Algorithm-based Prediction Research Articles

Carboxylesterase 1-Based Drug-Drug Interaction Potential of Remimazolam: In-Vitro Studies and Literature Review.

The ultra-short-acting benzodiazepine remimazolam, approved for procedural sedation and general anesthesia, is inactivated by carboxylesterase 1 (CES1). Remimazolam´s involvement in CES1-mediated drug-drug interactions (DDIs) was investigated. Possible interactions of remimazolam were studied in co-exposure experiments with eleven different drugs. Further, substrates and inhibitors of CES1, identified in the literature, were evaluated for possible in-vivo inhibition using pharmacokinetic and Ki or IC50 values. Compounds with only one published inhibitory concentration and CES1 substrates lacking inhibition data were assigned conservative Ki values. In human liver homogenates and/or blood cells, remimazolam showed no significant inhibition of esmolol and landiolol metabolism, which, in turn, at up to 98 and 169 μM, respectively, did not inhibit remimazolam hydrolysis by human liver homogenates. In human liver S9 fractions, IC50 values ranged from 0.69 μM (simvastatin) and 57 μM (diltiazem) to > 100 μM (atorvastatin) and, for the remaining test items (bupropion, carvedilol, nelfinavir, nitrendipine, and telmisartan), they ranged from 126 to 658 μM. Remifentanil was ineffective even at 1250 μM. Guidance-conforming evaluation revealed no relevant drug-drug interactions with remimazolam via CES1. The algorithm-based predictions were consistent with human study data. Among CES1 inhibitors and substrates identified in the literature, only dapsone and rufinamide were found to be possible in-vivo inhibitors of remimazolam metabolism. Data and analyses suggest a very low potential of remimazolam for pharmacokinetic DDIs mediated by CES1. The theoretical approach and compiled data are not specific to remimazolam and, hence, applicable in the evaluation of other CES1 substrates.

Algorithm Versus Expert: Machine Learning Versus Surgeon-Predicted Symptom Improvement After Carpal Tunnel Release.

Surgeons rely on clinical experience when making predictions about treatment effects. Incorporating algorithm-based predictions of symptom improvement after carpal tunnel release (CTR) could support medical decision-making. However, these algorithm-based predictions need to outperform predictions made by surgeons to add value. We compared predictions of a validated prediction model for symptom improvement after CTR with predictions made by surgeons. This cohort study included 97 patients scheduled for CTR. Preoperatively, surgeons estimated each patient's probability of improvement 6 months after surgery, defined as reaching the minimally clinically important difference on the Boston Carpal Tunnel Syndrome Symptom Severity Score. We assessed model and surgeon performance using calibration (calibration belts), discrimination (area under the curve [AUC]), sensitivity, and specificity. In addition, we assessed the net benefit of decision-making based on the prediction model's estimates vs the surgeon's judgement. The surgeon predictions had poor calibration and suboptimal discrimination (AUC 0.62, 95%-CI 0.49-0.74), while the prediction model showed good calibration and appropriate discrimination (AUC 0.77, 95%-CI 0.66-0.89, P = .05). The accuracy of surgeon predictions was 0.65 (95%-CI 0.37-0.78) vs 0.78 (95%-CI 0.67-0.89) for the prediction model (P = .03). The sensitivity of surgeon predictions and the prediction model was 0.72 (95%-CI 0.15-0.96) and 0.85 (95%-CI 0.62-0.97), respectively (P = .04). The specificity of the surgeon predictions was similar to the model's specificity (P = .25). The net benefit analysis showed better decision-making based on the prediction model compared with the surgeons' decision-making (ie, more correctly predicted improvements and/or fewer incorrectly predicted improvements). The prediction model outperformed surgeon predictions of improvement after CTR in terms of calibration, accuracy, and sensitivity. Furthermore, the net benefit analysis indicated that using the prediction model instead of relying solely on surgeon decision-making increases the number of patients who will improve after CTR, without increasing the number of unnecessary surgeries.

