gradient Boosting Machine (LightGBM), Linear Discriminant Analysis, support vector machines, XGBoost, TabNet, and multilayer perceptron were used to develop the predictor model.

RESULTS

In IDH-A model, we identified the LightGBM method as the best-performing and interpretable model with C- statistics of 0.82 in Fall30Nadir90 definitions, which was higher than those obtained using the other models (P<0.01). In other IDH standards of Nadir90, Nadir100, Fall20, Fall30, and Fall20Nadir90, the LightGBM method had a performance with C- statistics ranged 0.77 to 0.89. As a complementary application, the LightGBM model in IDH-B model achieved C- statistics of 0.68 in Fall30Nadir90 definitions and 0.69 to 0.78 in the other five IDH standards, which were also higher than the other methods, respectively.

CONCLUSION

Use ML, we identified the LightGBM method as the good-performing and interpretable model. We identified the top variables as the high-risk factors for IDH incident in HD-patient. IDH-A and IDH-B model can usefully complement each other for risk prediction and further facilitate timely intervention through applied into different clinical setting.

Prediction of intradialytic hypotension using pre-dialysis features-a deep learning-based artificial intelligence model

BACKGROUND

Intradialytic hypotension (IDH) is a serious complication of hemodialysis (HD) that is associated with increased risks of cardiovascular morbidity and mortality. However, its accurate prediction remains a clinical challenge. The aim of this study was to develop a deep learning-based artificial intelligence (AI) model to predict IDH using pre-dialysis features.

METHODS

Data from 2007 patients with 943 220 HD sessions at seven university hospitals were used. The performance of the deep learning model was compared with three machine learning models (logistic regression, random forest and XGBoost).

RESULTS

IDH occurred in 5.39% of all studied HD sessions. A lower pre-dialysis blood pressure (BP), and a higher ultrafiltration (UF) target rate and interdialytic weight gain in IDH sessions compared with non-IDH sessions, and the occurrence of IDH in previous sessions was more frequent among IDH sessions compared with non-IDH sessions. Matthews correlation coefficient and macro-averaged F1 score were used to evaluate both positive and negative prediction performances. Both values were similar in logistic regression, random forest, XGBoost and deep learning models, developed with data from a single session. When combining data from the previous three sessions, the prediction performance of the deep learning model improved and became superior to that of other models. The common top-ranked features for IDH prediction were mean systolic BP (SBP) during the previous session, UF target rate, pre-dialysis SBP, and IDH experience during the previous session.

CONCLUSIONS

Our AI model predicts IDH accurately, suggesting it as a reliable tool for HD treatment.

Regulation of Skeletal Muscle Resistance Arteriolar Tone: Integration of Multiple Mechanisms

INTRODUCTION

Physiological system complexity represents an imposing challenge to gaining insight into how arteriolar behavior emerges. Further, mechanistic complexity in arteriolar tone regulation requires that a systematic determination of how these processes interact to alter vascular diameter be undertaken.

METHODS

The present study evaluated the reactivity of ex vivo proximal and in situ distal resistance arterioles in skeletal muscle with challenges across the full range of multiple physiologically relevant stimuli and determined the stability of responses over progressive alterations to each other parameter. The five parameters chosen for examination were (1) metabolism (adenosine concentration), (2) adrenergic activation (norepinephrine concentration), (3) myogenic activation (intravascular pressure), (4) oxygen (superfusate PO2), and (5) wall shear rate (altered intraluminal flow). Vasomotor tone of both arteriole groups following challenge with individual parameters was determined; subsequently, responses were determined following all two- and three-parameter combinations to gain deeper insight into how stimuli integrate to change arteriolar tone. A hierarchical ranking of stimulus significance for establishing arteriolar tone was performed using mathematical and statistical analyses in conjunction with machine learning methods.

RESULTS

Results were consistent across methods and indicated that metabolic and adrenergic influences were most robust and stable across all conditions. While the other parameters individually impact arteriolar tone, their impact can be readily overridden by the two dominant contributors.

CONCLUSION

These data suggest that mechanisms regulating arteriolar tone are strongly affected by acute changes to the local environment and that ongoing investigation into how microvessels integrate stimuli regulating tone will provide a more thorough understanding of arteriolar behavior emergence across physiological and pathological states.

Human Endocrine-Disrupting Effects of Phthalate Esters through Adverse Outcome Pathways: A Comprehensive Mechanism Analysis

Phthalate esters (PAEs) are widely exposed in the environment as plasticizers in plastics, and they have been found to cause significant environmental and health hazards, especially in terms of endocrine disruption in humans. In order to investigate the processes underlying the endocrine disruption effects of PAEs, three machine learning techniques were used in this study to build an adverse outcome pathway (AOP) for those effects on people. According to the results of the three machine learning techniques, the random forest and XGBoost models performed well in terms of prediction. Subsequently, sensitivity analysis was conducted to identify the initial events, key events, and key features influencing the endocrine disruption effects of PAEs on humans. Key features, such as Mol.Wt, Q+, QH+, ELUMO, minHCsats, MEDC-33, and EG, were found to be closely related to the molecular structure. Therefore, a 3D-QSAR model for PAEs was constructed, and, based on the three-dimensional potential energy surface information, it was discovered that the hydrophobic, steric, and electrostatic fields of PAEs significantly influence their endocrine disruption effects on humans. Lastly, an analysis of the contributions of amino acid residues and binding energy (BE) was performed, identifying and confirming that hydrogen bonding, hydrophobic interactions, and van der Waals forces are important factors affecting the AOP of PAEs' molecular endocrine disruption effects. This study defined and constructed a comprehensive AOP for the endocrine disruption effects of PAEs on humans and developed a method based on theoretical simulation to characterize the AOP, providing theoretical guidance for studying the mechanisms of toxicity caused by other pollutants.

