scholarly journals A novel survival algorithm in COVID-19 intensive care patients: the classification and regression tree (CRT) method

2021 ◽  
Vol 21 (3) ◽  
pp. 1083-1092
Author(s):  
Sevinç Dağıstanlı ◽  
Süleyman Sönmez ◽  
Murat Ünsel ◽  
Emre Bozdağ ◽  
Ali Kocataş ◽  
...  
Keyword(s):  
Intensive Care ◽  
Decision Tree ◽  
Clinical Laboratory ◽  
Laboratory Data ◽  
Regression Tree ◽  
Classification And Regression Tree ◽  
Care Process ◽  
Intensive Care Patients ◽  
Training And Research ◽  
The City

Background/aim: The present study aimed to create a decision tree for the identification of clinical, laboratory and radio- logical data of individuals with COVID-19 diagnosis or suspicion of Covid-19 in the Intensive Care Units of a Training and Research Hospital of the Ministry of Health on the European side of the city of Istanbul. Materials and methods: The present study, which had a retrospective and sectional design, covered all the 97 patients treated with Covid-19 diagnosis or suspicion of COVID-19 in the intensive care unit between 12 March and 30 April 2020. In all cases who had symptoms admitted to the COVID-19 clinic, nasal swab samples were taken and thoracic CT was per- formed when considered necessary by the physician, radiological findings were interpreted, clinical and laboratory data were included to create the decision tree. Results: A total of 61 (21 women, 40 men) of the cases included in the study died, and 36 were discharged with a cure from the intensive care process. By using the decision tree algorithm created in this study, dead cases will be predicted at a rate of 95%, and those who survive will be predicted at a rate of 81%. The overall accuracy rate of the model was found at 90%. Conclusions: There were no differences in terms of gender between dead and live patients. Those who died were older, had lower MON, MPV, and had higher D-Dimer values than those who survived. Keywords: Survival algorithm; COVID-19 intensive care patients; CRT method.

scholarly journals Improvement of Adequate Digoxin Dosage: An Application of Machine Learning Approach

2018 ◽  
Vol 2018 ◽  
pp. 1-9 ◽  
Author(s):  
Ya-Han Hu ◽  
Chun-Tien Tai ◽  
Chih-Fong Tsai ◽  
Min-Wei Huang
Keyword(s):  
Machine Learning ◽  
Decision Tree ◽  
Prediction Models ◽  
Medical Center ◽  
Laboratory Data ◽  
Regression Tree ◽  
Digoxin Concentration ◽  
Classification And Regression Tree ◽  
Machine Learning Techniques ◽  
Learning Techniques

Digoxin is a high-alert medication because of its narrow therapeutic range and high drug-to-drug interactions (DDIs). Approximately 50% of digoxin toxicity cases are preventable, which motivated us to improve the treatment outcomes of digoxin. The objective of this study is to apply machine learning techniques to predict the appropriateness of initial digoxin dosage. A total of 307 inpatients who had their conditions treated with digoxin between 2004 and 2013 at a medical center in Taiwan were collected in the study. Ten independent variables, including demographic information, laboratory data, and whether the patients had CHF were also noted. A patient with serum digoxin concentration being controlled at 0.5–0.9 ng/mL after his/her initial digoxin dosage was defined as having an appropriate use of digoxin; otherwise, a patient was defined as having an inappropriate use of digoxin. Weka 3.7.3, an open source machine learning software, was adopted to develop prediction models. Six machine learning techniques were considered, including decision tree (C4.5), k-nearest neighbors (kNN), classification and regression tree (CART), randomForest (RF), multilayer perceptron (MLP), and logistic regression (LGR). In the non-DDI group, the area under ROC curve (AUC) of RF (0.912) was excellent, followed by that of MLP (0.813), CART (0.791), and C4.5 (0.784); the remaining classifiers performed poorly. For the DDI group, the AUC of RF (0.892) was the best, followed by CART (0.795), MLP (0.777), and C4.5 (0.774); the other classifiers’ performances were less than ideal. The decision tree-based approaches and MLP exhibited markedly superior accuracy performance, regardless of DDI status. Although digoxin is a high-alert medication, its initial dose can be accurately determined by using data mining techniques such as decision tree-based and MLP approaches. Developing a dosage decision support system may serve as a supplementary tool for clinicians and also increase drug safety in clinical practice.

