Specific patterns and potential risk factors to predict 3-year risk of death among non-cancer patients with advanced chronic kidney disease by machine learning.

Chronic kidney disease (CKD) is a major public health concern. But there are limited machine learning studies on non-cancer patients with advanced CKD, and the results of machine learning studies on cancer patients with CKD may not apply directly on non-cancer patients. We aimed to conduct a comprehensive investigation of risk factors for a 3-year risk of death among non-cancer advanced CKD patients with an estimated glomerular filtration rate < 60.0 mL/min/1.73m2 by several machine learning algorithms. In this retrospective cohort study, we collected data from in-hospital and emergency care patients from 2 hospitals in Taiwan from 2009 to 2019, including their international classification of disease at admission and laboratory data from the hospital's electronic medical records (EMRs). Several machine learning algorithms were used to analyze the potential impact and degree of influence of each factor on mortality and survival. Data from 2 hospitals in northern Taiwan were collected with 6565 enrolled patients. After data cleaning, 26 risk factors and approximately 3887 advanced CKD patients from Shuang Ho Hospital were used as the training set. The validation set contained 2299 patients from Taipei Medical University Hospital. Predictive variables, such as albumin, PT-INR, and age, were the top 3 significant risk factors with paramount influence on mortality prediction. In the receiver operating characteristic curve, the random forest had the highest values for accuracy above 0.80. MLP, and Adaboost had better performance on sensitivity and F1-score compared to other methods. Additionally, SVM with linear kernel function had the highest specificity of 0.9983, while its sensitivity and F1-score were poor. Logistic regression had the best performance, with an area under the curve of 0.8527. Evaluating Taiwanese advanced CKD patients' EMRs could provide physicians with a good approximation of the patients' 3-year risk of death by machine learning algorithms.

Exploring the Potential Performance of Fibroscan for Predicting and Evaluating Metabolic Syndrome using a Feature Selected Strategy of Machine Learning.

Metabolic syndrome (MetS) includes several conditions that can increase an individual's predisposition to high-risk cardiovascular events, morbidity, and mortality. Non-alcoholic fatty liver disease (NAFLD) is a predominant cause of cirrhosis, which is a global indicator of liver transplantation and is considered the hepatic manifestation of MetS. FibroScan® provides an accurate and non-invasive method for assessing liver steatosis and fibrosis in patients with NAFLD, via a controlled attenuation parameter (CAP) and liver stiffness measurement (LSM or E) scores and has been widely used in current clinical practice. Several machine learning (ML) models with a recursive feature elimination (RFE) algorithm were applied to evaluate the importance of the CAP score. Analysis by ANOVA revealed that five symptoms at different CAP and E score levels were significant. All eight ML models had accuracy scores > 0.9, while treebags and random forest had the best kappa values (0.6439 and 0.6533, respectively). The CAP score was the most important variable in the seven ML models. Machine learning models with RFE demonstrated that using the CAP score to identify patients with MetS may be feasible. Thus, a combination of CAP scores and other significant biomarkers could be used for early detection in predicting MetS.

Association between liver stiffness measurement by transient elastography and chronic kidney disease.

ABSTRACT: Transient elastography or elastometry (TE) is widely used for clinically cirrhosis and liver steatosis examination. Liver fibrosis and fatty liver had been known to share some co-morbidities that may result in chronic impairment in renal function. We conducted a study to analyze the association between scores of 2 TE parameters, liver stiffness measurement (LSM) and controlled attenuation parameter (CAP), with chronic kidney disease among health checkup population.This was a retrospective, cross-sectional study. Our study explored the data of the health checkup population between January 2009 and the end of June 2018 in a regional hospital. All patients were aged more than 18 year-old. Data from a total of 1940 persons were examined in the present study. The estimated glomerular filtration rate (eGFR) was calculated by the modification of diet in renal disease (MDRD-simplify-GFR) equation. Chronic kidney disease (CKD) was defined as eGFR < 60 mL/min/1.73 m2.The median of CAP and LSM score was 242, 265.5, and 4.3, 4.95 in non-CKD (eGFR > 60) and CKD (eGFR < 60) group, respectively. In stepwise regression model, we adjust for LSM, CAP, inflammatory markers, serum biochemistry markers of liver function, and metabolic risks factors. The P value of LSM score, ALT, AST, respectively is .005, <.001, and <.001 in this model.The LSM score is an independent factor that could be used to predict renal function impairment according to its correlation with eGFR. This result can further infer that hepatic fibrosis may be a risk factor for CKD.