Establishment of a scoring model for the differential diagnosis of white coat hypertension and sustained hypertension

OBJECTIVES

This study aimed to establish a scoring model for the differential diagnosis of white coat hypertension (WCH) and sustained hypertension (SHT).

METHODS

This study comprised 553 adults with elevated office blood pressure, normal renal function, and no antihypertensive medications. Through questionnaire investigation and biochemical detection, 17 parameters, such as gender and age, were acquired. WCH and SHT were distinguished by 24 h ambulatory blood pressure monitoring. The participants were randomly divided into a training set (445 cases) and a validation set (108 cases). The above parameters were screened using least absolute shrinkage and selection operator regression and univariate logistic regression analysis in the training set. Afterward, a scoring model was constructed through multivariate logistic regression analysis.

RESULTS

Finally, six parameters were selected, including isolated systolic hypertension, office systolic blood pressure, office diastolic blood pressure, triglyceride, serum creatinine, and cardiovascular and cerebrovascular diseases. Multivariate logistic regression was used to establish a scoring model. The R2 and area under the ROC curve (AUC) of the scoring model in the training set were 0.163 and 0.705, respectively. In the validation set, the R2 of the scoring model was 0.206, and AUC was 0.718. The calibration test results revealed that the scoring model had good stability in both the training and validation sets (mean square error = 0.001, mean absolute error = 0.014; mean square error = 0.001, mean absolute error = 0.025).

CONCLUSION

A stable scoring model for distinguishing WCH was established, which can assist clinicians in identifying WCH at the first diagnosis.

Predicting dry weight change in Hemodialysis patients using machine learning

BACKGROUND

Machine Learning has been increasingly used in the medical field, including managing patients undergoing hemodialysis. The random forest classifier is a Machine Learning method that can generate high accuracy and interpretability in the data analysis of various diseases. We attempted to apply Machine Learning to adjust dry weight, the appropriate volume status of patients undergoing hemodialysis, which requires a complex decision-making process considering multiple indicators and the patient's physical conditions.

METHODS

All medical data and 69,375 dialysis records of 314 Asian patients undergoing hemodialysis at a single dialysis center in Japan between July 2018 and April 2020 were collected from the electronic medical record system. Using the random forest classifier, we developed models to predict the probabilities of adjusting the dry weight at each dialysis session.

RESULTS

The areas under the receiver-operating-characteristic curves of the models for adjusting the dry weight upward and downward were 0.70 and 0.74, respectively. The average probability of upward adjustment of the dry weight had sharp a peak around the actual change over time, while the average probability of downward adjustment of the dry weight formed a gradual peak. Feature importance analysis revealed that median blood pressure decline was a strong predictor for adjusting the dry weight upward. In contrast, elevated serum levels of C-reactive protein and hypoalbuminemia were important indicators for adjusting the dry weight downward.

CONCLUSIONS

The random forest classifier should provide a helpful guide to predict the optimal changes to the dry weight with relative accuracy and may be useful in clinical practice.

Efficacy of endoscopic therapy for T1b esophageal cancer and construction of prognosis prediction model: a retrospective cohort study

BACKGROUND

The efficacy of endoscopic therapy on the long-term survival outcomes of T1b oesophageal cancer (EC) is unclear, this study was designed to clarify the survival outcomes of endoscopic therapy and to construct a model for predicting the prognosis in T1b EC patients.

METHODS

This study was performed using the Surveillance, Epidemiology, and End Results (SEER) database from 2004 to 2017 of patients with T1bN0M0 EC. Cancer-specific survival (CSS) and overall survival (OS) were compared between endoscopic therapy group, esophagectomy group and chemoradiotherapy group, respectively. Stabilized inverse probability treatment weighting was used as the main analysis method. The propensity score matching method and an independent dataset from our hospital were used as sensitivity analysis. The least absolute shrinkage and selection operator regression (Lasso) was employed to sift variables. A prognostic model was then established and was verified in two external validation cohorts.

RESULTS

The unadjusted 5-year CSS was 69.5% (95% CI, 61.5-77.5) for endoscopic therapy, 75.0% (95% CI, 71.5-78.5) for esophagectomy and 42.4% (95% CI, 31.0-53.8) for chemoradiotherapy. After stabilized inverse probability treatment weighting adjustment, CSS and OS were similar in endoscopic therapy and esophagectomy groups ( P =0.32, P =0.83), while the CSS and OS of chemoradiotherapy patients were inferior to endoscopic therapy patients ( P <0.01, P <0.01). Age, histology, grade, tumour size, and treatment were selected to build the prediction model. The area under the curve of receiver operating characteristics of 1, 3, and 5 years in the validation cohort 1 were 0.631, 0.618, 0.638, and 0.733, 0.683, 0.768 in the validation cohort 2. The calibration plots also demonstrated the consistency of predicted and actual values in the two external validation cohorts.