Patient Risk Factor Profiles Associated With the Timing of Goals-of-Care Consultation Before Death: A Classification and Regression Tree Analysis

2020 ◽  
Vol 37 (10) ◽  
pp. 767-778
Author(s):  
Lauren T. Starr ◽  
Connie M. Ulrich ◽  
Paul Junker ◽  
Liming Huang ◽  
Nina R. O’Connor ◽  
...  
Keyword(s):  
Risk Factor ◽  
Intensive Care ◽  
Medical Center ◽  
Regression Tree ◽  
Illness Severity ◽  
Classification And Regression Tree ◽  
Goals Of Care ◽  
Intensive Care Patients ◽  
Care Consultation ◽  
Classification And Regression

Background: Early palliative care consultation (“PCC”) to discuss goals-of-care benefits seriously ill patients. Risk factor profiles associated with the timing of conversations in hospitals, where late conversations most likely occur, are needed. Objective: To identify risk factor patient profiles associated with PCC timing before death. Methods: Secondary analysis of an observational study was conducted at an urban, academic medical center. Patients aged 18 years and older admitted to the medical center, who had PCC, and died July 1, 2014 to October 31, 2016, were included. Patients admitted for childbirth or rehabilitationand patients whose date of death was unknown were excluded. Classification and Regression Tree modeling was employed using demographic and clinical variables. Results: Of 1141 patients, 54% had PCC “close to death” (0-14 days before death); 26% had PCC 15 to 60 days before death; 21% had PCC >60 days before death (median 13 days before death). Variables associated with receiving PCC close to death included being Hispanic or “Other” race/ethnicity intensive care patients with extreme illness severity (85%), with age <46 or >75 increasing this probability (98%). Intensive care patients with extreme illness severity were also likely to receive PCC close to death (64%) as were 50% of intensive care patients with less than extreme illness severity. Conclusions: A majority of patients received PCC close to death. A complex set of variable interactions were associated with PCC timing. A systematic process for engaging patients with PCC earlier in the care continuum, and in intensive care regardless of illness severity, is needed.

scholarly journals Analisis Kecenderungan Keterlambatan Pembayaran Pengecekan Kapal di Pelabuhan Regional Riau

2019 ◽  
Vol 2 (2) ◽  
pp. 92-98
Author(s):  
Hespri Yomeldi ◽  
Moh Roufiq Azmy ◽  
Ryche Pranita
Keyword(s):  
Logistic Regression ◽  
Decision Tree ◽  
Regression Tree ◽  
Classification And Regression Tree ◽  
Ministry Of Health ◽  
Tree Algorithm ◽  
Health Checks ◽  
Depth Study ◽  
Classification And Regression ◽  
Port Health

Ship health checks must be carried out which function to provide a sailing permit. The implementation of ship health checks is carried out in collaboration with the ministry of health and transportation. The implementation of the activity, commonly known as Port Health Quarantine Clearance (PHQC) requires time to check and the ship makes a payment check to be able to issue a sailing permit. The problem that arises in the field is that the ship delays PHQC payments and then impacts on the buildup of ships in the port, besides that officers also need longer time to process the issuance of sailing permits. This of course has an impact on other port services such as dwelling time and scheduled departures that can be delayed. In overcoming this problem, an in-depth study is needed to analyze the trend of late payment of ship health checks, what variables influence it and how treatment is done to overcome these problems. Using logistic regression and decision tree with Classification and Regression Tree algorithm , a model is then developed that determines the variables that affect the delay of the ship making PHQC payments.

scholarly journals Assessing the influence of environmental context on responses to noise exposure in the city of Toronto

2021 ◽  
Author(s):  
Desislava Stevanova
Keyword(s):  
Spatial Data ◽  
Green Space ◽  
Noise Exposure ◽  
Regression Tree ◽  
Environmental Context ◽  
Classification And Regression Tree ◽  
Bivariate Analysis ◽  
Night Time ◽  
Cart Analysis ◽  
The City