Correction: A COVID-19 Pandemic Artificial Intelligence-Based System With Deep Learning Forecasting and Automatic Statistical Data Acquisition: Development and Implementation Study.

[This corrects the article DOI: 10.2196/27806.].

Identify the Characteristics of Metabolic Syndrome and Non-obese Phenotype: Data Visualization and a Machine Learning Approach.

Introduction: A third of the world's population is classified as having Metabolic Syndrome (MetS). Traditional diagnostic criteria for MetS are based on three or more of five components. However, the outcomes of patients with different combinations of specific metabolic components are undefined. It is challenging to be discovered and introduce treatment in advance for intervention, since the related research is still insufficient. Methods: This retrospective cohort study attempted to establish a method of visualizing metabolic components by using unsupervised machine learning and treemap technology to discover the relations between predicting factors and different metabolic components. Several supervised machine-learning models were used to explore significant predictors of MetS and to construct a powerful prediction model for preventive medicine. Results: The random forest had the best performance with accuracy and c-statistic of 0.947 and 0.921, respectively, and found that body mass index, glycated hemoglobin, and controlled attenuation parameter (CAP) score were the optimal primary predictors of MetS. In treemap, high triglyceride level plus high fasting blood glucose or large waist circumference group had higher CAP scores (>260) than other groups. Moreover, 32.2% of patients with high CAP scores during 3 years of follow-up had metabolic diseases are observed. This reveals that the CAP score may be used for detecting MetS, especially for the non-obese MetS phenotype. Conclusions: Machine learning and data visualization can illustrate the complicated relationships between metabolic components and potential risk factors for MetS.

Exploring and predicting mortality among patients with end-stage liver disease without cancer: a machine learning approach.

OBJECTIVE: End-stage liver disease is a global public health problem with a high mortality rate. Early identification of people at risk of poor prognosis is fundamental for decision-making in clinical settings. This study created a machine learning prediction system that provides several related models with visualized graphs, including decision trees, ensemble learning and clustering, to predict mortality in patients with end-stage liver disease. METHODS: A retrospective cohort study was conducted: the training data were from patients enrolled from January 2009 to December 2010 and followed up to December 2014; validation data were from patients enrolled from January 2015 to December 2016 and followed up to January 2019. Hospitalized patients with noncancer-related chronic liver disease were identified from the hospital's electrical medical records. RESULTS: In traditional multivariable logistic regression and Cox proportional hazard model, prothrombin time of international normalized ratio, which was significant with P value = 0.002, odds ratio = 2.790 and hazard ratio 1.363. Besides, blood urea nitrogen and C-reactive protein were also significant, with P value <0.001 and 0.026. The area under the curve was 0.771 in the receiver operating characteristic curve. In machine learning, blood urea nitrogen and age were regarded as the primary factors for predicting mortality. Creatinine, prothrombin time of international normalized ratio and bilirubin were also significant mortality predictors. The area under the curve of the random forest and AdaBoost was 0.838 and 0.792. CONCLUSION: The machine learning techniques provided a comprehensive assessment of patient conditions; it could help physicians make an accurate diagnosis of chronic liver disease and improve healthcare management.

A COVID-19 Pandemic Artificial Intelligence-Based System With Deep Learning Forecasting and Automatic Statistical Data Acquisition: Development and Implementation Study.