CONCLUSION

Endoscopic therapy achieved comparable long-term survival outcomes to esophagectomy for T1b EC patients. The prediction model developed performed well in calculating the OS of patients with T1b EC.

Monitoring Knee Contact Force with Force-Sensing Insoles

Numerous applications exist for monitoring knee contact force (KCF) throughout activities of daily living. However, the ability to estimate these forces is restricted to a laboratory setting. The purposes of this study are to develop KCF metric estimation models and explore the feasibility of monitoring KCF metrics via surrogate measures derived from force-sensing insole data. Nine healthy subjects (3F, age 27 ± 5 years, mass 74.8 ± 11.8 kg, height 1.7 ± 0.08 m) walked at multiple speeds (0.8-1.6 m/s) on an instrumented treadmill. Thirteen insole force features were calculated as potential predictors of peak KCF and KCF impulse per step, estimated with musculoskeletal modeling. The error was calculated with median symmetric accuracy. Pearson product-moment correlation coefficients defined the relationship between variables. Models develop per-limb demonstrated lower prediction error than those developed per-subject (KCF impulse: 2.2% vs 3.4%; peak KCF: 3.50% vs. 6.5%, respectively). Many insole features are moderately to strongly associated with peak KCF, but not KCF impulse across the group. We present methods to directly estimate and monitor changes in KCF using instrumented insoles. Our results carry promising implications for internal tissue loads monitoring outside of a laboratory with wearable sensors.

A Machine Learning Approach for Walking Classification in Elderly People with Gait Disorders

Walking ability of elderly individuals, who suffer from walking difficulties, is limited, which restricts their mobility independence. The physical health and well-being of the elderly population are affected by their level of physical activity. Therefore, monitoring daily activities can help improve the quality of life. This becomes especially a huge challenge for those, who suffer from dementia and Alzheimer's disease. Thus, it is of great importance for personnel in care homes/rehabilitation centers to monitor their daily activities and progress. Unlike normal subjects, it is required to place the sensor on the back of this group of patients, which makes it even more challenging to detect walking from other activities. With the latest advancements in the field of health sensing and sensor technology, a huge amount of accelerometer data can be easily collected. In this study, a Machine Learning (ML) based algorithm was developed to analyze the accelerometer data collected from patients with walking difficulties, who live in one of the municipalities in Denmark. The ML algorithm is capable of accurately classifying the walking activity of these individuals with different walking abnormalities. Various statistical, temporal, and spectral features were extracted from the time series data collected using an accelerometer sensor placed on the back of the participants. The back sensor placement is desirable in patients with dementia and Alzheimer's disease since they may remove visible sensors to them due to the nature of their diseases. Then, an evolutionary optimization algorithm called Particle Swarm Optimization (PSO) was used to select a subset of features to be used in the classification step. Four different ML classifiers such as k-Nearest Neighbors (kNN), Random Forest (RF), Stacking Classifier (Stack), and Extreme Gradient Boosting (XGB) were trained and compared on an accelerometry dataset consisting of 20 participants. These models were evaluated using the leave-one-group-out cross-validation (LOGO-CV) technique. The Stack model achieved the best performance with average sensitivity, positive predictive values (precision), F₁-score, and accuracy of 86.85%, 93.25%, 88.81%, and 93.32%, respectively, to classify walking episodes. In general, the empirical results confirmed that the proposed models are capable of classifying the walking episodes despite the challenging sensor placement on the back of the patients, who suffer from walking disabilities.

Model for Predicting Complications of Hemodialysis Patients Using Data From the Internet of Medical Things and Electronic Medical Records

Intelligent models for predicting hemodialysis-related complications, i.e., hypotension and the deterioration of the quality or obstruction of the AV fistula, based on machine learning (ML) methods were established to offer early warnings to medical staff and give them enough time to provide pre-emptive treatment. A novel integration platform collected data from the Internet of Medical Things (IoMT) at a dialysis center and inspection results from electronic medical records (EMR) to train ML algorithms and build models. The selection of the feature parameters was implemented using Pearson's correlation method. Then, the eXtreme Gradient Boost (XGBoost) algorithm was chosen to create the predictive models and optimize the feature choice. 75% of collected data are used as a training dataset and the other 25% are used as a testing dataset. We adopted the prediction precision and recall rate of hypotension and AV fistula obstruction to measure the effectiveness of the predictive models. These rates were sufficiently high at approximately 71%-90%. In the context of hemodialysis, hypotension and the deterioration of the quality or obstruction of the arteriovenous (AV) fistula affect treatment quality and patient safety and may lead to a poor prognosis. Our prediction models with high accuracies can provide excellent references and signals for clinical healthcare service providers. Clinical and Translational Impact Statement-With the integrated dataset collected from IoMT and EMR, the superior predictive results of our models for complications of hemodialysis patients are demonstrated. We believe, after enough clinical tests are implemented as planned, these models can assist the healthcare team in making appropriate preparations in advance or adjusting the medical procedures to avoid these adverseevents.

Latent trajectories of frailty and risk prediction models among geriatric community dwellers: an interpretable machine learning perspective

BACKGROUND

This study aimed to identify long-term frailty trajectories among older adults (≥65) and construct interpretable prediction models to assess the risk of developing abnormal frailty trajectory among older adults and examine significant factors related to the progression of frailty.