Data-driven, two-stage machine learning algorithm-based prediction scheme for assessing 1-year and 3-year mortality risk in chronic hemodialysis patients

Life expectancy is likely to be substantially reduced in patients undergoing chronic hemodialysis (CHD). However, machine learning (ML) may predict the risk factors of mortality in patients with CHD by analyzing the serum laboratory data from regular dialysis routine. This study aimed to establish the mortality prediction model of CHD patients by adopting two-stage ML algorithm-based prediction scheme, combined with importance of risk factors identified by different ML methods. This is a retrospective, observational cohort study. We included 800 patients undergoing CHD between December 2006 and December 2012 in Shin-Kong Wu Ho-Su Memorial Hospital. This study analyzed laboratory data including 44 indicators. We used five ML methods, namely, logistic regression (LGR), decision tree (DT), random forest (RF), gradient boosting (GB), and eXtreme gradient boosting (XGB), to develop a two-stage ML algorithm-based prediction scheme and evaluate the important factors that predict CHD mortality. LGR served as a bench method. Regarding the validation and testing datasets from 1- and 3-year mortality prediction model, the RF had better accuracy and area-under-curve results among the five different ML methods. The stepwise RF model, which incorporates the most important factors of CHD mortality risk based on the average rank from DT, RF, GB, and XGB, exhibited superior predictive performance compared to LGR in predicting mortality among CHD patients over both 1-year and 3-year periods. We had developed a two-stage ML algorithm-based prediction scheme by implementing the stepwise RF that demonstrated satisfactory performance in predicting mortality in patients with CHD over 1- and 3-year periods. The findings of this study can offer valuable information to nephrologists, enhancing patient-centered decision-making and increasing awareness about risky laboratory data, particularly for patients with a high short-term mortality risk.

Artificial Neural Network and Genetic Algorithm-based prediction of photovoltaic panel performance with porous foam gradient and nano-enhanced phase change material

Photovoltaic panels (PV) become less efficient as the temperature rises. The use of porous foam gradient, phase change materials (PCMs), and nanoparticle-PCM (nePCM) can effectively lower the temperature of PV panels. This study employed numerical analysis using the finite volume method to investigate the effects of different types of PCMs (n-octadecane, decanoic acid natural, lauric acid, paraffin wax, and RT-42), mono and hybrid nanoparticles/PCM, and the position of PCM in the design system. The results indicate that the highest electrical efficiency of 9 % is achieved when the PCM is placed under the PV panel and the thickness of the PCM layer is 0.6 times the height of the microchannel. Furthermore, the use of a porous medium with a variable porosity coefficient that increases along the y-direction can significantly improve thermal and electrical efficiency (approximately 5–42 % and 5–48 % respectively). Additionally, incorporating PCM and hybrid nanoparticles-PCM enhances PV performance compared to the absence of nanoparticles. For TiO2, Ag, Al2O3, GO, and MWCNT nanoparticles, the thermal efficiency, electrical efficiency, and exergetic efficiency increase by approximately 24.18 %, 26.15 %, 26.28 %, 26.30 %, and 26.33 % respectively. Similarly, for hybrid nanoparticles such as GO-Al2O3, GO-Al2O3-MWCNT, MWCNT-GO, and MWCNT-Al2O3, these values increase by approximately 26.32 %, 26.3419 %, 26.3432 %, and 26.335 % respectively for thermal efficiency, and 5.16 %, 5.1702 %, 5.1704 %, and 5.1689 % respectively for electrical efficiency. Moreover, n-octadecane exhibits the highest melting fraction, while paraffin wax has the lowest melting fraction. Lastly, the genetic algorithm-neural network method was employed to estimate the PV electrical efficiency.

A random forest algorithm-based prediction model for moderate to severe acute postoperative pain after orthopedic surgery under general anesthesia