BACKGROUND: More than 79.2 million confirmed COVID-19 cases and 1.7 million deaths were caused by SARS-CoV-2; the disease was named COVID-19 by the World Health Organization. Control of the COVID-19 epidemic has become a crucial issue around the globe, but there are limited studies that investigate the global trend of the COVID-19 pandemic together with each country's policy measures. OBJECTIVE: We aimed to develop an online artificial intelligence (AI) system to analyze the dynamic trend of the COVID-19 pandemic, facilitate forecasting and predictive modeling, and produce a heat map visualization of policy measures in 171 countries. METHODS: The COVID-19 Pandemic AI System (CPAIS) integrated two data sets: the data set from the Oxford COVID-19 Government Response Tracker from the Blavatnik School of Government, which is maintained by the University of Oxford, and the data set from the COVID-19 Data Repository, which was established by the Johns Hopkins University Center for Systems Science and Engineering. This study utilized four statistical and deep learning techniques for forecasting: autoregressive integrated moving average (ARIMA), feedforward neural network (FNN), multilayer perceptron (MLP) neural network, and long short-term memory (LSTM). With regard to 1-year records (ie, whole time series data), records from the last 14 days served as the validation set to evaluate the performance of the forecast, whereas earlier records served as the training set. RESULTS: A total of 171 countries that featured in both databases were included in the online system. The CPAIS was developed to explore variations, trends, and forecasts related to the COVID-19 pandemic across several counties. For instance, the number of confirmed monthly cases in the United States reached a local peak in July 2020 and another peak of 6,368,591 in December 2020. A dynamic heat map with policy measures depicts changes in COVID-19 measures for each country. A total of 19 measures were embedded within the three sections presented on the website, and only 4 of the 19 measures were continuous measures related to financial support or investment. Deep learning models were used to enable COVID-19 forecasting; the performances of ARIMA, FNN, and the MLP neural network were not stable because their forecast accuracy was only better than LSTM for a few countries. LSTM demonstrated the best forecast accuracy for Canada, as the root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) were 2272.551, 1501.248, and 0.2723075, respectively. ARIMA (RMSE=317.53169; MAPE=0.4641688) and FNN (RMSE=181.29894; MAPE=0.2708482) demonstrated better performance for South Korea. CONCLUSIONS: The CPAIS collects and summarizes information about the COVID-19 pandemic and offers data visualization and deep learning-based prediction. It might be a useful reference for predicting a serious outbreak or epidemic. Moreover, the system undergoes daily updates and includes the latest information on vaccination, which may change the dynamics of the pandemic.

Machine-Learning Monitoring System for Predicting Mortality Among Patients With Noncancer End-Stage Liver Disease: Retrospective Study.

BACKGROUND: Patients with end-stage liver disease (ESLD) have limited treatment options and have a deteriorated quality of life with an uncertain prognosis. Early identification of ESLD patients with a poor prognosis is valuable, especially for palliative care. However, it is difficult to predict ESLD patients that require either acute care or palliative care. OBJECTIVE: We sought to create a machine-learning monitoring system that can predict mortality or classify ESLD patients. Several machine-learning models with visualized graphs, decision trees, ensemble learning, and clustering were assessed. METHODS: A retrospective cohort study was conducted using electronic medical records of patients from Wan Fang Hospital and Taipei Medical University Hospital. A total of 1214 patients from Wan Fang Hospital were used to establish a dataset for training and 689 patients from Taipei Medical University Hospital were used as a validation set. RESULTS: The overall mortality rate of patients in the training set and validation set was 28.3% (257/907) and 22.6% (145/643), respectively. In traditional clinical scoring models, prothrombin time-international normalized ratio, which was significant in the Cox regression (P<.001, hazard ratio 1.288), had a prominent influence on predicting mortality, and the area under the receiver operating characteristic (ROC) curve reached approximately 0.75. In supervised machine-learning models, the concordance statistic of ROC curves reached 0.852 for the random forest model and reached 0.833 for the adaptive boosting model. Blood urea nitrogen, bilirubin, and sodium were regarded as critical factors for predicting mortality. Creatinine, hemoglobin, and albumin were also significant mortality predictors. In unsupervised learning models, hierarchical clustering analysis could accurately group acute death patients and palliative care patients into different clusters from patients in the survival group. CONCLUSIONS: Medical artificial intelligence has become a cutting-edge tool in clinical medicine, as it has been found to have predictive ability in several diseases. The machine-learning monitoring system developed in this study involves multifaceted analyses, which include various aspects for evaluation and diagnosis. This strength makes the clinical results more objective and reliable. Moreover, the visualized interface in this system offers more intelligible outcomes. Therefore, this machine-learning monitoring system provides a comprehensive approach for assessing patient condition, and may help to classify acute death patients and palliative care patients. Upon further validation and improvement, the system may be used to help physicians in the management of ESLD patients.