METHODS

This study retrospectively collected data from the Chinese Longitudinal Healthy Longevity and Happy Family Study between 2002 and 2018 (N = 4083). Frailty was defined by the frailty index. The whole study consisted of two phases of tasks. First, group-based trajectory modeling was used to identify frailty trajectories. Second, easy-to-access epidemiological data was utilized to construct machine learning algorithms including naïve bayes, logistic regression, decision tree, support vector machine, random forest, artificial neural network, and extreme gradient boosting to predict the risk of long-term frailty trajectories. Further, Shapley additive explanations was employed to identify feature importance and open-up the black box model of machine learning to further strengthen decision makers' trust in the model.

RESULTS

Two distinct frailty trajectories (stable-growth: 82.54%, rapid-growth: 17.46%) were identified. Compared with other algorithms, random forest performed relatively better in distinguishing the stable-growth and rapid-growth groups. Physical function including activities of daily living and instrumental activities of daily living, marital status, weight, and cognitive function were top five predictors.

CONCLUSIONS

Interpretable machine learning can achieve the primary goal of risk stratification and make it more transparent in individual prediction beneficial to primary screening and tailored prevention.

Implementation of ensemble machine learning algorithms on exome datasets for predicting early diagnosis of cancers

Classification of different cancer types is an essential step in designing a decision support model for early cancer predictions. Using various machine learning (ML) techniques with ensemble learning is one such method used for classifications. In the present study, various ML algorithms were explored on twenty exome datasets, belonging to 5 cancer types. Initially, a data clean-up was carried out on 4181 variants of cancer with 88 features, and a derivative dataset was obtained using natural language processing and probabilistic distribution. An exploratory dataset analysis using principal component analysis was then performed in 1 and 2D axes to reduce the high-dimensionality of the data. To significantly reduce the imbalance in the derivative dataset, oversampling was carried out using SMOTE. Further, classification algorithms such as K-nearest neighbour and support vector machine were used initially on the oversampled dataset. A 4-layer artificial neural network model with 1D batch normalization was also designed to improve the model accuracy. Ensemble ML techniques such as bagging along with using KNN, SVM and MLPs as base classifiers to improve the weighted average performance metrics of the model. However, due to small sample size, model improvement was challenging. Therefore, a novel method to augment the sample size using generative adversarial network (GAN) and triplet based variational auto encoder (TVAE) was employed that reconstructed the features and labels generating the data. The results showed that from initial scrutiny, KNN showed a weighted average of 0.74 and SVM 0.76. Oversampling ensured that the accuracy of the derivative dataset improved significantly and the ensemble classifier augmented the accuracy to 82.91%, when the data was divided into 70:15:15 ratio (training, test and holdout datasets). The overall evaluation metric value when GAN and TVAE increased the sample size was found to be 0.92 with an overall comparison model of 0.66. Therefore, the present study designed an effective model for classifying cancers which when implemented to real world samples, will play a major role in early cancer diagnosis.

Cuffless blood pressure estimation using chaotic features of photoplethysmograms and parallel convolutional neural network

BACKGROUND AND OBJECTIVE

As a nonlinear framework in dynamical system analysis, chaotic approaches are mainly applied to evolve the complexity of biological systems. Due to the chaotic nature of the cardiovascular systems, the nonlinear features can intuitively provide a reliable framework in blood pressure (BP) estimation. Cuffless BP estimation is usually carried out by establishing deep neural network models estimating the BP values through machine-learned features of photoplethysmogram (PPG) signals.

METHODS

In this study, a novel parallel deep architecture is proposed to handle the machine-learned and chaotic features of PPG signals in estimating the actual BP values. The chaotic handcrafted features were the signal properties associated with the Poincare sections in the phase space and the recurrence plot-based measures called recurrence quantification analysis (RQA). Moreover, the measures quantifying the nonlinear properties of the temporal sequences such as correlation dimension, fractal dimension, Lyapunov exponent, and entropy-based quantities were also employed. The parallel architecture not only embedded the chaotic nature of PPG signals but also provided a facility to include the pseudo-periodic variations of PPGs by utilizing a concatenating layer.

RESULTS

Our framework was examined on the public dataset, namely, Multi-Parameter Intelligent in Intensive Care II contained the recording of PPG, ECG and arterial blood pressure. The performance of the employed handcrafted features in distinguishing between the levels of BP values was investigated based on Spearman's statistics. In addition, our proposed scheme is evaluated in terms of Pearson's correlation coefficient (R), root mean square error (RMSE), and mean absolute error (MAE). The best performance was achieved when the employed handcrafted features accompanied by PPG sequences were applied to the parallel deep network. In particular, the values of R, RMSE, and MAE were obtained 0.9529, 2.76 mmHg, and 1.73 mmHg for diastolic BP, and 0.9444, 6.18 mmHg, and 3.8 mmHg for systolic BP, respectively. Moreover, based on the requirements of the standards set by the British Hypertension Society (BHS), the proposed scheme achieved a grade of A.

CONCLUSIONS

Our proposed scheme outperformed the state-of-the-art BP estimation methods. In addition, the results confirmed that the concatenation of the PPG-related machine-learned and nonlinear handcrafted features can be properly applied in continuous BP monitoring.