BackgroundPostoperative pain is one of the most common complications after surgery. In order to detect early and intervene in time for moderate to severe postoperative pain, it is necessary to identify risk factors and construct clinical prediction models. This study aimed to identify significant risk factors and establish a better-performing model to predict moderate to severe acute postoperative pain after orthopedic surgery under general anesthesia.MethodsPatients who underwent orthopedic surgery under general anesthesia were divided into patients with moderate to severe pain group (group P) and patients without moderate to severe pain group (group N) based on VAS scores. The features selected by Lasso regression were processed by the random forest and multivariate logistic regression models to predict pain outcomes. The classification performance of the two models was evaluated through the testing set. The area under the curves (AUC), the accuracy of the classifiers, and the classification error rate for both classifiers were calculated, the better-performing model was used to predict moderate to severe acute postoperative pain after orthopedic surgery under general anesthesia.ResultsA total of 327 patients were enrolled in this study (228 in the training set and 99 in the testing set). The incidence of moderate to severe postoperative pain was 41.3%. The random forest model revealed a classification error rate of 25.2% and an AUC of 0.810 in the testing set. The multivariate logistic regression model revealed a classification error rate of 31.3% and an AUC of 0.764 in the testing set. The random forest model was chosen for predicting clinical outcomes in this study. The risk factors with the greatest and second contribution were immobilization and duration of surgery, respectively.ConclusionsThe random forest model can be used to predict moderate to severe acute postoperative pain after orthopedic surgery under general anesthesia, which is of potential clinical application value.

Identification of high-risk factors for recurrence of colon cancer following complete mesocolic excision: An 8-year retrospective study.

Colon cancer recurrence is a common adverse outcome for patients after complete mesocolic excision (CME) and greatly affects the near-term and long-term prognosis of patients. This study aimed to develop a machine learning model that can identify high-risk factors before, during, and after surgery, and predict the occurrence of postoperative colon cancer recurrence. The study included 1187 patients with colon cancer, including 110 patients who had recurrent colon cancer. The researchers collected 44 characteristic variables, including patient demographic characteristics, basic medical history, preoperative examination information, type of surgery, and intraoperative information. Four machine learning algorithms, namely extreme gradient boosting (XGBoost), random forest (RF), support vector machine (SVM), and k-nearest neighbor algorithm (KNN), were used to construct the model. The researchers evaluated the model using the k-fold cross-validation method, ROC curve, calibration curve, decision curve analysis (DCA), and external validation. Among the four prediction models, the XGBoost algorithm performed the best. The ROC curve results showed that the AUC value of XGBoost was 0.962 in the training set and 0.952 in the validation set, indicating high prediction accuracy. The XGBoost model was stable during internal validation using the k-fold cross-validation method. The calibration curve demonstrated high predictive ability of the XGBoost model. The DCA curve showed that patients who received interventional treatment had a higher benefit rate under the XGBoost model. The external validation set's AUC value was 0.91, indicating good extrapolation of the XGBoost prediction model. The XGBoost machine learning algorithm-based prediction model for colon cancer recurrence has high prediction accuracy and clinical utility.

Machine-learning algorithm-based prediction of a diagnostic model based on oxidative stress-related genes involved in immune infiltration in diabetic nephropathy patients.

Diabetic nephropathy (DN) is the most prevalent microvascular consequence of diabetes and has recently risen to the position of the world's second biggest cause of end-stage renal diseases. Growing studies suggest that oxidative stress (OS) responses are connected to the advancement of DN. This study aimed to developed a novel diagnostic model based on OS-related genes. The differentially expressed oxidative stress-related genes (DE-OSRGs) experiments required two human gene expression datasets, which were given by the GEO database (GSE30528 and GSE96804, respectively). The potential diagnostic genes were identified using the SVM-RFE assays and the LASSO regression model. CIBERSORT was used to determine the compositional patterns of the 22 different kinds of immune cell fraction seen in DN. These estimates were based on the combined cohorts. DN serum samples and normal samples were both subjected to RT-PCR in order to investigate the degree to which certain genes were expressed. In this study, we were able to locate 774 DE-OSRGs in DN. The three marker genes (DUSP1, PRDX6 and S100A8) were discovered via machine learning on two different machines. The high diagnostic value was validated by ROC tests, which focused on distinguishing DN samples from normal samples. The results of the CIBERSORT study suggested that DUSP1, PRDX6, and S100A8 may be associated to the alterations that occur in the immunological microenvironment of DN patients. Besides, the results of RT-PCR indicated that the expression of DUSP1, PRDX6, and S100A8 was much lower in DN serum samples compared normal serum samples. The diagnostic value of the proposed model was likewise verified in our cohort, with an area under the curve of 9.946. Overall, DUSP1, PRDX6, and S100A8 were identified to be the three diagnostic characteristic genes of DN. It's possible that combining these genes will be effective in diagnosing DN and determining the extent of immune cell infiltration.