Correction: Development of an Online Health Care Assessment for Preventive Medicine: A Machine Learning Approach.

[This corrects the article DOI: 10.2196/18585.].

Development of an Online Health Care Assessment for Preventive Medicine: A Machine Learning Approach.

BACKGROUND: In the era of information explosion, the use of the internet to assist with clinical practice and diagnosis has become a cutting-edge area of research. The application of medical informatics allows patients to be aware of their clinical conditions, which may contribute toward the prevention of several chronic diseases and disorders. OBJECTIVE: In this study, we applied machine learning techniques to construct a medical database system from electronic medical records (EMRs) of subjects who have undergone health examination. This system aims to provide online self-health evaluation to clinicians and patients worldwide, enabling personalized health and preventive health. METHODS: We built a medical database system based on the literature, and data preprocessing and cleaning were performed for the database. We utilized both supervised and unsupervised machine learning technology to analyze the EMR data to establish prediction models. The models with EMR databases were then applied to the internet platform. RESULTS: The validation data were used to validate the online diagnosis prediction system. The accuracy of the prediction model for metabolic syndrome reached 91%, and the area under the receiver operating characteristic (ROC) curve was 0.904 in this system. For chronic kidney disease, the prediction accuracy of the model reached 94.7%, and the area under the ROC curve (AUC) was 0.982. In addition, the system also provided disease diagnosis visualization via clustering, allowing users to check their outcome compared with those in the medical database, enabling increased awareness for a healthier lifestyle. CONCLUSIONS: Our web-based health care machine learning system allowed users to access online diagnosis predictions and provided a health examination report. Users could understand and review their health status accordingly. In the future, we aim to connect hospitals worldwide with our platform, so that health care practitioners can make diagnoses or provide patient education to remote patients. This platform can increase the value of preventive medicine and telemedicine.

Predicting Metabolic Syndrome With Machine Learning Models Using a Decision Tree Algorithm: Retrospective Cohort Study.

BACKGROUND: Metabolic syndrome is a cluster of disorders that significantly influence the development and deterioration of numerous diseases. FibroScan is an ultrasound device that was recently shown to predict metabolic syndrome with moderate accuracy. However, previous research regarding prediction of metabolic syndrome in subjects examined with FibroScan has been mainly based on conventional statistical models. Alternatively, machine learning, whereby a computer algorithm learns from prior experience, has better predictive performance over conventional statistical modeling. OBJECTIVE: We aimed to evaluate the accuracy of different decision tree machine learning algorithms to predict the state of metabolic syndrome in self-paid health examination subjects who were examined with FibroScan. METHODS: Multivariate logistic regression was conducted for every known risk factor of metabolic syndrome. Principal components analysis was used to visualize the distribution of metabolic syndrome patients. We further applied various statistical machine learning techniques to visualize and investigate the pattern and relationship between metabolic syndrome and several risk variables. RESULTS: Obesity, serum glutamic-oxalocetic transaminase, serum glutamic pyruvic transaminase, controlled attenuation parameter score, and glycated hemoglobin emerged as significant risk factors in multivariate logistic regression. The area under the receiver operating characteristic curve values for classification and regression trees and for the random forest were 0.831 and 0.904, respectively. CONCLUSIONS: Machine learning technology facilitates the identification of metabolic syndrome in self-paid health examination subjects with high accuracy.

Clustering Heatmap for Visualizing and Exploring Complex and High-dimensional Data Related to Chronic Kidney Disease.