Intradialytic hypotension prediction using covariance matrix-driven whale optimizer with orthogonal structure-assisted extreme learning machine

Intradialytic hypotension (IDH) is an adverse event occurred during hemodialysis (HD) sessions with high morbidity and mortality. The key to preventing IDH is predicting its pre-dialysis and administering a proper ultrafiltration prescription. For this purpose, this paper builds a prediction model (bCOWOA-KELM) to predict IDH using indices of blood routine tests. In the study, the orthogonal learning mechanism is applied to the first half of the WOA to improve the search speed and accuracy. The covariance matrix is applied to the second half of the WOA to enhance the ability to get out of local optimum and convergence accuracy. Combining the above two improvement methods, this paper proposes a novel improvement variant (COWOA) for the first time. More, the core of bCOWOA-KELM is that the binary COWOA is utilized to improve the performance of the KELM. In order to verify the comprehensive performance of the study, the paper sets four types of comparison experiments for COWOA based on 30 benchmark functions and a series of prediction experiments for bCOWOA-KELM based on six public datasets and the HD dataset. Finally, the results of the experiments are analyzed separately in this paper. The results of the comparison experiments prove fully that the COWOA is superior to other famous methods. More importantly, the bCOWOA performs better than its peers in feature selection and its accuracy is 92.41%. In addition, bCOWOA improves the accuracy by 0.32% over the second-ranked bSCA and by 3.63% over the worst-ranked bGWO. Therefore, the proposed model can be used for IDH prediction with future applications.

Real-Time Cuffless Continuous Blood Pressure Estimation Using 1D Squeeze U-Net Model: A Progress toward mHealth

Measuring continuous blood pressure (BP) in real time by using a mobile health (mHealth) application would open a new door in the advancement of the healthcare system. This study aimed to propose a real-time method and system for measuring BP without using a cuff from a digital artery. An energy-efficient real-time smartphone-application-friendly one-dimensional (1D) Squeeze U-net model is proposed to estimate systolic and diastolic BP values, using only raw photoplethysmogram (PPG) signal. The proposed real-time cuffless BP prediction method was assessed for accuracy, reliability, and potential usefulness in the hypertensive assessment of 100 individuals in two publicly available datasets: Multiparameter Intelligent Monitoring in Intensive Care (MIMIC-I) and Medical Information Mart for Intensive Care (MIMIC-III) waveform database. The proposed model was used to build an android application to measure BP at home. This proposed deep-learning model performs best in terms of systolic BP, diastolic BP, and mean arterial pressure, with a mean absolute error of 4.42, 2.25, and 2.56 mmHg and standard deviation of 4.78, 2.98, and 3.21 mmHg, respectively. The results meet the grade A performance requirements of the British Hypertension Society and satisfy the AAMI error range. The result suggests that only using a short-time PPG signal is sufficient to obtain accurate BP measurements in real time. It is a novel approach for real-time cuffless BP estimation by implementing an mHealth application and can measure BP at home and assess hypertension.

Machine Learning Improves the Prediction Rate of Non-Curative Resection of Endoscopic Submucosal Dissection in Patients with Early Gastric Cancer

Non-curative resection (NCR) of early gastric cancer (EGC) after endoscopic submucosal dissection (ESD) can increase the burden of additional treatment and medical expenses. We aimed to develop a machine-learning (ML)-based NCR prediction model for EGC prior to ESD. We obtained data from 4927 patients with EGC who underwent ESD between January 2006 and February 2020. Ten clinicopathological characteristics were selected using extreme gradient boosting (XGBoost) and were used to develop a ML-based model. Dataset was divided into the training and internal validation sets and verified using an external validation set. Sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) were evaluated. The performance of each model was compared by using the Delong test. A total of 1100 (22.1%) patients were identified as being treated non-curatively with ESD. Seven ML-based NCR prediction models were developed. The performance of NCR prediction was highest in the XGBoost model (AUROC, 0.851; 95% confidence interval, 0.837–0.864). When we compared the prediction performance by the Delong test, XGBoost (p = 0.02) and support vector machine (p = 0.02) models showed a significantly higher performance among the NCR prediction models. We developed an ML model capable of accurately predicting the NCR of EGC before ESD. This ML model can provide useful information for decision-making regarding the appropriate treatment of EGC before ESD.

Modeling of energy consumption factors for an industrial cement vertical roller mill by SHAP-XGBoost: a "conscious lab" approach

Cement production is one of the most energy-intensive manufacturing industries, and the milling circuit of cement plants consumes around 4% of a year's global electrical energy production. It is well understood that modeling and digitalizing industrial-scale processes would help control production circuits better, improve efficiency, enhance personal training systems, and decrease plants' energy consumption. This tactical approach could be integrated using conscious lab (CL) as an innovative concept in the internet age. Surprisingly, no CL has been reported for the milling circuit of a cement plant. A robust CL interconnect datasets originated from monitoring operational variables in the plants and translating them to human basis information using explainable artificial intelligence (EAI) models. By initiating a CL for an industrial cement vertical roller mill (VRM), this study conducted a novel strategy to explore relationships between VRM monitored operational variables and their representative energy consumption factors (output temperature and motor power). Using SHapley Additive exPlanations (SHAP) as one of the most recent EAI models accurately helped fill the lack of information about correlations within VRM variables. SHAP analyses highlighted that working pressure and input gas rate with positive relationships are the key factors influencing energy consumption. eXtreme Gradient Boosting (XGBoost) as a powerful predictive tool could accurately model energy representative factors by R-square ever 0.80 in the testing phase. Comparison assessments indicated that SHAP-XGBoost could provide higher accuracy for VRM-CL structure than conventional modeling tools (Pearson correlation, Random Forest, and Support vector regression.