This thesis examines individual and community noise perception of environmental noise in three neighbourhoods in the city of Toronto. The significance of this research is based on a relative absence of literature on how noise sensitivity and annoyance are affected by non-acoustic factors such as the built environment, demographic, and socio-economic factors. Data from a neighbourhood noise survey (n=552) were combined with spatial data on exposures to noise. Bivariate analysis, multivariate regression, and classification and regression tree (CART) analysis were used. The results showed that participants in Downtown and Don Valley have similar noise responses (64% and 67% high annoyance) despite differences in noise exposure (LAeq 24h: 66.8 and 59.3). Estimation of Community Tolerance Levels (CTL) confirmed that participants exposed to lower sound levels have a lower tolerance of noise. Further results showed that a neighbourhood with high socioeconomic status and access to green space, and relatively low night time noise levels were still two times more likely to report high annoyance, compared with neighbourhood with moderate socio-economic status and lower access to green space. The results suggest that environmental context influences expectations and sensitivity to noise.

scholarly journals Detection of Subclinical Keratoconus Using a Novel Combined Tomographic and Biomechanical Model Based on an Automated Decision Tree

2021 ◽  
Author(s):  
Peng Song ◽  
Shengwei Ren ◽  
Yu Liu ◽  
Pei Li ◽  
Qingyan Zeng
Keyword(s):  
Decision Tree ◽  
Amplitude Ratio ◽  
Regression Tree ◽  
Biomechanical Model ◽  
Study Data ◽  
Classification And Regression Tree ◽  
Stiffness Parameter ◽  
Chi Square ◽  
Interaction Detection ◽  
Cart Model

Abstract The aim of this study was to develop a predictive model for subclinical keratoconus (SKC) based on decision tree (DT) algorithms. A total of 194 eyes (including 105 normal eyes and 89 SKC) were included in the double-center retrospective study. Data were separately used for training and validation databases. The baseline variables were derived from tomography and biomechanical imaging. DT models were generated in the training database using Chi-square automatic interaction detection (CHAID) and classification and regression tree (CART) algorithms. The discriminating rules of the CART model selected variables of the Belin/Ambrósio deviation (BAD-D), stiffness parameter at first applanation (SPA1), back eccentricity (Becc), and maximum pachymetric progression index in order, while the CHAID model selected BAD-D, deformation amplitude ratio, SPA1, and Becc. The CART model allowed discrimination between normal and SKC eyes with 92.2% accuracy, which was higher than that of the CHAID model (88.3%), BAD-D (82.0%), Corvis biomechanical index (CBI, 77.3%), and tomographic and biomechanical index (TBI, 78.1%). The discriminating performance of the CART model was validated with 92.4% accuracy, while the CHAID model was validated with 86.4% accuracy in the validation database. Thus, the CART model using tomography and biomechanical imaging was an excellent model for SKC screening and provided easy-to-understand discriminating rules.

scholarly journals Beef Tenderness Prediction by a Combination of Statistical Methods: Chemometrics and Supervised Learning to Manage Integrative Farm-To-Meat Continuum Data

Foods ◽  
2019 ◽  
Vol 8 (7) ◽  
pp. 274 ◽  
Author(s):  
Mohammed Gagaoua ◽  
Valérie Monteils ◽  
Sébastien Couvreur ◽  
Brigitte Picard
Keyword(s):  
Decision Tree ◽  
Supervised Learning ◽  
Predictive Accuracy ◽  
Regression Tree ◽  
Classification And Regression Tree ◽  
Beef Tenderness ◽  
Interaction Detection ◽  
Fiber Area ◽  
Total Collagen ◽  
Protected Designation Of Origin

