Classification and regression tree analysis in acute coronary syndrome patients

Tung, Heng Hsin; Chen, Chiang-Yi; Lin, Kuan‐Chia; Chou, Nai‐Kuan; Lee, Jyun-Yi; Clinciu, Daniel L.; Lien, Ru-Yu

doi:10.4236/wjcd.2012.23030

Cited by 7 publications

(5 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Classification and Regression Tree (CART) [ 14 ], unlike ID3 algorithm, uses Gini coefficient (Classification Tree) and squared error (Regression Tree) as the basis for the optimal feature delineation, and the complete algorithm includes decision tree pruning in addition to feature selection and segmentation of feature values. CART decision trees utilize the Gini coefficient to classify features, where the Gini value measures the uncertainty of a dataset by indicating the probability of two randomly drawn samples having different categories.…”

Section: Concepts and Methodsmentioning

confidence: 99%

Comparative analysis of supervised learning algorithms for prediction of cardiovascular diseases

Dou,

Liu,

Meng

et al. 2024

THC

View full text Add to dashboard Cite

BACKGROUND: With the advent of artificial intelligence technology, machine learning algorithms have been widely used in the area of disease prediction. OBJECTIVE: Cardiovascular disease (CVD) seriously jeopardizes human health worldwide, thereby needing the establishment of an effective CVD prediction model that can be of great significance for controlling the risk of the disease and safeguarding the physical and mental health of the population. METHODS: Considering the UCI heart disease dataset as an example, initially, a single machine learning prediction model was constructed. Subsequently, six methods such as Pearson, chi-squared, RFE and LightGBM were comprehensively used for the feature screening. On the basis of the base classifiers, Soft Voting fusion and Stacking fusion was carried out to build a prediction model for cardiovascular diseases, in order to realize an early warning and disease intervention for high-risk populations. To address the data imbalance problem, the SMOTE method was adopted to process the data set, and the prediction effect of the model was analyzed using multi-dimensional and multi-indicators. RESULTS: In the single classifier model, the MLP algorithm performed optimally on the preprocessed heart disease dataset. After feature selection, five features eliminated. The ENSEM_SV algorithm that combines the base classifiers to determine the prediction results by soft voting on the results of the classifiers achieved the optimal value on five metrics such as Accuracy, Jaccard_Score, Hamm_Loss, AUC, etc., and the AUC value reached 0.951. The RF, ET, GBDT, and LGB algorithms were employed in the first stage sub-model composed of base classifiers. The AB algorithm was selected as the second stage model, and the ensemble algorithm ENSEM_ST, obtained by Stacking fusion of the two stages exhibited the best performance on 7 indicators such as Accuracy, Sensitivity, F1_Score, Mathew_Corrcoef, etc., and the AUC reached 0.952. Furthermore, a comparison of the algorithms’ classification effects based on different training set occupancy was carried out. The results indicated that the prediction performance of both the fusion models was better than the single models, and the overall effect of ENSEM_ST fusion was stronger than the ENSEM_SV fusion. CONCLUSIONS: The fusion model established in this study improved the overall classification accuracy and stability of the model to a significant extent. It has a good application value in the predictive analysis of CVD diagnosis, and can provide a valuable reference in the disease diagnosis and intervention strategies.

show abstract

Section: Concepts and Methodsmentioning

confidence: 99%

Comparative analysis of supervised learning algorithms for prediction of cardiovascular diseases

Dou,

Liu,

Meng

et al. 2024

THC

View full text Add to dashboard Cite

show abstract

“…A Classification and Regression Tree (CART) model (based on an iterative algorithm) was used to estimate the covariates associated with SST false positivity. CART is a nonparametric multiple regression statistical model to perform supervised classifications with regard to explanatory variables compared with a categorical variable [ 28 , 29 ]. The modeling is carried out in 3 stages: (1) identification of the covariates associated with the response variable, (2) classification of each covariate to discriminate the response variable into 2 distinct groups, and (3) iterative repetition of the 2 previous steps until it is no longer possible to perform segmentation [ 30 ].…”

Section: Methodsmentioning

confidence: 99%

Unusual Localization of Blood-Borne Loa loa Microfilariae in the Skin Depends on Microfilarial Density in the Blood: Implications for Onchocerciasis Diagnosis in Coendemic Areas

