Grant: Project Grant
McMaster University
$153,000.00
Machine Learning enhanced Central Statistical Monitoring of Trials to ensure Robust & Assured Quality (TRAQ)
FY 2020-21
Summary
This work develops new and easy machine learning tools to find problems in data in clinical studies. It enables more robust data and better quality studies, as well as lower costs.
Regulatory agencies ask sponsors of studies to ensure the safety of study participants and the quality of the study data. Industry sponsors carry out costly resource-intensive procedures like onsite monitoring visits and complete source data verification. Academic investigators cannot afford to do so, and focus on a few key data elements, which is still resource and effort intensive. It is important to detect irregularities in data early, to investigate them quickly, take prompt necessary corrective action, and prevent future irregularities.
Specific Aim: Develop an east to implement artificial intelligence (AI) machine learning tool for detection of irregularities in data of multi-center collaborative clinical trials.
Approach:
1. Develop an unsupervised machine learning algorithm
2. Verify and improve its performance using existing data from prior conducted global RCTs at the Population Health Research Institute (PHRI)
3. Apply the algorithm to ongoing RCTs at PHRI
4. Disseminate the developed methodology in the trials research community
5. Conduct a 3-day knowledge translation workshop
Expected outcomes:
1. Easy to implement machine learning based tool for comprehensive and thorough centralized statistical monitoring and identification of irregularities in clinical trials
2. More robust, efficient and better data quality in clinical trials
3. Cost savings since can do targeted onsite monitoring
Department
Program
Project Grant
Recipient
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database
As published
Id
434422
Name
Bangdiwala Shrikant I
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Population and Public Health
Theme2
Not applicable / Specified
Prcname2
Randomized Controlled Trials 2
Projecttitle
Machine Learning enhanced Central Statistical Monitoring of Trials to ensure Robust & Assured Quality (TRAQ)
Programtype2
Operating Grants
Orgnameinp2
McMaster University
Cihramount2
$153,000
Cihrequipment2
$0
Programname2
Project Grant
Keyworddelim
Data Quality Assurance; Randomized Controlled Studies; Statistical Methodology
Abstract
This work develops new and easy machine learning tools to find problems in data in clinical studies. It enables more robust data and better quality studies, as well as lower costs.
Regulatory agencies ask sponsors of studies to ensure the safety of study participants and the quality of the study data. Industry sponsors carry out costly resource-intensive procedures like onsite monitoring visits and complete source data verification. Academic investigators cannot afford to do so, and focus on a few key data elements, which is still resource and effort intensive. It is important to detect irregularities in data early, to investigate them quickly, take prompt necessary corrective action, and prevent future irregularities.
Specific Aim: Develop an east to implement artificial intelligence (AI) machine learning tool for detection of irregularities in data of multi-center collaborative clinical trials.
Approach:
1. Develop an unsupervised machine learning algorithm
2. Verify and improve its performance using existing data from prior conducted global RCTs at the Population Health Research Institute (PHRI)
3. Apply the algorithm to ongoing RCTs at PHRI
4. Disseminate the developed methodology in the trials research community
5. Conduct a 3-day knowledge translation workshop
Expected outcomes:
1. Easy to implement machine learning based tool for comprehensive and thorough centralized statistical monitoring and identification of irregularities in clinical trials
2. More robust, efficient and better data quality in clinical trials
3. Cost savings since can do targeted onsite monitoring
Competitiondate
202010
Approvedterm2
2 yrs 0 mth
Pinamesdelim
Bangdiwala, Shrikant I
Approvedterm
2 ans 0 mois
Cihrcontribution2
~ $153,000~ $0~