Concatenated convolutional neural network model for cuffless blood pressure estimation using fuzzy recurrence properties of photoplethysmogram signals

Due to the importance of continuous monitoring of blood pressure (BP) in controlling hypertension, the topic of cuffless BP estimation has been widely studied in recent years. A most important approach is to explore the nonlinear mapping between the recorded peripheral signals and the BP values which is usually conducted by deep neural networks. Because of the sequence-based pseudo periodic nature of peripheral signals such as photoplethysmogram (PPG), a proper estimation model needed to be equipped with the 1-dimensional (1-D) and recurrent layers. This, in turn, limits the usage of 2-dimensional (2-D) layers adopted in convolutional neural networks (CNN) for embedding spatial information in the model. In this study, considering the advantage of chaotic approaches, the recurrence characterization of peripheral signals was taken into account by a visual 2-D representation of PPG in phase space through fuzzy recurrence plot (FRP). FRP not only provides a beneficial framework for capturing the spatial properties of input signals but also creates a reliable approach for embedding the pseudo periodic properties to the neural models without using recurrent layers. Moreover, this study proposes a novel deep neural network architecture that combines the morphological features extracted simultaneously from two upgraded 1-D and 2-D CNNs capturing the temporal and spatial dependencies of PPGs in systolic and diastolic BP estimation. The model has been fed with the 1-D PPG sequences and the corresponding 2-D FRPs from two separate routes. The performance of the proposed framework was examined on the well-known public dataset, namely, multi-parameter intelligent in Intensive Care II. Our scheme is analyzed and compared with the literature in terms of the requirements of the standards set by the British Hypertension Society (BHS) and the Association for the Advancement of Medical Instrumentation (AAMI). The proposed model met the AAMI requirements, and it achieved a grade of A as stated by the BHS standard. In addition, its mean absolute errors and standard deviation for both systolic and diastolic blood pressure estimations were considerably low, 3.05 ± 5.26 mmHg and 1.58 ± 2.6 mmHg, in turn.

Integrated Machine Learning and Bioinformatic Analyses Constructed a Novel Stemness-Related Classifier to Predict Prognosis and Immunotherapy Responses for Hepatocellular Carcinoma Patients

Immunotherapy has made great progress in hepatocellular carcinoma (HCC), yet there is still a lack of biomarkers for predicting response to it. Cancer stem cells (CSCs) are the primary cause of the tumorigenesis, metastasis, and multi-drug resistance of HCC. This study aimed to propose a novel CSCs-related cluster of HCC to predict patients' response to immunotherapy. Based on RNA-seq datasets from The Cancer Genome Atlas (TCGA) and Progenitor Cell Biology Consortium (PCBC), one-class logistic regression (OCLR) algorithm was applied to compute the stemness index (mRNAsi) of HCC patients. Unsupervised consensus clustering was performed to categorize HCC patients into two stemness subtypes which further proved to be a predictor of tumor immune microenvironment (TIME) status, immunogenomic expressions and sensitivity to neoadjuvant therapies. Finally, four machine learning algorithms (LASSO, RF, SVM-RFE and XGboost) were applied to distinguish different stemness subtypes. Thus, a five-hub-gene based classifier was constructed in TCGA and ICGC HCC datasets to predict patients' stemness subtype in a more convenient and applicable way, and this novel stemness-based classification system could facilitate the prognostic prediction and guide clinical strategies of immunotherapy and targeted therapy in HCC.

Development and validation of a machine learning algorithm-based risk prediction model of pressure injury in the intensive care unit

The study aimed to establish a machine learning-based scoring nomogram for early recognition of likely pressure injuries in an intensive care unit (ICU) using large-scale clinical data. A retrospective cohort study design was employed to develop and validate a top-performing clinical feature panel accessibly in the electronic medical records (EMRs), which was in the mode of a quantifiable nomogram. Clinical factors regarding demographics, admission cause, clinical laboratory index, medical history and nursing scales were extracted as risk candidates. The performance improvement was based on the application of the machine learning technique, comprising logistic regression, decision tree and random forest algorithm with five-fold cross-validation (CV) technique. The comprehensive assessment of sensitivity, specificity and the area under the receiver operating characteristic curve (AUROC) was considered in the evaluation of predictive performance. The receiver operating characteristic curves revealed the top performance for the logistic regression model in respect to machine learning improvement, achieving the highest sensitivity and AUC among three types of classifiers. Compared against the 23-point Braden scale routinely recorded online, an incorporated nomogram of logistic regression model and Braden scale achieved the best performance with an AUC of 0.87 ± 0.07 and 0.84 ± 0.05 in training and test cohort, respectively. Our findings suggest that the machine learning technique potentiated the limited predictive validity of routinely recorded clinical data on pressure injury development during ICU hospitalisation. Easily accessible electronic records held the potentials to substitute the traditional Braden score in the prediction of pressure injury in intensive care unit. Preoperative prediction of pressure injury facilitates the exemption from the severe consequences.