This trial aimed to integrate metadata that spread over farm-to-fork continuum of 110 Protected Designation of Origin (PDO)Maine-Anjou cows and combine two statistical approaches that are chemometrics and supervised learning; to identify the potential predictors of beef tenderness analyzed using the instrumental Warner-Bratzler Shear force (WBSF). Accordingly, 60 variables including WBSF and belonging to 4 levels of the continuum that are farm-slaughterhouse-muscle-meat were analyzed by Partial Least Squares (PLS) and three decision tree methods (C&RT: classification and regression tree; QUEST: quick, unbiased, efficient regression tree and CHAID: Chi-squared Automatic Interaction Detection) to select the driving factors of beef tenderness and propose predictive decision tools. The former method retained 24 variables from 59 to explain 75% of WBSF. Among the 24 variables, six were from farm level, four from slaughterhouse level, 11 were from muscle level which are mostly protein biomarkers, and three were from meat level. The decision trees applied on the variables retained by the PLS model, allowed identifying three WBSF classes (Tender (WBSF ≤ 40 N/cm2), Medium (40 N/cm2 < WBSF < 45 N/cm2), and Tough (WBSF ≥ 45 N/cm2)) using CHAID as the best decision tree method. The resultant model yielded an overall predictive accuracy of 69.4% by five splitting variables (total collagen, µ-calpain, fiber area, age of weaning and ultimate pH). Therefore, two decision model rules allow achieving tender meat on PDO Maine-Anjou cows: (i) IF (total collagen < 3.6 μg OH-proline/mg) AND (µ-calpain ≥ 169 arbitrary units (AU)) AND (ultimate pH < 5.55) THEN meat was very tender (mean WBSF values = 36.2 N/cm2, n = 12); or (ii) IF (total collagen < 3.6 μg OH-proline/mg) AND (µ-calpain < 169 AU) AND (age of weaning < 7.75 months) AND (fiber area < 3100 µm2) THEN meat was tender (mean WBSF values = 39.4 N/cm2, n = 30).

Using machine learning techniques to develop prediction models for detecting unpaid credit card customers

2020 ◽  
Vol 39 (5) ◽  
pp. 6073-6087
Author(s):  
Meltem Yontar ◽  
Özge Hüsniye Namli ◽  
Seda Yanik
Keyword(s):  
Decision Tree ◽  
Credit Card ◽  
Banking Sector ◽  
Performance Metrics ◽  
Prediction Models ◽  
Regression Tree ◽  
Classification And Regression Tree ◽  
Machine Learning Techniques ◽  
Support Vector ◽  
Cart Algorithm

Customer behavior prediction is gaining more importance in the banking sector like in any other sector recently. This study aims to propose a model to predict whether credit card users will pay their debts or not. Using the proposed model, potential unpaid risks can be predicted and necessary actions can be taken in time. For the prediction of customers’ payment status of next months, we use Artificial Neural Network (ANN), Support Vector Machine (SVM), Classification and Regression Tree (CART) and C4.5, which are widely used artificial intelligence and decision tree algorithms. Our dataset includes 10713 customer’s records obtained from a well-known bank in Taiwan. These records consist of customer information such as the amount of credit, gender, education level, marital status, age, past payment records, invoice amount and amount of credit card payments. We apply cross validation and hold-out methods to divide our dataset into two parts as training and test sets. Then we evaluate the algorithms with the proposed performance metrics. We also optimize the parameters of the algorithms to improve the performance of prediction. The results show that the model built with the CART algorithm, one of the decision tree algorithm, provides high accuracy (about 86%) to predict the customers’ payment status for next month. When the algorithm parameters are optimized, classification accuracy and performance are increased.

Enhancement of diuresis with metolazone in infant paediatric cardiac intensive care patients

2017 ◽  
Vol 28 (1) ◽  
pp. 27-31 ◽  
Author(s):  
Russell T. Wise ◽  
Brady S. Moffett ◽  
Ayse Akcan-Arikan ◽  
Marianne Galati ◽  
Natasha Afonso ◽  
...  
Keyword(s):  
Intensive Care ◽  
Urine Output ◽  
Multivariable Analysis ◽  
Laboratory Data ◽  
Demographic Variables ◽  
Intensive Care Patients ◽  
Select Group ◽  
Cardiac Intensive Care ◽  
Few Data ◽  
Selection Of

AbstractBackgroundFew data are available regarding the use of metolazone in infants in cardiac intensive care. Researchers need to carry out further evaluation to characterise the effects of this treatment in this population.MethodsThis is a descriptive, retrospective study carried out in patients less than a year old. These infants had received metolazone over a 2-year period in the paediatric cardiac intensive care unit at our institution. The primary goal was to measure the change in urine output from 24 hours before the start of metolazone therapy to 24 hours after. Patient demographic variables, laboratory data, and fluid-balance data were analysed.ResultsThe study identified 97 infants with a mean age of 0.32±0.25 years. Their mean weight was 4.9±1.5 kg, and 58% of the participants were male. An overall 63% of them had undergone cardiovascular surgery. The baseline estimated creatinine clearance was 93±37 ml/minute/1.73 m2. Initially, the participants had received a metolazone dose of 0.27±0.10 mg/kg/day, the maximum dose being 0.43 mg/kg/day. They had also received other diuretics during metolazone initiation, such as furosemide (87.6%), spironolactone (58.8%), acetazolamide (11.3%), bumetanide (7.2%), and ethacrynic acid (1%). The median change in urine output after metolazone was 0.9 ml/kg/hour (interquartile range 0.15–1.9). The study categorised a total of 66 patients (68.0%) as responders. Multivariable analysis identified acetazolamide use (p=0.002) and increased fluid input in the 24 hours after metolazone initiation (p<0.001) as being significant for increased urine output. Changes in urine output were not associated with the dose of metolazone (p>0.05).ConclusionsMetolazone increased urine output in a select group of patients. Efficacy can be maximised by strategic selection of patients.