A data-driven approach to predict the compressive strength of alkali-activated materials and correlation of influencing parameters using SHapley Additive exPlanations (SHAP) analysis

This research used gene expression programming (GEP) and multi expression programming (MEP) to determine the compressive strength (CS) of alkali-activated material (AAM) to compare and develop more reliable genetic algorithm-based prediction models. To learn more about how raw ingredients affect and interact with the CS of AAM, a SHapley Additive exPlanations (SHAP) analysis was conducted. A comprehensive dataset containing 676 points with fifteen influential parameters was formulated from the previously published literature. According to this study, considering the impact of 15 input variables, both genetic algorithms produced results close to the experimental CS (retrieved from the literature). When the performance of the GEP and MEP models were compared, it was found that the MEP model, with an R2 of 0.86, performed better than the GEP model, with an R2 of 0.82. The assessment of the statistical parameters of generated models revealed that the MEP model was more effective. Additionally, SHAP analysis revealed that slag content, followed by the specimen's age, sodium silicate, and curing temperature, showed a positive correlation with CS of AAM, which were the most important parameters. The results also revealed the importance of chemical contents, i.e., CaO, SiO2, Al2O3, of FA and slag on the CS of AAM. The built models might be used to compute the CS of AAMs with varying input parameter values, minimizing the effort, time, and cost of unnecessary lab tests. Furthermore, the outcomes of the SHAP study might help researchers and the industry determine the quantity or composition of raw ingredients when producing AAMs.

Research on indoor light environment optimization of nursing home community based on artificial neural network

ABSTRACT China's ageing population is set to peak in 2040 and nursing homes are where most elderly people live. In order to improve the comfort of the elderly in residential homes, it is necessary to investigate a greener lighting solution. The greenness of a lighting scheme is reflected in the fact that it meets the two needs of economy and comfort. The design of indoor light will directly affect their quality of life, as well as their physiological and psychological changes, so the elderly have special needs for light. To address the problem of optimizing the indoor light environment in nursing home communities, this study proposes an artificial neural network algorithm-based prediction method for optimizing the indoor light environment, which in turn improves the comfort of indoor lighting for the elderly. The results show that the mean absolute error, mean square error and root mean square error are 0.49445, 0.49984 and 0.70699 respectively after using the artificial neural network algorithm for light control. after 217 training sessions, the method can meet the pre-set error rate of 10−2 and achieve the expected purpose. The R-values for the training, validation and test sets hovered around 0.9, indicating a good fit.

Random Forest Algorithm-Based Prediction of Solvation Gibbs Energies

Random Forest Algorithm-Based Prediction of Solvation Gibbs Energies

Influence of Input Features and EMG Type on Ankle Joint Torque Prediction with Support Vector Regression.

Reliable and accurate EMG-driven prediction of joint torques are instrumental in the control of wearable robotic systems. This study investigates how different EMG input features affect the machine learning algorithm-based prediction of ankle joint torque in isometric and dynamic conditions. High-density electromyography (HD-EMG) of five lower leg muscles were recorded during isometric contractions and dynamic tasks. Four datasets (HD-EMG, HD-EMG with reduced dimensionality, features extracted from HD-EMG with Convolutional Neural Network, and bipolar EMG) were created and used alone or in combination with joint kinematic information for the prediction of ankle joint torque using Support Vector Regression. The performance was evaluated under intra-session, inter-subject, and inter-session cases. All HD-EMG-derived datasets led to significantly more accurate isometric ankle torque prediction than the bipolar EMG datasets. The highest torque prediction accuracy for the dynamic tasks was achieved using bipolar EMG or HD-EMG with reduced dimensionality in combination with kinematic features. The findings of this study contribute to the knowledge allowing an informed selection of appropriate features for EMG-driven torque prediction.

Integrating Health Data-Driven Machine Learning Algorithms to Evaluate Risk Factors of Early Stage Hypertension at Different Levels of HDL and LDL Cholesterol.

