國立虎尾科技大學 |

Predicting 30-Day Diabetic Patient Readmissions via Heterogeneous Ensemble Feature Selection.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Predicting 30-Day Diabetic Patient Readmissions via Heterogeneous Ensemble Feature Selection./
作者:	Tipu, Amina.
面頁冊數:	1 online resource (221 pages)
附註:	Source: Masters Abstracts International, Volume: 84-04.
Contained By:	Masters Abstracts International84-04.
標題:	Systems science. -
電子資源:	click for full text (PQDT)
ISBN:	9798351457024

Predicting 30-Day Diabetic Patient Readmissions via Heterogeneous Ensemble Feature Selection.
Tipu, Amina.

Predicting 30-Day Diabetic Patient Readmissions via Heterogeneous Ensemble Feature Selection. - 1 online resource (221 pages)

Source: Masters Abstracts International, Volume: 84-04.

Thesis (M.S.)--State University of New York at Binghamton, 2022.

Includes bibliographical references

This research study focuses on predicting 30-day hospital readmissions for diabetic patients via the knowledge discovery process and a heterogeneous ensemble feature selection algorithm. Patient readmissions are a huge financial cost to hospitals and health care systems at large and are linked to poorer patient health outcomes. The Centers for Medicare and Medicaid recent regulation surrounding financial penalizations towards hospitals with considerable amounts of patient readmissions for certain disease states as well as avoidable readmissions has brought readmissions to the forefront of health services research. This study aims to identify factors that are linked to patient readmissions via the heterogeneous ensemble feature selection method as well as predict patients most likely to be readmitted. To examine diabetic patient readmissions a publicly available dataset from the UCI Machine Learning Repository was used. The dataset contains patient information from 130 hospitals across the United States from 1999-2008. The research methodology was carried out via three unique implementations, in the primary no feature selection or tuning is carried out, and in the second three feature selection techniques: recursive feature elimination, random forest, and mutual information are applied with their results averaged. Lastly, the HEFS algorithm was applied to the dataset in the third implementation. Ten prediction algorithms are employed: Logistic Regression, Gaussian Naive Bayes, Decision Tree, Random Forest, K-Nearest Neighbors Support Vector Machine, Bagging, Adaptive Boosting, Gradient Boosting, and Multi-Layer Perceptron. The performance metric results show an improvement is achieved based on the HEFS algorithm in the case of LR, RF, KNN, SVM, BG, ADA, GB, and MLP in comparison to LR and GNB in differentiating between patients likely to be readmitted and not. The research output will act as a decision support tool for medical health professionals and executives in allocating and assigning resources to relevant patients regarding readmissions.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798351457024Subjects--Topical Terms:

1148479
Systems science.
Subjects--Index Terms:

Heterogeneous ensemble feature selectionIndex Terms--Genre/Form:

554714
Electronic books.

Predicting 30-Day Diabetic Patient Readmissions via Heterogeneous Ensemble Feature Selection.
LDR:03395ntm a22003497 4500 001 1144233
005 20240531083806.5
006 m o d
007 cr mn ---uuuuu
008 250605s2022 xx obm 000 0 eng d
020 $a 9798351457024
035 $a (MiAaPQ)AAI29328102
035 $a AAI29328102
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Tipu, Amina. $3 1469195
245 1 0 $a Predicting 30-Day Diabetic Patient Readmissions via Heterogeneous Ensemble Feature Selection.
264 0 $c 2022
300 $a 1 online resource (221 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Masters Abstracts International, Volume: 84-04.
500 $a Advisor: Wang, Yong.
502 $a Thesis (M.S.)--State University of New York at Binghamton, 2022.
504 $a Includes bibliographical references
520 $a This research study focuses on predicting 30-day hospital readmissions for diabetic patients via the knowledge discovery process and a heterogeneous ensemble feature selection algorithm. Patient readmissions are a huge financial cost to hospitals and health care systems at large and are linked to poorer patient health outcomes. The Centers for Medicare and Medicaid recent regulation surrounding financial penalizations towards hospitals with considerable amounts of patient readmissions for certain disease states as well as avoidable readmissions has brought readmissions to the forefront of health services research. This study aims to identify factors that are linked to patient readmissions via the heterogeneous ensemble feature selection method as well as predict patients most likely to be readmitted. To examine diabetic patient readmissions a publicly available dataset from the UCI Machine Learning Repository was used. The dataset contains patient information from 130 hospitals across the United States from 1999-2008. The research methodology was carried out via three unique implementations, in the primary no feature selection or tuning is carried out, and in the second three feature selection techniques: recursive feature elimination, random forest, and mutual information are applied with their results averaged. Lastly, the HEFS algorithm was applied to the dataset in the third implementation. Ten prediction algorithms are employed: Logistic Regression, Gaussian Naive Bayes, Decision Tree, Random Forest, K-Nearest Neighbors Support Vector Machine, Bagging, Adaptive Boosting, Gradient Boosting, and Multi-Layer Perceptron. The performance metric results show an improvement is achieved based on the HEFS algorithm in the case of LR, RF, KNN, SVM, BG, ADA, GB, and MLP in comparison to LR and GNB in differentiating between patients likely to be readmitted and not. The research output will act as a decision support tool for medical health professionals and executives in allocating and assigning resources to relevant patients regarding readmissions.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Systems science. $3 1148479
653 $a Heterogeneous ensemble feature selection
653 $a Diabetic patient readmissions
653 $a Financial penalizations
655 7 $a Electronic books. $2 local $3 554714
690 $a 0790
710 2 $a State University of New York at Binghamton. $b Systems Science Industrial Engineering. $3 1182293
710 2 $a ProQuest Information and Learning Co. $3 1178819
773 0 $t Masters Abstracts International $g 84-04.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29328102 $z click for full text (PQDT)