scholarly journals KEY FACTORS AFFECTING RAIL SERVICE QUALITY IN THE NORTHERN ITALY: A DECISION TREE APPROACH

Transport ◽  
2014 ◽  
Vol 29 (1) ◽  
pp. 75-83 ◽  
Author(s):  
Rocío De Oña ◽  
Laura Eboli ◽  
Gabriella Mazzulla
Keyword(s):  
Service Quality ◽  
Regression Tree ◽  
Northern Italy ◽  
Classification And Regression Tree ◽  
Key Factors ◽  
Factors Affecting ◽  
Service Characteristics ◽  
Transit Services ◽  
The City ◽  
Tree Approach

This work concerns with the analysis of transit service quality on the basis of the perceptions directly expressed by the passengers of the services. The transit services supporting the research are offered by rail operators of the Northern Italy, and particularly by regional and suburban lines connecting different towns of the hinterland of the city of Milan, and express lines connecting Milan with the Malpensa airport. The experimental data were collected in a survey conducted in May 2012, and addressed to a sample of more than 16,000 passengers. Passengers expressed their opinions about service characteristics such as safety, cleanliness, comfort, information, personnel. The tool chosen for evaluating service quality is a Classification and Regression Tree Approach (CART), useful for identifying the characteristics mostly influencing the overall service quality. We found that service characteristics like ‘Windows and Doors Working’, ‘Courtesy and Competence on Board’, ‘Information at Stations’, ‘Punctuality of Runs’, ‘Courtesy and Competence in Station’ and ‘Regularity of Runs’ mainly influence service quality.

scholarly journals Classification of Fatigued and Drunk Driving Based on Decision Tree Methods: A Simulator Study

2019 ◽  
Vol 16 (11) ◽  
pp. 1935 ◽  
Author(s):  
Ying Yao ◽  
Xiaohua Zhao ◽  
Hongji Du ◽  
Yunlong Zhang ◽  
Guohui Zhang ◽  
...  
Keyword(s):  
Decision Tree ◽  
Regression Tree ◽  
Drunk Driving ◽  
Driving Performance ◽  
Classification And Regression Tree ◽  
Vehicle Speed ◽  
Road Geometry ◽  
Lane Position ◽  
Straight Segments

It is a commonly known fact that both alcohol and fatigue impair driving performance. Therefore, the identification of fatigue and drinking status is very important. In this study, each of the 22 participants finished five driving tests in total. The control condition, serving as the benchmark in the five driving tests, refers to alert driving. The other four test conditions include driving with three blood alcohol content (BAC) levels (0.02%, 0.05%, and 0.08%) and driving in a fatigued state. The driving scenario included straight and curved roads. The straight roads connected the curved ones with radii of 200 m, 500 m, and 800 m with two turning directions (left and right). Driving performance indicators such as the average and standard deviation of longitudinal speed and lane position were selected to identify drunk driving and fatigued driving. In the process of identification, road geometry (straight segments, radius, and direction of curves) was also taken into account. Alert vs. abnormal and fatigued vs. drunk driving with various BAC levels were analyzed separately using the Classification and Regression Tree (CART) model, and the significance of the variables on the binary response variable was determined. The results showed that the decision tree could be used to distinguish normal driving from abnormal driving, fatigued driving, and drunk driving based on the indexes of vehicle speed and lane position at curves with different radii. The overall accuracy of classification of “alert” and “abnormal” driving was 90.9%, and that of “fatigued” and “drunk” driving was 94.4%. The accuracy was relatively low in identifying different BAC degrees. This experiment is designed to provide a reference for detecting dangerous driving states.

Sign in / Sign up

Export Citation Format

Share Document