BACKGROUND: Preventive medicine and primary health care are essential for patients with chronic kidney disease (CKD) because the symptoms of CKD may not appear until the renal function is severely compromised. Early identification of the risk factors of CKD is critical for preventing kidney damage and adverse outcomes. Early recognition of rapid progression to advanced CKD in certain high-risk populations is vital. METHODS: This is a retrospective cohort study, the population screened and the site where the study has been performed. Multivariate statistical analysis was used to assess the prediction of CKD as many potential risk factors are involved. The clustering heatmap and random forest provides an interactive visualization for the classification of patients with different CKD stages. RESULTS: uric acid, blood urea nitrogen, waist circumference, serum glutamic oxaloacetic transaminase, and hemoglobin A1c (HbA1c) were significantly associated with CKD. CKD was highly associated with obesity, hyperglycemia, and liver function. Hypertension and HbA1c were in the same cluster with a similar pattern, whereas high-density lipoprotein cholesterol had an opposite pattern, which was also verified using heatmap. Early staged CKD patients who are grouped into the same cluster as advanced staged CKD patients could be at high risk for rapid decline of kidney function and should be closely monitored. CONCLUSIONS: The clustering heatmap provided a new predictive model of health care management for patients at high risk of rapid CKD progression. This model could help physicians make an accurate diagnosis of this progressive and complex disease.

Dynamic change of surface microbiota with different environmental cleaning methods between two wards in a hospital.

Terminal disinfection and daily cleaning have been performed in hospitals in Taiwan for many years to reduce the risks of healthcare-associated infections. However, the effectiveness of these cleaning approaches and dynamic changes of surface microbiota upon cleaning remain unclear. Here, we report the surface changes of bacterial communities with terminal disinfection and daily cleaning in a medical intensive care unit (MICU) and only terminal disinfection in a respiratory care center (RCC) using 16s ribosomal RNA (rRNA) metagenomics. A total of 36 samples, including 9 samples per sampling time, from each ward were analysed. The clinical isolates were recorded during the sampling time. A large amount of microbial diversity was detected, and human skin microbiota (HSM) was predominant in both wards. In addition, the colonization rate of the HSM in the MICU was higher than that in the RCC, especially for Moraxellaceae. A higher alpha-diversity (p = 0.005519) and a lower UniFrac distance was shown in the RCC due to the lack of daily cleaning. Moreover, a significantly higher abundance among Acinetobacter sp., Streptococcus sp. and Pseudomonas sp. was shown in the RCC compared to the MICU using the paired t test. We concluded that cleaning changes might contribute to the difference in diversity between two wards.

First-ever ischemic stroke in elderly patients: predictors of functional outcome following carotid artery stenting.

Age is an important risk factor for stroke, and carotid artery stenosis is the primary cause of first-ever ischemic stroke. Timely intervention with stenting procedures can effectively prevent secondary stroke; however, the impact of stenting on various periprocedural physical functionalities has never been thoroughly investigated. The primary aim of this study was to investigate whether prestenting characteristics were associated with long-term functional outcomes in patients presenting with first-ever ischemic stroke. The secondary aim was to investigate whether patient age was an important factor in outcomes following stenting, measured by the modified Rankin scale (mRS). In total, 144 consecutive patients with first-ever ischemic stroke who underwent carotid artery stenting from January 2010 to November 2014 were included. Clinical data were obtained by review of medical records. The Barthel index (BI) and mRS were used to assess disability before stenting and at 12-month follow-up. In total, 72/144 patients showed improvement (mRS[+]), 71 showed stationary and one showed deterioration in condition (mRS[-]). The prestenting parameters, ratio of cerebral blood volume (1.41 vs 1.2 for mRS[-] vs mRS[+]), BI (75 vs 85), and high-sensitivity C-reactive protein (hsCRP 5.0 vs 3.99), differed significantly between the two outcome groups (P<0.05). The internal carotid artery/common carotid artery ratio (P=0.011), BI (P=0.019), ipsilateral internal carotid artery resistance index (P=0.003), and HbA1c (P=0.039) were all factors significantly associated with patient age group. There was no significant association between age and poststenting outcome measured by mRS with 57% of patients in the ≥75 years age group showing mRS(-) and 43% showing mRS(+) (P=0.371). Our findings indicate that in our elderly patient series, carotid artery stenting may benefit a significant proportion of carotid stenotic patients regardless of age. Ratio of cerebral blood volume, BI, and admission hsCRP could serve as important predictors of mRS improvement and may facilitate differentiation of patients at baseline.

Role of Extracranial Carotid Duplex and Computed Tomography Perfusion Scanning in Evaluating Perfusion Status of Pericarotid Stenting.

