國立虎尾科技大學 |

Leveraging Real World Data and Machine Learning to Improve Clinical Management of Methicillin-Resistant Staphylococcus aureus Infections.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Leveraging Real World Data and Machine Learning to Improve Clinical Management of Methicillin-Resistant Staphylococcus aureus Infections./
Author:	Jun, Inyoung.
Description:	1 online resource (90 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 85-08, Section: B.
Contained By:	Dissertations Abstracts International85-08B.
Subject:	Public health. -
Online resource:	click for full text (PQDT)
ISBN:	9798381692860

Leveraging Real World Data and Machine Learning to Improve Clinical Management of Methicillin-Resistant Staphylococcus aureus Infections.
Jun, Inyoung.

Leveraging Real World Data and Machine Learning to Improve Clinical Management of Methicillin-Resistant Staphylococcus aureus Infections. - 1 online resource (90 pages)

Source: Dissertations Abstracts International, Volume: 85-08, Section: B.

Thesis (Ph.D.)--University of Florida, 2023.

Includes bibliographical references

Identifying an optimal antibiotic treatment regimen is key to increase the chance of favorable clinical outcome for individuals and can also have a role in reducing AMR at the population level, by stopping the spread of drug-resistant strains in the community or hospitals. However, most of the treatment guidelines are based on randomized clinical trials (RCTs), which are very powerful to unveil causal effects, but limited in sample size and population settings. Further, RCTs are inherently unable to explore the whole space of treatment and care options due to cost/resource or ethical constraints. Those unmet needs could be fulfilled and complemented if we can leverage real world, large-scale, observational data because they offer a larger sample size and allow the testing of multiple hypotheses. In this dissertation, we propose to fill knowledge gaps in MRSA treatment and clinical management by: (1) building an EHR-based machine learning prediction model of treatment change, mortality, and recurrence for MRSA bacteremia; (2) quantifying social and racial disparity in MRSA infection risk and outcomes using individual-level clinical information from EHR and contextual-level SDoH, using algorithmic fairness methods; (3) developing a 'counterfactual' (i.e., causal) prediction model for optimizing antibiotic choice of MRSA patients by comparing vancomycin only versus replacing/adding other antibiotics, leveraging RCT emulation methods and machine learning. This study builds upon: (1) a comprehensive statewide EHR records dataset, UF Health IDR, which includes patient-level information and follow-ups about demographics, admissions, vitals, diagnosis, procedures, lab tests (including antimicrobial resistance testing results), and medications -across almost a decade; and (2) our prior works in the field using EHR and causal machine learning (e.g., trial emulation and individualized treatment effect prediction).

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

Mode of access: World Wide Web

ISBN: 9798381692860Subjects--Topical Terms:

560998
Public health.
Subjects--Index Terms:

Real world dataIndex Terms--Genre/Form:

554714
Electronic books.

Leveraging Real World Data and Machine Learning to Improve Clinical Management of Methicillin-Resistant Staphylococcus aureus Infections.
LDR:03393ntm a22003977 4500 001 1149401
005 20241015112543.5
006 m o d
007 cr bn ---uuuuu
008 250605s2023 xx obm 000 0 eng d
020 $a 9798381692860
035 $a (MiAaPQ)AAI30815086
035 $a AAI30815086
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Jun, Inyoung. $3 1475622
245 1 0 $a Leveraging Real World Data and Machine Learning to Improve Clinical Management of Methicillin-Resistant Staphylococcus aureus Infections.
264 0 $c 2023
300 $a 1 online resource (90 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: Dissertations Abstracts International, Volume: 85-08, Section: B.
500 $a Advisor: Prosperi, Mattia.
502 $a Thesis (Ph.D.)--University of Florida, 2023.
504 $a Includes bibliographical references
520 $a Identifying an optimal antibiotic treatment regimen is key to increase the chance of favorable clinical outcome for individuals and can also have a role in reducing AMR at the population level, by stopping the spread of drug-resistant strains in the community or hospitals. However, most of the treatment guidelines are based on randomized clinical trials (RCTs), which are very powerful to unveil causal effects, but limited in sample size and population settings. Further, RCTs are inherently unable to explore the whole space of treatment and care options due to cost/resource or ethical constraints. Those unmet needs could be fulfilled and complemented if we can leverage real world, large-scale, observational data because they offer a larger sample size and allow the testing of multiple hypotheses. In this dissertation, we propose to fill knowledge gaps in MRSA treatment and clinical management by: (1) building an EHR-based machine learning prediction model of treatment change, mortality, and recurrence for MRSA bacteremia; (2) quantifying social and racial disparity in MRSA infection risk and outcomes using individual-level clinical information from EHR and contextual-level SDoH, using algorithmic fairness methods; (3) developing a 'counterfactual' (i.e., causal) prediction model for optimizing antibiotic choice of MRSA patients by comparing vancomycin only versus replacing/adding other antibiotics, leveraging RCT emulation methods and machine learning. This study builds upon: (1) a comprehensive statewide EHR records dataset, UF Health IDR, which includes patient-level information and follow-ups about demographics, admissions, vitals, diagnosis, procedures, lab tests (including antimicrobial resistance testing results), and medications -across almost a decade; and (2) our prior works in the field using EHR and causal machine learning (e.g., trial emulation and individualized treatment effect prediction).
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Public health. $3 560998
650 4 $a Microbiology. $3 591510
650 4 $a Epidemiology. $3 635923
653 $a Real world data
653 $a Clinical management
653 $a Staphylococcus aureus
653 $a Antibiotic treatment
653 $a Machine learning prediction
655 7 $a Electronic books. $2 local $3 554714
690 $a 0573
690 $a 0410
690 $a 0766
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a University of Florida. $b Epidemiology. $3 1475623
773 0 $t Dissertations Abstracts International $g 85-08B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30815086 $z click for full text (PQDT)