Purpose: Cardiovascular disease (CVD) is a major worldwide health burden. As the risk factors of CVD, hypertension, and hyperlipidemia are most mentioned. Early stage hypertension in the population with dyslipidemia is an important public health hazard. This study was the application of data-driven machine learning (ML), demonstrating complex relationships between risk factors and outcomes and promising predictive performance with vast amounts of medical data, aimed to investigate the association between dyslipidemia and the incidence of early stage hypertension in a large cohort with normal blood pressure at baseline. Methods: This study analyzed annual health screening data for 71,108 people from 2005 to 2017, including data for 27 risk-related indicators, sourced from the MJ Group, a major health screening center in Taiwan. We used five machine learning (ML) methods—stochastic gradient boosting (SGB), multivariate adaptive regression splines (MARS), least absolute shrinkage and selection operator regression (Lasso), ridge regression (Ridge), and gradient boosting with categorical features support (CatBoost)—to develop a multi-stage ML algorithm-based prediction scheme and then evaluate important risk factors at the early stage of hypertension, especially for groups with high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C) levels within or out of the reference range. Results: Age, body mass index, waist circumference, waist-to-hip ratio, fasting plasma glucose, and C-reactive protein (CRP) were associated with hypertension. The hemoglobin level was also a positive contributor to blood pressure elevation and it appeared among the top three important risk factors in all LDL-C/HDL-C groups; therefore, these variables may be important in affecting blood pressure in the early stage of hypertension. A residual contribution to blood pressure elevation was found in groups with increased LDL-C. This suggests that LDL-C levels are associated with CPR levels, and that the LDL-C level may be an important factor for predicting the development of hypertension. Conclusion: The five prediction models provided similar classifications of risk factors. The results of this study show that an increase in LDL-C is more important than the start of a drop in HDL-C in health screening of sub-healthy adults. The findings of this study should be of value to health awareness raising about hypertension and further discussion and follow-up research.

Machine learning algorithms’ accuracy in predicting kidney disease progression: a systematic review and meta-analysis

BackgroundKidney disease progression rates vary among patients. Rapid and accurate prediction of kidney disease outcomes is crucial for disease management. In recent years, various prediction models using Machine Learning (ML) algorithms have been established in nephrology. However, their accuracy have been inconsistent. Therefore, we conducted a systematic review and meta-analysis to investigate the diagnostic accuracy of ML algorithms for kidney disease progression.MethodsWe searched PubMed, EMBASE, Cochrane Central Register of Controlled Trials, the Chinese Biomedicine Literature Database, Chinese National Knowledge Infrastructure, Wanfang Database, and the VIP Database for diagnostic studies on ML algorithms’ accuracy in predicting kidney disease prognosis, from the establishment of these databases until October 2020. Two investigators independently evaluate study quality by QUADAS-2 tool and extracted data from single ML algorithm for data synthesis using the bivariate model and the hierarchical summary receiver operating characteristic (HSROC) model.ResultsFifteen studies were left after screening, only 6 studies were eligible for data synthesis. The sample size of these 6 studies was 12,534, and the kidney disease types could be divided into chronic kidney disease (CKD) and Immunoglobulin A Nephropathy, with 5 articles using end-stage renal diseases occurrence as the primary outcome. The main results indicated that the area under curve (AUC) of the HSROC was 0.87 (0.84–0.90) and ML algorithm exhibited a strong specificity, 95% confidence interval and heterogeneity (I2) of (0.87, 0.84–0.90, [I2 99.0%]) and a weak sensitivity of (0.68, 0.58–0.77, [I2 99.7%]) in predicting kidney disease deterioration. And the the results of subgroup analysis indicated that ML algorithm’s AUC for predicting CKD prognosis was 0.82 (0.79–0.85), with the pool sensitivity of (0.64, 0.49–0.77, [I2 99.20%]) and pool specificity of (0.84, 0.74–0.91, [I2 99.84%]). The ML algorithm’s AUC for predicting IgA nephropathy prognosis was 0.78 (0.74–0.81), with the pool sensitivity of (0.74, 0.71–0.77, [I2 7.10%]) and pool specificity of (0.93, 0.91–0.95, [I2 83.92%]).ConclusionTaking advantage of big data, ML algorithm-based prediction models have high accuracy in predicting kidney disease progression, we recommend ML algorithms as an auxiliary tool for clinicians to determine proper treatment and disease management strategies.

Machine Learning Algorithm-Based Prediction Model for the Augmented Use of Clozapine with Electroconvulsive Therapy in Patients with Schizophrenia.

