東華大學圖書館 |

Improving Patient Outcomes Through Data Driven Medical Decision Making.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Improving Patient Outcomes Through Data Driven Medical Decision Making./
作者:	Swan, Breanna Patrice.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2021,
面頁冊數:	271 p.
附註:	Source: Dissertations Abstracts International, Volume: 83-02, Section: B.
Contained By:	Dissertations Abstracts International83-02B.
標題:	Diabetes. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28552538
ISBN:	9798522941734

Improving Patient Outcomes Through Data Driven Medical Decision Making.
Swan, Breanna Patrice.

Improving Patient Outcomes Through Data Driven Medical Decision Making. - Ann Arbor : ProQuest Dissertations & Theses, 2021 - 271 p.

Source: Dissertations Abstracts International, Volume: 83-02, Section: B.

Thesis (Ph.D.)--North Carolina State University, 2021.

This item must not be sold to any third party vendors.

Medical decision-making is reliant on the power of data. Physicians, health insurers, and epidemiologists, among other stakeholders, are actively seeking out ways to engage with data and embrace the shift towards personalized, preventative, predictive, and participatory (P4) medicine. Health datasets continue to grow in number, availability, and size, where the volume, veracity, variety, and value can drastically change from one dataset to the next. Shifting data distributions, like population health, further compound the complexity and uncertainty around data-driven medical decisions. This dissertation develops frameworks that integrate methodologies from industrial engineering, statistics, epidemiology, and bioinformatics to predict risk and recommended care interventions. Our frameworks embrace different datasets' unique characteristics and prioritize stakeholder engagement to develop data-driven decision support tools.First, we focus on the predictive and personalized arms of P4 medicine. There are hundreds of health-related machine learning (ML) models in the literature and an increasing number and availability of open-source ML methodologies. Selecting the 'best' prediction model for a given disease, patient population, and clinical practice is challenging. To support this decision process, we developed the Selection of Machine-learning Algorithms with ReplicaTions (SMART) Framework that integrates building and selecting ML models with decision theory. We extract theoretical insights while illustrating the power and flexibility of the SMART Framework to predict patients at high risk of complications due to diabetes using two data sources. The datasets are unique in volume and variety, each requiring a unique set of analytical methods for building viable prediction models. We outline opportunities for stakeholder engagement throughout the SMART Framework to empower medical decision-makers and build their confidence in leveraging ML models to enhance patient care. We demonstrate that combining ML and decision theory incorporates uncertainty and stakeholder engagement, resulting in more robust data-driven medical decision-making support tools.Next, we focus on the preventative and personalized arms of P4 medicine in two clinical operations case studies related to diabetic retinopathy (DR) and an emergency department (ED). To prevent blindness due to DR, we developed a microsimulation model of diabetes progression and patient's interaction with the care system. We amplify the effects of care interventions, tested in a small clinical trial, to population-level impacts using patient health outcomes, adherence behavior, and cumulative totals of visits. To increase patient flow through an ED, we develop an optimization model and a discrete event simulation model to compare alternate staffing schedules to a team-based care staffing model. Complex interactions between workload variability, uncertain and increasing arrival rates, and resource constraints make it difficult to improve flow and reduce crowding in the ED. We merged patient information and operations variables to give a more holistic view of the patient experience. In both case studies, we continuously iterate our data interpretation, analyses, and models with stakeholders. As a result, we establish data-driven recommendations for increasing the number of patients treated for DR, thus reducing the incidence of blindness, and for efficiently matching ED staffing to patient demand.The volume, variety, and veracity of datasets in each chapter were unique, yet, combined with stakeholders' participation, we built models that transformed data into medical decision-making support tools.