Long short-term memory stacking model to predict the number of cases and deaths caused by COVID-19

The long short-term memory (LSTM) is a high-efficiency model for forecasting time series, for being able to deal with a large volume of data from a time series with nonlinearities. As a case study, the stacked LSTM will be used to forecast the growth of the pandemic of COVID-19, based on the increase in the number of contaminated and deaths in the State of Santa Catarina, Brazil. COVID-19 has been spreading very quickly, causing great concern in relation to the ability to care for critically ill patients. Control measures are being imposed by governments with the aim of reducing the contamination and the spreading of viruses. The forecast of the number of contaminated and deaths caused by COVID-19 can help decision making regarding the adopted restrictions, making them more or less rigid depending on the pandemic’s control capacity. The use of LSTM stacking shows an R2 of 0.9625 for confirmed cases and 0.9656 for confirmed deaths caused by COVID-19, being superior to the combinations among other evaluated models.

Accuracy versus reliability-based modelling approaches for medical decision making

Forecasting in the medical domain is critical to the quality of decisions made by physicians, patients, and health planners. Modeling is one of the most important components of decision support systems, which are frequently used to simulate and analyze under-studied systems in order to make more appropriate decisions in medical science. In the medical modeling literature, various approaches with varying structures and characteristics have been proposed to cover a wide range of application categories and domains. Regardless of the differences between modeling approaches, all of them aim to maximize the accuracy or reliability of the results in order to achieve the most generalizable model and, as a result, a higher level of profitability decisions. Despite the theoretical significance and practical impact of reliability on generalizability, particularly in high-risk decisions and applications, a significant number of models in the fields of medical forecasting, classification, and time series prediction have been developed to maximize accuracy in mind. In other words, given the volatility of medical variables, it is also necessary to have stable and reliable forecasts in order to make sound decisions. The quality of medical decisions resulting from accuracy and reliability-based intelligent and statistical modeling approaches is compared and evaluated in this paper in order to determine the relative importance of accuracy and reliability on the quality of made decisions in decision support systems. For this purpose, 33 different case studies from the UCI in three categories of supervised modeling, namely causal forecasting, time series prediction, and classification, were considered. These cases were chosen from various domains, such as disease diagnosis (obesity, Parkinson's disease, diabetes, hepatitis, stenosis of arteries, orthopedic disease, autism) and cancer (lung, breast, cervical), experiments, therapy (immunotherapy, cryotherapy), fertility prediction, and predicting the number of patients in the emergency room and ICU. According to empirical findings, the reliability-based strategy outperformed the accuracy-based strategy in causal forecasting cases by 2.26%, classification cases by 13.49%, and time series prediction cases by 3.08%. Furthermore, compared to similar accuracy-based models, the reliability-based models can generate a 6.28% improvement. As a result, they can be considered an appropriate alternative to traditional accuracy-based models for medical decision support systems modeling purposes.

Machine Learning Prediction Models for Mortality in Intensive Care Unit Patients with Lactic Acidosis

Background: Lactic acidosis is the most common cause of anion gap metabolic acidosis in the intensive care unit (ICU), associated with poor outcomes including mortality. We sought to compare machine learning (ML) approaches versus logistic regression analysis for prediction of mortality in lactic acidosis patients admitted to the ICU. Methods: We used the Medical Information Mart for Intensive Care (MIMIC-III) database to identify ICU adult patients with lactic acidosis (serum lactate ≥4 mmol/L). The outcome of interest was hospital mortality. We developed prediction models using four ML approaches consisting of random forest (RF), decision tree (DT), extreme gradient boosting (XGBoost), artificial neural network (ANN), and statistical modeling with forward stepwise logistic regression using the testing dataset. We then assessed model performance using area under the receiver operating characteristic curve (AUROC), accuracy, precision, error rate, Matthews correlation coefficient (MCC), F1 score, and assessed model calibration using the Brier score, in the independent testing dataset. Results: Of 1919 lactic acidosis ICU patients, 1535 and 384 were included in the training and testing dataset, respectively. Hospital mortality was 30%. RF had the highest AUROC at 0.83, followed by logistic regression 0.81, XGBoost 0.81, ANN 0.79, and DT 0.71. In addition, RF also had the highest accuracy (0.79), MCC (0.45), F1 score (0.56), and lowest error rate (21.4%). The RF model was the most well-calibrated. The Brier score for RF, DT, XGBoost, ANN, and multivariable logistic regression was 0.15, 0.19, 0.18, 0.19, and 0.16, respectively. The RF model outperformed multivariable logistic regression model, SOFA score (AUROC 0.74), SAP II score (AUROC 0.77), and Charlson score (AUROC 0.69). Conclusion: The ML prediction model using RF algorithm provided the highest predictive performance for hospital mortality among ICU patient with lactic acidosis.

Remaining Useful Life Prediction of the Concrete Piston Based on Probability Statistics and Data Driven

This paper proposes a method on predicting the remaining useful life (RUL) of a concrete piston of a concrete pump truck based on probability statistics and data-driven approaches. Firstly, the average useful life of the concrete piston is determined by probability distribution fitting using actual life data. Secondly, according to condition monitoring data of the concrete pump truck, a concept of life coefficient of the concrete piston is proposed to represent the influence of the loading condition on the actual useful life of individual concrete pistons, and different regression models are established to predict the RUL of the concrete pistons. Finally, according to the prediction result of the concrete piston at different life stages, a replacement warning point is established to provide support for the inventory management and replacement plan of the concrete piston.