The augmentation of clozapine with electroconvulsive therapy (ECT) has been an optimal treatment option for patients with treatment- or clozapine-resistant schizophrenia. Using data from the Research on Asian Psychotropic Prescription Patterns for Antipsychotics survey, which was the largest international psychiatry research collaboration in Asia, our study aimed to develop a machine learning algorithm-based substantial prediction model for the augmented use of clozapine with ECT in patients with schizophrenia in terms of precision medicine. A random forest model and least absolute shrinkage and selection operator (LASSO) model were used to develop a substantial prediction model for the augmented use of clozapine with ECT. Among the 3744 Asian patients with schizophrenia, those treated with a combination of clozapine and ECT were characterized by significantly greater proportions of females and inpatients, a longer duration of illness, and a greater prevalence of negative symptoms and social or occupational dysfunction than those not treated. In the random forest model, the area under the curve (AUC), which was the most preferred indicator of the prediction model, was 0.774. The overall accuracy was 0.817 (95% confidence interval, 0.793–0.839). Inpatient status was the most important variable in the substantial prediction model, followed by BMI, age, social or occupational dysfunction, persistent symptoms, illness duration > 20 years, and others. Furthermore, the AUC and overall accuracy of the LASSO model were 0.831 and 0.644 (95% CI, 0.615–0.672), respectively. Despite the subtle differences in both AUC and overall accuracy of the random forest model and LASSO model, the important variables were commonly shared by the two models. Using the machine learning algorithm, our findings allow the development of a substantial prediction model for the augmented use of clozapine with ECT in Asian patients with schizophrenia. This substantial prediction model can support further studies to develop a substantial prediction model for the augmented use of clozapine with ECT in patients with schizophrenia in a strict epidemiological context.

Assessing the Usability of a Novel Wearable Remote Patient Monitoring Device for the Early Detection of In-Hospital Patient Deterioration: Observational Study.

BackgroundPatients admitted to general wards are inherently at risk of deterioration. Thus, tools that can provide early detection of deterioration may be lifesaving. Frequent remote patient monitoring (RPM) has the potential to allow such early detection, leading to a timely intervention by health care providers.ObjectiveThis study aimed to assess the potential of a novel wearable RPM device to provide timely alerts in patients at high risk for deterioration.MethodsThis prospective observational study was conducted in two general wards of a large tertiary medical center. Patients determined to be at high risk to deteriorate upon admission and assigned to a telemetry bed were included. On top of the standard monitoring equipment, a wearable monitor was attached to each patient, and monitoring was conducted in parallel. The data gathered by the wearable monitors were analyzed retrospectively, with the medical staff being blinded to them in real time. Several early warning scores of the risk for deterioration were used, all calculated from frequent data collected by the wearable RPM device: these included (1) the National Early Warning Score (NEWS), (2) Airway, Breathing, Circulation, Neurology, and Other (ABCNO) score, and (3) deterioration criteria defined by the clinical team as a “wish list” score. In all three systems, the risk scores were calculated every 5 minutes using the data frequently collected by the wearable RPM device. Data generated by the early warning scores were compared with those obtained from the clinical records of actual deterioration among these patients.ResultsIn total, 410 patients were recruited and 217 were included in the final analysis. The median age was 71 (IQR 62-78) years and 130 (59.9%) of them were male. Actual clinical deterioration occurred in 24 patients. The NEWS indicated high alert in 16 of these 24 (67%) patients, preceding actual clinical deterioration by 29 hours on average. The ABCNO score indicated high alert in 18 (75%) of these patients, preceding actual clinical deterioration by 38 hours on average. Early warning based on wish list scoring criteria was observed for all 24 patients 40 hours on average before clinical deterioration was detected by the medical staff. Importantly, early warning based on the wish list scoring criteria was also observed among all other patients who did not deteriorate.ConclusionsFrequent remote patient monitoring has the potential for early detection of a high risk to deteriorate among hospitalized patients, using both grouped signal-based scores and algorithm-based prediction. In this study, we show the ability to formulate scores for early warning by using RPM. Nevertheless, early warning scores compiled on the basis of these data failed to deliver reasonable specificity. Further efforts should be directed at improving the specificity and sensitivity of such tools.Trial RegistrationClinicalTrials.gov NCT04220359; https://clinicaltrials.gov/ct2/show/NCT04220359

