Grant: Discovery Grants Program - Individual
Queen's University
$42,000.00
Predictive and Prescriptive Analytics for Delivery of Mental Health Care
EF 2023-24
Résumé
The health sector constitutes a rich context for unique and challenges for predictive and prescriptive analytics. While operations researchers and artificial intelligence experts have identified and developed solutions for many distinctive problems in this domain, the field of mental care has been mostly overlooked by scholars in these fields. During the period of this grant, our contributions in NSE will be grounded in the reality of serious and persistent mental illnesses (SPMIs). In particular, the optimal stopping time models have a wide variety of applications in this context. The determination of when to request and enforce community treatment orders for schizophrenia patients is one example. Another fruitful application is the timing of discharge of an SPMI patient from the hospital. We plan to contribute several methodological improvements in decision-making for optimal timing. These problems are specifically difficult in the mental care domain due to the presence of significant uncertainties. The use of machine learning in this context has been mostly for predicting treatment outcomes for psychiatric medications, predominantly using clinical trial data. As one of the novel features of this research program, we will be using observational data. One of the well-known methodological issues is balancing the distributions of observable covariates to obtain unbiased causal estimates. We intend to further investigate the use of optimization-based models for this purpose. A third area of contributions involve analyses of the free-hand clinical notes in designing preventive care for the SPMI patient. These notes are taken by the psychiatrist during the clinic visit and they have the subjective, objective, assessment, and plan fields. The use of quantitative markers (e.g., blood pressure, heart rate) have dominated NSE studies using electronic medical records data. We are not aware of any work using the text data extracted from clinical notes, particularly in mental care. We will use natural language processing and text mining techniques to predict the hospitalization risk soon after the clinical visit. This will enable us to develop methods for optimizing the interventions to prevent the patient from being hospitalized, which is an undesirable and costly event for all stakeholders.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
RGPIN-2023-03840
Name nom
Verter, Vedat
Department département
Smith School of Business
Organization id
29
Institution établissement
Queen's University
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2023
Competition year année de concours
2023
Award amount
42000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1509
Committee name en
Civil, Industrial and Systems Engineering
Committee name fr
Génie civil, industriel et des systèmes
Area of application code
520
Area of application group en
Manufacturing processes and products
Area of application group fr
Procédés et produits industriels
Area of application en
Production and operations management
Area of application fr
Production et gestion des opérations
Research subject code
9999
Research subject group en
Not available
Research subject group fr
Not available
Research subject en
Non disponible
Research subject fr
Non disponible
Application title
Predictive and Prescriptive Analytics for Delivery of Mental Health Care
Keyword
Empirical Methods; Health care facility networks; Health care operations management; Machine Learning; Meta-heuristic Algorithms; Mixed-Integer Programming; Process Design and Improvement
Application summary
The health sector constitutes a rich context for unique and challenges for predictive and prescriptive analytics. While operations researchers and artificial intelligence experts have identified and developed solutions for many distinctive problems in this domain, the field of mental care has been mostly overlooked by scholars in these fields. During the period of this grant, our contributions in NSE will be grounded in the reality of serious and persistent mental illnesses (SPMIs). In particular, the optimal stopping time models have a wide variety of applications in this context. The determination of when to request and enforce community treatment orders for schizophrenia patients is one example. Another fruitful application is the timing of discharge of an SPMI patient from the hospital. We plan to contribute several methodological improvements in decision-making for optimal timing. These problems are specifically difficult in the mental care domain due to the presence of significant uncertainties. The use of machine learning in this context has been mostly for predicting treatment outcomes for psychiatric medications, predominantly using clinical trial data. As one of the novel features of this research program, we will be using observational data. One of the well-known methodological issues is balancing the distributions of observable covariates to obtain unbiased causal estimates. We intend to further investigate the use of optimization-based models for this purpose. A third area of contributions involve analyses of the free-hand clinical notes in designing preventive care for the SPMI patient. These notes are taken by the psychiatrist during the clinic visit and they have the subjective, objective, assessment, and plan fields. The use of quantitative markers (e.g., blood pressure, heart rate) have dominated NSE studies using electronic medical records data. We are not aware of any work using the text data extracted from clinical notes, particularly in mental care. We will use natural language processing and text mining techniques to predict the hospitalization risk soon after the clinical visit. This will enable us to develop methods for optimizing the interventions to prevent the patient from being hospitalized, which is an undesirable and costly event for all stakeholders.