Grant: CIHR Fellowship
University of Calgary
$90,000.00
The Development of Machine Learning Algorithms to Identify Perioperative Adverse Events Using Real-world Data
EF 2022-23
Résumé
In Canada, one in ten patients experiences adverse events (AEs) during hospitalization. Over half of AEs are related to surgery. The economic burden of hospital AEs were estimated to be 1.1 billion dollars in Canada, importantly, most of AEs are preventable. Study on the AEs could dramatically save the costs, but one first necessary step is to identify them. Traditional methods for AE detection including medical experts chart reviews and ICD codes. However, it is well known that chart reviews often require extensive human resources and time-consuming, and the ICD codes for AEs that based on patient discharge databases are often under-reported and lack of accuracy.
Electronic medical records (EMR) contain detailed information regarding various aspects of patient care and has been widely implemented across Canada. As such, EMRs contain doctors' and nurses' notes and are a valuable resource for the timely detection of surgery related AEs. However, when EMR systems are set up, they have not yet been coupled with powerful computational tools to automatically detect AEs.
To bridge this gap, we propose the current study, which aims to create EMR-based algorithms that detect surgery related AEs. We will use artificial intelligence techniques (i.e., natural language processing) to automate the detection of surgery related AEs (Surgical Site Infections, Postoperative hip fracture, Postoperative hemorrhage or hematoma, Postoperative physiologic and metabolic derangements, Postoperative respiratory failure, Postoperative pulmonary embolism or deep vein thrombosis, Postoperative sepsis, Postoperative wound dehiscence in abdominopelvic surgical patients) using linked EMR and coded hospital data from Alberta. We will then assess the validity of the developed algorithms by comparing them with clinical expert chart reviews. The developed EMR-based algorithms will ultimately be used to routinely detect AEs for patient safety surveillance and monitoring and to improve patient care.
Programme
CIHR Fellowship
Bénéficiaire
Emplacement
Alberta, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database
Tel que publié
Id
471987
Name
Wu Guosong
Region
Alberta
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Health Services and Policy Research
Theme2
Health systems / services
Prcname2
Health Research Training A - Post-PhD (HTA)
Projecttitle
The Development of Machine Learning Algorithms to Identify Perioperative Adverse Events Using Real-world Data
Programtype2
Fellowship Programs
Orgnameinp2
University of Calgary
Cihramount2
$90,000
Cihrequipment2
$0
Programname2
CIHR Fellowship
Keyworddelim
Care Quality; Patient Safety
Abstract
In Canada, one in ten patients experiences adverse events (AEs) during hospitalization. Over half of AEs are related to surgery. The economic burden of hospital AEs were estimated to be 1.1 billion dollars in Canada, importantly, most of AEs are preventable. Study on the AEs could dramatically save the costs, but one first necessary step is to identify them. Traditional methods for AE detection including medical experts chart reviews and ICD codes. However, it is well known that chart reviews often require extensive human resources and time-consuming, and the ICD codes for AEs that based on patient discharge databases are often under-reported and lack of accuracy.
Electronic medical records (EMR) contain detailed information regarding various aspects of patient care and has been widely implemented across Canada. As such, EMRs contain doctors' and nurses' notes and are a valuable resource for the timely detection of surgery related AEs. However, when EMR systems are set up, they have not yet been coupled with powerful computational tools to automatically detect AEs.
To bridge this gap, we propose the current study, which aims to create EMR-based algorithms that detect surgery related AEs. We will use artificial intelligence techniques (i.e., natural language processing) to automate the detection of surgery related AEs (Surgical Site Infections, Postoperative hip fracture, Postoperative hemorrhage or hematoma, Postoperative physiologic and metabolic derangements, Postoperative respiratory failure, Postoperative pulmonary embolism or deep vein thrombosis, Postoperative sepsis, Postoperative wound dehiscence in abdominopelvic surgical patients) using linked EMR and coded hospital data from Alberta. We will then assess the validity of the developed algorithms by comparing them with clinical expert chart reviews. The developed EMR-based algorithms will ultimately be used to routinely detect AEs for patient safety surveillance and monitoring and to improve patient care.
Competitiondate
202210
Approvedterm2
2 yrs 0 mth
Pinamesdelim
Wu, Guosong
Approvedterm
2 ans 0 mois
Cihrcontribution2
~ $90,000~ $0~