Prediction of Quality Food Sale in Mart Using the AI-Based TOR Method

“The golden highways of the technological route serve as the ultimate metaphor for the society of artificial intelligence,” explains the author of this present excursion. Unsupervised predictor is a technique that is used in all upbringings and algometry computations, and it is the foundation of all of them. The accuracy-based artificial intelligence and genetic algorithm-based prediction are becoming more important in business intelligence, with the crossover revealing the highest percentage of interest-based profit in the food industry. Our suggested approach aims to determine the utmost amplification of the food business profit with the help of food sale prediction. Hence, owners of outlet sale, supermarket sale, grocery store, and so on can maintain the stock. Therefore, our study strives to elucidate the relationship between food business intelligence and artificial intelligence in the context of soft computing. The study combines the previous data with the record of the current data and compares the existing development of business management in order to create the topic of the TOR method, which will be the artificial intelligence buzzword in the town-based food business intelligence. This paper uses artificial intelligence, business intelligence, and neural networks to forecast the food sale prediction by evaluating the data of each individual client. Consider the situation in which this technology is applied in every grocery store in the country. Our suggested model has achieved low mean squared error and low variance. Food mart owners will have a far better grasp of their clientele and will be able to categorise them in this situation, enabling them to proceed with future sales forecasting.

A New Random Forest Algorithm-Based Prediction Model of Post-operative Mortality in Geriatric Patients With Hip Fractures.

BackgroundPost-operative mortality risk assessment for geriatric patients with hip fractures (HF) is a challenge for clinicians. Early identification of geriatric HF patients with a high risk of post-operative death is helpful for early intervention and improving clinical prognosis. However, a single significant risk factor of post-operative death cannot accurately predict the prognosis of geriatric HF patients. Therefore, our study aims to utilize a machine learning approach, random forest algorithm, to fabricate a prediction model for post-operative death of geriatric HF patients.MethodsThis retrospective study enrolled consecutive geriatric HF patients who underwent treatment for surgery. The study cohort was divided into training and testing datasets at a 70:30 ratio. The random forest algorithm selected or excluded variables according to the feature importance. Least absolute shrinkage and selection operator (Lasso) was utilized to compare feature selection results of random forest. The confirmed variables were used to create a simplified model instead of a full model with all variables. The prediction model was then verified in the training dataset and testing dataset. Additionally, a prediction model constructed by logistic regression was used as a control to evaluate the efficiency of the new prediction model.ResultsFeature selection by random forest algorithm and Lasso regression demonstrated that seven variables, including age, time from injury to surgery, chronic obstructive pulmonary disease (COPD), albumin, hemoglobin, history of malignancy, and perioperative blood transfusion, could be used to predict the 1-year post-operative mortality. The area under the curve (AUC) of the random forest algorithm-based prediction model in training and testing datasets were 1.000, and 0.813, respectively. While the prediction tool constructed by logistic regression in training and testing datasets were 0.895, and 0.797, respectively.ConclusionsCompared with logistic regression, the random forest algorithm-based prediction model exhibits better predictive ability for geriatric HF patients with a high risk of death within post-operative 1 year.

An Integrated Machine Learning Model for Heart Disease Classification and Categorization

Progressions in the coordination among the machine learning algorithms, helped to achieve high accuracy and reliability in decision-making systems. Due to the impact and importance of heart diseases in real life, designing the efficient heart disease prediction model become a pivotal aspect today. Former research scholars applied the popular supervised machine learning models like Decision Trees, Naïve Bayes, ANN's, and FNN's to implement the heart disease prediction systems. As the heart disease prediction process is a multi-layered operation, each layer is expecting the optimal machine learning algorithm and the coordination among the algorithms of different layers to minimize the errors in prediction results. In this study, we proposed a new fuzzy neural-genetic algorithm to design an efficient and accurate heart disease prediction system. In our prediction system, we integrated the genetic algorithm, neural networks, and fuzzy logic technologies for training, classification, and categorization processes respectively. Experimental evaluations of Cleveland's heart disease dataset proved that the proposed fuzzy neural-genetic algorithm-based prediction system achieved high accuracy and low error rate when compared with the other machine learning models.