Carotid stenting is an effective treatment of choice in terms of treating ischemic stroke patients with concomitant carotid stenosis. Though computed tomography perfusion scan has been recognized as a standard tool to monitor/follow up this group of patients, not everyone could endure due to underlying medical illness. In contrast, carotid duplex is a noninvasive assessment tool and could track patient clinical condition in real time. In this study we found that "resistance index" of the carotid ultrasound could detect flow changes before and after the stenting procedure, thus having great capacity to replace the role of computed tomography perfusion exam.

Impact of baseline characteristics on outcomes of carotid artery stenting in acute ischemic stroke patients.

Carotid artery stenting is an effective treatment for ischemic stroke patients with moderate-to-severe carotid artery stenosis. However, the midterm outcome for patients undergoing this procedure varies considerably with baseline characteristics. To determine the impact of baseline characteristics on outcomes following carotid artery stenting, data from 107 eligible patients with a first episode of ischemic stroke were collected by retrospective chart review. A modified Rankin Scale (mRS) was used to divide patients into two baseline groups, mRS ≤2 and mRS >2. A three-step decision-tree statistical analysis was conducted. After weighting the decision-tree parameters, the following impact hierarchy was obtained: admission low-density lipoprotein, gouty arthritis, chronic kidney disease, ipsilateral common carotid artery resistance index, contralateral ophthalmic artery resistance index, sex, and dyslipidemia. The finite-state machine model demonstrated that, in patients with baseline mRS ≤2, 46% had an improved mRS score at follow-up, whereas 54% had a stable mRS score. In patients with baseline mRS >2, a stable mRS score was observed in 75%, improved score in 23%, and a poorer score in 2%. Admission low-density lipoprotein was the strongest predictive factor influencing poststenting outcome. In addition, our study provides further evidence that carotid artery stenting can be of benefit in first-time ischemic stroke patients with baseline mRS scores >2.

Limited sampling strategy models for estimating AUC for amlodipine in Chinese healthy volunteers / 药学学报

This study aims to save cost of sampling for estimating the area under the amlodipine plasma concentration versus time curve in 24 hours (AUC(0-24 h)). Limited sampling strategy (LSS) models was developed and validated by mutiple regression model within 4 or fewer amlodipine concentration values. Absolute prediction error (APE), root of mean square error (RMSE) and visual predict check were used as criterion. The results of Jackknife validation showed that fifteen (9.4%) of the 160 LSS based on regression analysis were not within an APE of 15% by using one concentration-time point. 156 (97.5%), 159 (99.4%) and 160 (100%) of the 160 LSS model were capable of predicting within an APE 15% by using 2, 3, 4 points, separately. Limited sampling strategies have been developed and validated for estimating AUC(0-24 h) of amlodipine. The present study indicated that the implemention of both 5 mg and 10 mg dosage could enable accurate predictions of AUC(0-24 h) by the same LSS model. This study shows that 12, 4, 24, 2 h after administration are key sampling time points. The combination of (12, 4), (12, 4, 24) or (12, 4, 24, 2 h) might be chosen as sampling hours for predicting AUC(0-24 h) in practical application according to requirement.

Optimal design of resonant piezoelectric buzzer from a perspective of vibration-absorber theory.

In this paper, an optimization technique is presented for the design of piezoelectric buzzers. This design technique aims at finding the optimal configuration of the coupled cavity and diaphragm structure to maximize the sound pressure output. Instead of measuring the material constants of the piezoelectric ceramic and the metal diaphragm, an "added-mass method" is developed to estimate the equivalent electromechanical parameters of the system on which an analogous circuit can be established. The electrical impedance and on-axis sound pressure level of the piezoelectric buzzer can be simulated by solving the loop equations of the electromechanoacoustical analogous circuit. An interesting finding of this research is that the nature of the piezoelectric buzzer bears remarkable resemblance to that in the dynamic vibration absorber theory. Much physical insight can be gained by exploiting this resemblance in search of the optimal configuration. According to the system characteristic equation, a design chart was devised to "lock" the critical frequency at which the system delivers the maximal output. On the basis of the analogous circuit and the vibration absorber theory, an optimal design was found with constrained optimization formalism. Experiments were conducted to justify the optimal design. The results showed that the performance was significantly improved using the optimal design over the original design. Design guidelines for the piezoelectric buzzers are summarized.