Treatment of patients with multimorbidity is one of the greatest challenges for clinical decision support. While evidence-based management of specific diseases is supported by clinical practice guidelines, concurrent application of multiple guidelines requires checking for possible adverse interactions between interventions and mitigating them, before a management plan is constructed. In earlier work, we developed an approach that casts the problem of multimorbidity management as an AI planning problem. In this paper we build on this earlier work and make progress towards creating a pipeline that inputs disease and patient-specific information and outputs a management plan. We describe research focused on selected aspects of pipeline development and illustrate these aspects with a clinical case implemented using the PDDL planning language and the OPTIC planner.
|Original language||English (US)|
|Title of host publication||Artificial Intelligence in Medicine - 20th International Conference on Artificial Intelligence in Medicine, AIME 2022, Proceedings|
|Editors||Martin Michalowski, Syed Sibte Raza Abidi, Samina Abidi|
|Publisher||Springer Science and Business Media Deutschland GmbH|
|Number of pages||10|
|State||Published - 2022|
|Event||20th International Conference on Artificial Intelligence in Medicine, AIME 2022 - Halifax, Canada|
Duration: Jun 14 2022 → Jun 17 2022
|Name||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Conference||20th International Conference on Artificial Intelligence in Medicine, AIME 2022|
|Period||6/14/22 → 6/17/22|
Bibliographical noteFunding Information:
Acknowledgements. We thank Jean-Luc Blais-Amyot and Maxime Côté-Gagnéfor their programming work on the automated translation component. We thank the reviewers for their helpful feedback. This research was supported by funding from the Telfer Health Transformation Exchange and the Natural Sciences and Engineering Research Council of Canada.
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
- AI planning
- End-to-end pipeline