TY - GEN
T1 - Evaluating the TMR Model for Multimorbidity Decision Support Using a Community-of-Practice Based Methodology
AU - Grgurić, Josip
AU - Teije, Annette ten
AU - Harmelen, Frank van
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
PY - 2024
Y1 - 2024
N2 - Clinical practice guidelines are typically designed for treatment of a single disease, ignoring undesired interactions for comorbid patients. A number of methods for detecting such guideline interactions have been developed, based on computer interpretable representations of guidelines. A recently published paper by Van Woensel et al. [7] compared a number of methods for detecting and resolving interactions between multiple guidelines. The current paper contributes to this comparative corpus by applying the same functional features and evaluation dimensions to the TMR method for multimorbidity decision support. Our comparison shows that TMR allows for more complex reasoning compared to some of the methods discussed in [7]. It is one of the few that supports automated detection of adverse interactions. However, it falls short on temporal reasoning and reasoning about drug dosage. Our study also represents the first independent validation of the evaluation methodology published in [7].
AB - Clinical practice guidelines are typically designed for treatment of a single disease, ignoring undesired interactions for comorbid patients. A number of methods for detecting such guideline interactions have been developed, based on computer interpretable representations of guidelines. A recently published paper by Van Woensel et al. [7] compared a number of methods for detecting and resolving interactions between multiple guidelines. The current paper contributes to this comparative corpus by applying the same functional features and evaluation dimensions to the TMR method for multimorbidity decision support. Our comparison shows that TMR allows for more complex reasoning compared to some of the methods discussed in [7]. It is one of the few that supports automated detection of adverse interactions. However, it falls short on temporal reasoning and reasoning about drug dosage. Our study also represents the first independent validation of the evaluation methodology published in [7].
UR - https://www.scopus.com/pages/publications/85200770820
UR - https://www.scopus.com/pages/publications/85200770820#tab=citedBy
UR - https://link.springer.com/book/10.1007/978-3-031-66538-7
U2 - 10.1007/978-3-031-66538-7_6
DO - 10.1007/978-3-031-66538-7_6
M3 - Conference contribution
AN - SCOPUS:85200770820
SN - 9783031665370
VL - 1
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 52
EP - 63
BT - Artificial Intelligence in Medicine
A2 - Finkelstein, Joseph
A2 - Moskovitch, Robert
A2 - Parimbelli, Enea
PB - Springer Science and Business Media Deutschland GmbH
T2 - 22nd International Conference on Artificial Intelligence in Medicine, AIME 2024
Y2 - 9 July 2024 through 12 July 2024
ER -