Grant: Doctoral Research Award: Canada Graduate Scholarships
McGill University
$105,000.00
Predicting iron recovery after blood donation: comparing machine learning and biology-informed modeling approaches
EF 2022-23
Résumé
Iron deficiency may occur from repeat blood donation, but the risk varies considerably among individuals. While pre-donation fingerstick hemoglobin testing has been used to set a minimum hemoglobin threshold for blood donation, hemoglobin is not an accurate measure of iron stores, which subjects iron-deficient individuals to adverse outcomes after blood donation. I will develop and compare machine learning and biological systems models in predicting hemoglobin and ferritin levels using data from 2425 blood donors. I will test the predictions of both model types in blood banks with cohorts from three different countries to evaluate whether the models would work in different parts of the world. By creating statistical models that can be used by doctors making clinical decisions, I hope to provide better ways for blood banks to prevent iron deficiency from occurring in blood donors and to solve the competing challenges of meeting the constant demand for blood while managing the risk of iron deficiency in repeat donors.
Programme
Doctoral Research Award: Canada Graduate Scholarships
Bénéficiaire
Emplacement
Quebec, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database
Tel que publié
Id
475806
Name
Su Chen-Yang
Region
Québec
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Population and Public Health
Theme2
Health systems / services
Prcname2
Doctoral Research Awards - B
Projecttitle
Predicting iron recovery after blood donation: comparing machine learning and biology-informed modeling approaches
Programtype2
Studentship Programs
Orgnameinp2
McGill University
Cihramount2
$105,000
Cihrequipment2
$0
Programname2
Doctoral Research Award: Canada Graduate Scholarships
Keyworddelim
Biological Systems Modeling; Blood Donation; External Validation; Iron Deficiency Anemia; Iron Homeostasis; Machine Learning; Prediction Models
Abstract
Iron deficiency may occur from repeat blood donation, but the risk varies considerably among individuals. While pre-donation fingerstick hemoglobin testing has been used to set a minimum hemoglobin threshold for blood donation, hemoglobin is not an accurate measure of iron stores, which subjects iron-deficient individuals to adverse outcomes after blood donation. I will develop and compare machine learning and biological systems models in predicting hemoglobin and ferritin levels using data from 2425 blood donors. I will test the predictions of both model types in blood banks with cohorts from three different countries to evaluate whether the models would work in different parts of the world. By creating statistical models that can be used by doctors making clinical decisions, I hope to provide better ways for blood banks to prevent iron deficiency from occurring in blood donors and to solve the competing challenges of meeting the constant demand for blood while managing the risk of iron deficiency in repeat donors.
Competitiondate
202211
Approvedterm2
3 yrs 0 mth
Pinamesdelim
Su, Chen-Yang
Approvedterm
3 ans 0 mois
Cihrcontribution2
~ $105,000~ $0~