XGBoost algorithm-based prediction of safety assessment for pipelines

Pipeline safety is closely related to people's lives, environment and economic development. The traditional methods for pipeline inspection are laborious and very expensive, such as non-destructive examination (NDE) and engineering assessment (EA). In recent years, machine learning (ML) methods have been proven useful to solve pipeline assessment problems. These methods have mainly focused on a characteristic analysis of weld defects or pipeline life curves, and have not provided systematic and effective assessment methods for the pipeline industry. In this study, after comparing with other reference ML models, XGBoost algorithm-based predictive model is proposed for safety assessment of pipelines. First, two models based on XGBoost are established and tuned to create relationships between 9 parameters (e.g., internal diameters of pipelines) and NDE results/EA results. Then grid search and various statistical metrics (confusion matrices, AUC-ROC and confidence levels) are applied to optimize and assess these models. In addition, the prediction interpretability of XGBoost is analyzed to show its usefulness and applicability in industry. The results show that both of these two models satisfy the required predictive performance. The XGBoost model based on the EA results is finally selected thanks to its superiority by achieving an accuracy of 98.5%, AUC-ROC of 99% in testing, higher ratio of certain prediction (91.8%) and other metrics (precision of 98.8% and recall of 98.7%). In the analysis of model interpretations, a maximum operating pressure (MOP) is found to be the most important feature to affect the safety of pipelines. The results also show that when the current ML method based on the historical results of NDE and EA is established and proven stable, it can accurately predict the future risk of new pipelines in the absence of NDE and EA. In the studied research region, up to 91.8% of the funds and manpower spent on NDE and EA can be saved based on this ML method. Overall, the proposed XGBoost model provides an alternative to help industry evaluate pipeline welds with very few excavations and manual inspections. As more data from other regions is collected and used to train ML models, this method can be widely applied in industry.

Machine-Learning Algorithm-Based Prediction of Diagnostic Gene Biomarkers Related to Immune Infiltration in Patients With Chronic Obstructive Pulmonary Disease.

ObjectiveThis study aims to identify clinically relevant diagnostic biomarkers in chronic obstructive pulmonary disease (COPD) while exploring how immune cell infiltration contributes towards COPD pathogenesis.MethodsThe GEO database provided two human COPD gene expression datasets (GSE38974 and GSE76925; n=134) along with the relevant controls (n=49) for differentially expressed gene (DEG) analyses. Candidate biomarkers were identified using the support vector machine recursive feature elimination (SVM-RFE) analysis and the LASSO regression model. The discriminatory ability was determined using the area under the receiver operating characteristic curve (AUC) values. These candidate biomarkers were characterized in the GSE106986 dataset (14 COPD patients and 5 controls) in terms of their respective diagnostic values and expression levels. The CIBERSORT program was used to estimate patterns of tissue infiltration of 22 types of immune cells. Furthermore, the in vivo and in vitro model of COPD was established using cigarette smoke extract (CSE) to validated the bioinformatics results.Results80 genes were identified via DEG analysis that were primarily involved in cellular amino acid and metabolic processes, regulation of telomerase activity and phagocytosis, antigen processing and MHC class I-mediated peptide antigen presentation, and other biological processes. LASSO and SVM-RFE were used to further characterize the candidate diagnostic markers for COPD, SLC27A3, and STAU1. SLC27A3 and STAU1 were found to be diagnostic markers of COPD in the metadata cohort (AUC=0.734, AUC=0.745). Their relevance in COPD were validated in the GSE106986 dataset (AUC=0.900 AUC=0.971). Subsequent analysis of immune cell infiltration discovered an association between SLC27A3 and STAU1 with resting NK cells, plasma cells, eosinophils, activated mast cells, memory B cells, CD8+, CD4+, and helper follicular T-cells. The expressions of SLC27A3 and STAU1 were upregulated in COPD models both in vivo and in vitro. Immune infiltration activation was observed in COPD models, accompanied by the enhanced expression of SLC27A3 and STAU1. Whereas, the knockdown of SLC27A3 or STAU1 attenuated the effect of CSE on BEAS-2B cells.ConclusionSTUA1 and SLC27A3 are valuable diagnostic biomarkers of COPD. COPD pathogenesis is heavily influenced by patterns of immune cell infiltration. This study provides a molecular biology insight into COPD occurrence and in exploring new therapeutic means useful in COPD.