Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Discovery Grants Program - Individual

Carleton University

$18,000.00

Development of a diagnostic decision support tool

FY 2009-10

Summary
The research aims to provide an interactive multimedia computer application to train pediatric residents in effective diagnostic decision strategies. The approach includes research in human decision making, human computer interaction, biomedical engineering, information technology, and medicine. Application of knowledge of serious, systematic human decision biases as a basis for algorithm design in a medical diagnostic decision support system is, to the best of our knowledge, novel. Relatively inexperienced decision makers tend to jump to conclusions very rapidly, which is problematic when the stimulus information is ambiguous. As a consequence, the first item presented tends to weigh more heavily than justified by the evidence. In addition, research conducted over the past 25 years in different domains suggests that people have a poor understanding of the concept of diagnosticity, which is a measure of specificity and the frequency of occurrence of a given sign, symptom, or test result. Medical text books do not quantify symptom diagnosticity. Diagnosticians thus do not have access to this vital information but rely on cognitively taxing memorized lists of signs and symptoms associated with each disease. A proof-of-concept prototype has been developed with a database that is based on actual and complete patient records. The underlying algorithm determines the diagnosticity of each symptom or sign associated with a particular condition, respiratory distress in infants. Currently, the user interface is a simple GUI with a text-based interaction dialog. Upon user input of symptoms and signs, the system calculates the probability of different diseases associated with respiratory distress. It outputs the five topmost probabilities when the user elects to initiate a 'diagnose' function. The proposed research will expand the database and adopt a multimedia user interface that incorporates images such as X-rays as well as allowing free text user input. In addition, it will also be designed to offer suggestions for tests to conduct next or information to obtain to increase the probability of the most likely disease.
Program
Discovery Grants Program - Individual
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
371869-2009
Name nom
Lindgaard, Gitte
Department département
Psychology
Organization id
23
Institution établissement
Carleton University
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2009
Competition year année de concours
2009
Award amount
18000
Program id
RGPIN
Program nane en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
21
Committee name en
Interdisciplinary
Committee name fr
Interdisciplinaire
Area of application code
801
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Computer software
Area of application fr
Logiciel
Research subject code
6300
Research subject group en
Psychology
Research subject group fr
Psychologie
Research subject en
Psychology
Research subject fr
Psychologie
Partie
1
Nb partie
1
Application title
Development of a diagnostic decision support tool
Keyword
NA
Application summary
The research aims to provide an interactive multimedia computer application to train pediatric residents in effective diagnostic decision strategies. The approach includes research in human decision making, human computer interaction, biomedical engineering, information technology, and medicine. Application of knowledge of serious, systematic human decision biases as a basis for algorithm design in a medical diagnostic decision support system is, to the best of our knowledge, novel. Relatively inexperienced decision makers tend to jump to conclusions very rapidly, which is problematic when the stimulus information is ambiguous. As a consequence, the first item presented tends to weigh more heavily than justified by the evidence. In addition, research conducted over the past 25 years in different domains suggests that people have a poor understanding of the concept of diagnosticity, which is a measure of specificity and the frequency of occurrence of a given sign, symptom, or test result. Medical text books do not quantify symptom diagnosticity. Diagnosticians thus do not have access to this vital information but rely on cognitively taxing memorized lists of signs and symptoms associated with each disease. A proof-of-concept prototype has been developed with a database that is based on actual and complete patient records. The underlying algorithm determines the diagnosticity of each symptom or sign associated with a particular condition, respiratory distress in infants. Currently, the user interface is a simple GUI with a text-based interaction dialog. Upon user input of symptoms and signs, the system calculates the probability of different diseases associated with respiratory distress. It outputs the five topmost probabilities when the user elects to initiate a 'diagnose' function. The proposed research will expand the database and adopt a multimedia user interface that incorporates images such as X-rays as well as allowing free text user input. In addition, it will also be designed to offer suggestions for tests to conduct next or information to obtain to increase the probability of the most likely disease.