Erythropoiesis stimulating agent recommendation model using recurrent neural networks for patient with kidney failure with replacement therapy

In patients with kidney failure with replacement therapy (KFRT), optimizing anemia management in these patients is a challenging problem because of the complexities of the underlying diseases and heterogeneous responses to erythropoiesis-stimulating agents (ESAs). Therefore, we propose a ESA dose recommendation model based on sequential awareness neural networks. Data from 466 KFRT patients (12,907 dialysis sessions) in seven tertiary-care general hospitals were included in the experiment. First, a Hb prediction model was developed to simulate longitudinal heterogeneous ESA and Hb interactions. Based on the prediction model as a prospective study simulator, we built an ESA dose recommendation model to predict the required amount of ESA dose to reach a target hemoglobin level after 30 days. Each model's performance was evaluated in the mean absolute error (MAE). The MAEs presenting the best results of the prediction and recommendation model were 0.59 (95% confidence interval: 0.56-0.62) g/dL and 43.2 μg (ESAs dose), respectively. Compared to the results in the real-world clinical data, the recommendation model achieved a reduction of ESA dose (Algorithm: 140 vs. Human: 150 μg/month, P < 0.001), a more stable monthly Hb difference (Algorithm: 0.6 vs. Human: 0.8 g/dL, P < 0.001), and an improved target Hb success rate (Algorithm: 79.5% vs. Human: 62.9% for previous month's Hb < 10.0 g/dL; Algorithm: 95.7% vs. Human:73.0% for previous month's Hb 10.0-12.0 g/dL). We developed an ESA dose recommendation model for optimizing anemia management in patients with KFRT and showed its potential effectiveness in a simulated prospective study.

Machine learning-based prediction of acute kidney injury after nephrectomy in patients with renal cell carcinoma

The precise prediction of acute kidney injury (AKI) after nephrectomy for renal cell carcinoma (RCC) is an important issue because of its relationship with subsequent kidney dysfunction and high mortality. Herein we addressed whether machine learning (ML) algorithms could predict postoperative AKI risk better than conventional logistic regression (LR) models. A total of 4104 RCC patients who had undergone unilateral nephrectomy from January 2003 to December 2017 were reviewed. ML models such as support vector machine, random forest, extreme gradient boosting, and light gradient boosting machine (LightGBM) were developed, and their performance based on the area under the receiver operating characteristic curve, accuracy, and F1 score was compared with that of the LR-based scoring model. Postoperative AKI developed in 1167 patients (28.4%). All the ML models had higher performance index values than the LR-based scoring model. Among them, the LightGBM model had the highest value of 0.810 (0.783-0.837). The decision curve analysis demonstrated a greater net benefit of the ML models than the LR-based scoring model over all the ranges of threshold probabilities. The application of ML algorithms improves the predictability of AKI after nephrectomy for RCC, and these models perform better than conventional LR-based models.

Prediction of overall survival in patients with Stage I esophageal cancer: A novel web-based calculator

BACKGROUND AND AIMS

In this study, we aimed to develop a convenient web-based calculator to predict the overall survival (OS) of patients with Stage I esophageal cancer (EC).

METHODS

Data of 1664 patients, between 2004 and 2015, were extracted from the Surveillance, Epidemiology, and End Results database. Least absolute shrinkage and selection operator regression was employed to sift variables; subsequently, Cox proportional hazards regression model was built. We applied the enhanced bootstrap validation to appraise the discrimination and calibration of the model. Clinical benefit was measured using decision curve analysis (DCA). Thereafter, a web-based calculator based on the model, which could be used to predict the 1-, 3-, and 5-year OS rates, was developed.

RESULTS

Race, age, histologic type, grade, N stage, and therapeutic methods were selected. C-indices of the prediction model in the training and validation groups were 0.726 (95% confidence interval [CI], 0.679-0.773) and 0.724 (95% CI, 0.679-0.769), respectively. Calibration curves showed good agreement between the groups. The DCA demonstrated that the prediction model is clinically useful.

CONCLUSIONS

The prediction model we developed showed a good performance in calculating the OS rates in patients with Stage I EC. The web-based calculator is available at https://championship.shinyapps.io/dynnomapp/.

Waterlogging Resistance Evaluation Index and Photosynthesis Characteristics Selection: Using Machine Learning Methods to Judge Poplar’s Waterlogging Resistance

Flood disasters are the major natural disaster that affects the growth of agriculture and forestry crops. Due to rapid growth and strong waterlogging resistance characteristics, many studies have explained the waterlogging resistance mechanism of poplar from different perspectives. However, there is no accurate method to define the evaluation index of waterlogging resistance. In addition, there is also a lack of research on predicting the waterlogging resistance of poplars. Based on the changes of poplar biomass and seedling height, the evaluation index of poplar resistance to waterlogging was well determined, and the characteristics of photosynthesis were used to predict the waterlogging resistance of poplars. First, four methods of hierarchical clustering, lasso, stepwise regression and all-subsets regression were used to extract the photosynthesis characteristics. After that, the support vector regression model of poplar resistance to waterlogging was established by using the characteristic parameters of photosynthesis. Finally, the results show that the SVR model based on Stepwise regression and Lasso method has high precision. On the test set, the coefficient of determination (R2) was 0.8581 and 0.8492, the mean square error (MSE) was 0.0104 and 0.0341, and the mean relative error (MRE) was 9.78% and 9.85%, respectively. Therefore, using the characteristic parameters of photosynthesis to predict the waterlogging resistance of poplars is feasible.