Niamsi-Emalio¹,

Nana-Djeunga²,

Chesnais

et al. 2021

Clinical Infectious Diseases

View full text Add to dashboard Cite

Background The diagnostic gold standard for onchocerciasis relies on identification and enumeration of (skin-dwelling) Onchocerca volvulus microfilariae (mf) using the skin snip technique (SST). In a recent study, blood-borne Loa loa mf were found by SST in individuals heavily infected with L. loa, and microscopically misidentified as O. volvulus due to their superficially similar morphology. This study investigates the relationship between L. loa microfilarial density (Loa MFD) and the probability of testing SST positive. Methods A total of 1053 participants from the (onchocerciasis and loiasis co-endemic) East Region in Cameroon were tested for: i) Loa MFD in blood samples; ii) O. volvulus presence by SST, and iii) Ig[immunoglobulin]G4 antibody positivity to Ov16 by rapid diagnostic test (RDT). A Classification and Regression Tree (CART) model was used to perform a supervised classification of SST status and identify a Loa MFD threshold above which it is highly likely to find L. loa mf in skin snips. Results Of 1011 Ov16-negative individuals, 28 (2.8%) tested SST positive and 150 (14.8%) were L. loa positive. The range of Loa MFD was 0–85200mf/mL. The CART model subdivided the sample into two Loa MFD classes with a discrimination threshold of 4080 (95% CI: 2180–12240) mf/mL. The probability of being SST positive exceeded 27% when Loa MFD was >4080mf/mL. Conclusions The probability of finding L. loa mf by SST increases significantly with Loa MFD. Skin-snip polymerase chain reaction (PCR) would be useful when monitoring onchocerciasis prevalence by SST in onchocerciasis–loiasis co-endemic areas.

show abstract

“…The Classification and Regression Tree (CART) was used to assess the association between socio-demographic characteristics and respondents’ knowledge, attitude and practice scores for each of the targeted diseases. Indeed, CART is a non-parametric multiple regression approach that both avoids multicollinearity issues and explains a categorical dependent variable by defining groups of subjects with similar behaviors [15–16], while taking into account all interactions between different covariates [17]. CART then evaluates all the possible thresholds and separates the dependent variable into two groups, the procedure being repeated recursively until an optimal criterion is obtained [18].…”

Section: Methodsmentioning

confidence: 99%

An integrated approach to assess Knowledge, Attitude and Practices (KAP) regarding major Neglected Tropical Diseases endemic in the Mbengwi health district (North West Region, Cameroon)

Sumo

Lenou-Nanga²,

Ntonifor

et al. 2019

Preprint

View full text Add to dashboard Cite

2Background: Preventive chemotherapy (PCT) is the main strategy currently used to 3 control and/or eliminate onchocerciasis (Oncho), lymphatic filariasis (LF) and Soil 4 Transmitted Helminthiasis (STH), and community participation (through 5 implementation of MDA or adherence to PCT) is critical to achieve this goal. However, 6 these Neglected Tropical Diseases (NTDs) are still persisting in most endemic areas as a 7 consequence of sub-optimal treatment coverage, the presence of systematic non-8 compliers in communities ... This study aimed at investigating whether the knowledge, 9 attitudes and practices of populations about these NTDs can explain the poor trends 10 towards elimination. 11 12 Methodology: A cross-sectional survey was carried out in the Mbengwi Health District 13 (North West Region, Cameroon) using the cluster sampling technique. Clusters were 14 selected using the Probability Proportionate to Estimate Size strategy. In each cluster, 15 the random walk technique was used for the selection of households, and a structure 16 questionnaire was administered to 2-3 of its members. 17 18 Principal Findings: A total of 254 households from 26 clusters were visited, and 514 19 individuals were interviewed. The sex ratio of interviewees (1.08) was unbiased, and 20 their ages ranged between 10 and 99 years old. Though most of the respondents 21 declared having already heard of these NTDs (41.2%, 73.7% and 89.9% for Oncho, LF 22 and STH respectively), only a minority of them were aware of correct response of how 23 they are acquired/transmitted (3.7%, 6.8% and 12.5% for Oncho, LF and STH 24 respectively), and prevented (23.1%, 18.9% and 47.2% for Oncho, LF and STH 25respectively). Even when respondents were aware that medicines were useful to 3 1 prevent and/or treat these NTDs, almost none of them knew the drug used or the 2 treatment frequency. 3 4 Conclusion/Significance: This study reveals that interviewees exhibit poor knowledge, 5 attitudes and practices as regards to these NTDs, although they are endemic in the study 6 area and PCTs given yearly since a while. These misconceptions can seriously affect the 7 adherence and contribution of populations to the success of PCTs, and it appears 8 compulsory to improve individual knowledge, with a focus on the importance and 9 rationale behind MDA, to optimize their attitudes and practices, especially community 10 participation to PCTs. Mbengwi health district, Cameroon 14 15 16 Author summary 17 The control and/or elimination of Neglected Tropical Diseases (NTDs) are currently on 18 top of the agenda of endemic countries' control programs and stakeholders. Ivermectin-19 and Albendazole/Mebendazole-based mass distribution is currently the main strategy to 20 control/interrupt transmission of onchocerciasis, lymphatic filariasis and soil-21 transmitted helminthiasis, and adherence of communities is compulsory for the success 22 of this approach. Despite the success registered in the fight against these diseases, the 23 trend towards elimination remains unsatisfa...

show abstract

Classification and regression tree analysis in acute coronary syndrome patients

Cited by 7 publications

References 30 publications

Comparative analysis of supervised learning algorithms for prediction of cardiovascular diseases

Comparative analysis of supervised learning algorithms for prediction of cardiovascular diseases

Unusual Localization of Blood-Borne Loa loa Microfilariae in the Skin Depends on Microfilarial Density in the Blood: Implications for Onchocerciasis Diagnosis in Coendemic Areas

An integrated approach to assess Knowledge, Attitude and Practices (KAP) regarding major Neglected Tropical Diseases endemic in the Mbengwi health district (North West Region, Cameroon)

Contact Info

Product

Resources

About