ISBN: 9798522941734Subjects--Topical Terms:

544344
Diabetes.

Improving Patient Outcomes Through Data Driven Medical Decision Making.
LDR:04743nmm a2200313 4500 001 2347489
005 20220801062215.5
008 241004s2021 ||||||||||||||||| ||eng d
020 $a 9798522941734
035 $a (MiAaPQ)AAI28552538
035 $a (MiAaPQ)NCState_Univ18402038610
035 $a AAI28552538
040 $a MiAaPQ $c MiAaPQ
100 1 $a Swan, Breanna Patrice. $3 3686748
245 1 0 $a Improving Patient Outcomes Through Data Driven Medical Decision Making.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2021
300 $a 271 p.
500 $a Source: Dissertations Abstracts International, Volume: 83-02, Section: B.
500 $a Advisor: Mayorga, Maria;Ivy, Julie.
502 $a Thesis (Ph.D.)--North Carolina State University, 2021.
506 $a This item must not be sold to any third party vendors.
520 $a Medical decision-making is reliant on the power of data. Physicians, health insurers, and epidemiologists, among other stakeholders, are actively seeking out ways to engage with data and embrace the shift towards personalized, preventative, predictive, and participatory (P4) medicine. Health datasets continue to grow in number, availability, and size, where the volume, veracity, variety, and value can drastically change from one dataset to the next. Shifting data distributions, like population health, further compound the complexity and uncertainty around data-driven medical decisions. This dissertation develops frameworks that integrate methodologies from industrial engineering, statistics, epidemiology, and bioinformatics to predict risk and recommended care interventions. Our frameworks embrace different datasets' unique characteristics and prioritize stakeholder engagement to develop data-driven decision support tools.First, we focus on the predictive and personalized arms of P4 medicine. There are hundreds of health-related machine learning (ML) models in the literature and an increasing number and availability of open-source ML methodologies. Selecting the 'best' prediction model for a given disease, patient population, and clinical practice is challenging. To support this decision process, we developed the Selection of Machine-learning Algorithms with ReplicaTions (SMART) Framework that integrates building and selecting ML models with decision theory. We extract theoretical insights while illustrating the power and flexibility of the SMART Framework to predict patients at high risk of complications due to diabetes using two data sources. The datasets are unique in volume and variety, each requiring a unique set of analytical methods for building viable prediction models. We outline opportunities for stakeholder engagement throughout the SMART Framework to empower medical decision-makers and build their confidence in leveraging ML models to enhance patient care. We demonstrate that combining ML and decision theory incorporates uncertainty and stakeholder engagement, resulting in more robust data-driven medical decision-making support tools.Next, we focus on the preventative and personalized arms of P4 medicine in two clinical operations case studies related to diabetic retinopathy (DR) and an emergency department (ED). To prevent blindness due to DR, we developed a microsimulation model of diabetes progression and patient's interaction with the care system. We amplify the effects of care interventions, tested in a small clinical trial, to population-level impacts using patient health outcomes, adherence behavior, and cumulative totals of visits. To increase patient flow through an ED, we develop an optimization model and a discrete event simulation model to compare alternate staffing schedules to a team-based care staffing model. Complex interactions between workload variability, uncertain and increasing arrival rates, and resource constraints make it difficult to improve flow and reduce crowding in the ED. We merged patient information and operations variables to give a more holistic view of the patient experience. In both case studies, we continuously iterate our data interpretation, analyses, and models with stakeholders. As a result, we establish data-driven recommendations for increasing the number of patients treated for DR, thus reducing the incidence of blindness, and for efficiently matching ED staffing to patient demand.The volume, variety, and veracity of datasets in each chapter were unique, yet, combined with stakeholders' participation, we built models that transformed data into medical decision-making support tools.
590 $a School code: 0155.
650 4 $a Diabetes. $3 544344
650 4 $a Sensitivity analysis. $3 3560752
650 4 $a Diabetic retinopathy. $3 1086927
650 4 $a Decision making. $3 517204
650 4 $a Clinical trials. $3 724498
650 4 $a Data processing. $3 680224
650 4 $a Coronaviruses. $3 894828
650 4 $a Clinical outcomes. $3 3682127
650 4 $a COVID-19. $3 3554449
650 4 $a Artificial intelligence. $3 516317
650 4 $a Pharmacology. $3 634543
650 4 $a Simulation. $3 644748
650 4 $a Trends. $3 3560051
690 $a 0800
690 $a 0419
710 2 $a North Carolina State University. $3 1018772
773 0 $t Dissertations Abstracts International $g 83-02B.
790 $a 0155
791 $a Ph.D.
792 $a 2021
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28552538