Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Doctoral Award - Frederick Banting and Charles Best Canada Graduate Scholarships

University of Toronto

$105,000.00

Developing comprehensive predictive models of postoperative outcomes to improve surgical performance and perioperative care

FY 2017-18

Summary
Surgery is never without risks. Complications such as wound infections, blood clots and even death are not only possible but still common. The cause of these complications generally falls in three main categories: the patient's medical conditions, the surgeon's skill during the operation, and the care the patient receives after the surgery. Science has come a long way in reducing these risks, but there is still work to be done putting these areas together. In this study, we ask how medical conditions, surgical skill and care after the surgery intersect to cause complications. Currently, before surgery, a patient's risk of complications or death is calculated using a risk score, which assesses their medical conditions alone. Risk scores generally stop at the door to the operating room: they do not take into account the surgeon's performance or the care afterwards. The surgeon, therefore, generally does not know how to tailor her skill or postoperative care to provide her patient with the best outcomes. Programming techniques in the field of machine learning are capable of finding these connections between medical conditions, surgical skill, and care after an operation, to discover ways for surgeons to tailor care and improve their patients' outcomes. We plan to use our large existing database of surgical videos in the field of bariatric surgery and urology to provide insight into surgical skills, and combine this with data from the medical record system at St. Michael's Hospital to get information about patients' health, postoperative care and outcomes. The objective of this project is to build a tool that will teach surgeons how to better care for each individual patient, as certain health conditions may cause their needs to vary. By finding care pathways that are customized to individual patients, this project could change the way surgeons operate on and care for their patients, and in so doing reduce patients' complications and improve their quality of life.
Program
Doctoral Award - Frederick Banting and Charles Best Canada Graduate Scholarships
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database

As published

Id
383326
Name
Gordon Lauren E
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Population and Public Health
Theme2
Clinical
Prcname2
Doctoral Research Awards - B
Projecttitle
Developing comprehensive predictive models of postoperative outcomes to improve surgical performance and perioperative care
Programtype2
Studentship Programs
Orgnameinp2
University of Toronto
Cihramount2
$105,000
Cihrequipment2
$0
Programname2
Doctoral Award - Frederick Banting and Charles Best Canada Graduate Scholarships
Keyworddelim
Machine Learning; Predictive Modelling; Quality Improvement; Surgical Safety
Abstract
Surgery is never without risks. Complications such as wound infections, blood clots and even death are not only possible but still common. The cause of these complications generally falls in three main categories: the patient's medical conditions, the surgeon's skill during the operation, and the care the patient receives after the surgery. Science has come a long way in reducing these risks, but there is still work to be done putting these areas together. In this study, we ask how medical conditions, surgical skill and care after the surgery intersect to cause complications. Currently, before surgery, a patient's risk of complications or death is calculated using a risk score, which assesses their medical conditions alone. Risk scores generally stop at the door to the operating room: they do not take into account the surgeon's performance or the care afterwards. The surgeon, therefore, generally does not know how to tailor her skill or postoperative care to provide her patient with the best outcomes. Programming techniques in the field of machine learning are capable of finding these connections between medical conditions, surgical skill, and care after an operation, to discover ways for surgeons to tailor care and improve their patients' outcomes. We plan to use our large existing database of surgical videos in the field of bariatric surgery and urology to provide insight into surgical skills, and combine this with data from the medical record system at St. Michael's Hospital to get information about patients' health, postoperative care and outcomes. The objective of this project is to build a tool that will teach surgeons how to better care for each individual patient, as certain health conditions may cause their needs to vary. By finding care pathways that are customized to individual patients, this project could change the way surgeons operate on and care for their patients, and in so doing reduce patients' complications and improve their quality of life.
Competitiondate
201710
Approvedterm2
3 yrs 0 mth
Pinamesdelim
Gordon, Lauren E
Approvedterm
3 ans 0 mois
Cihrcontribution2
~ $105,000~ $0~