TY - GEN
T1 - Towards Modular and Formally-Verifiable Software Architecture for Clinical Guidance Systems
AU - Song, Shuang
AU - Saxena, Manasvi
AU - Tsai, Pei Hsuan
AU - Sha, Lui
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - Computer Science is being increasingly used in medicine to improve quality of care and patient outcome. Clinical Decision Support Systems (CDSSs) that codify clinical Best Practice Guidelines (BPGs) and provide situation-specific advice to physicians CDSSs have shown effectiveness in reducing adverse patient outcomes during clinical evaluations. However, representing both the BPG and the associated physical processes in software is complex and tedious, making CDSSs prone to bugs. This can be mitigated using a modular software architecture that encapsulates computational representation of physical processes for finer-grained development and verification leading to improved comprehensibility, shareability and maintainability. This paper discusses the sources of complexity in CDSSs, and proposes a software architecture that consists of an encoding of the BPG's medical knowledge into executable representations comprising of a patient digital twin, diagnosis and treatment workflows, an adherence monitor, a User Interface (UI) and a middleware for seamless integration with existing Hospital Information Systems. We developed a CDSS for Pediatric Sepsis Management co-designed by physicians using our approach. We use the Fluid Resuscitation therapy of this CDSS as a case study to illustrate the development process.
AB - Computer Science is being increasingly used in medicine to improve quality of care and patient outcome. Clinical Decision Support Systems (CDSSs) that codify clinical Best Practice Guidelines (BPGs) and provide situation-specific advice to physicians CDSSs have shown effectiveness in reducing adverse patient outcomes during clinical evaluations. However, representing both the BPG and the associated physical processes in software is complex and tedious, making CDSSs prone to bugs. This can be mitigated using a modular software architecture that encapsulates computational representation of physical processes for finer-grained development and verification leading to improved comprehensibility, shareability and maintainability. This paper discusses the sources of complexity in CDSSs, and proposes a software architecture that consists of an encoding of the BPG's medical knowledge into executable representations comprising of a patient digital twin, diagnosis and treatment workflows, an adherence monitor, a User Interface (UI) and a middleware for seamless integration with existing Hospital Information Systems. We developed a CDSS for Pediatric Sepsis Management co-designed by physicians using our approach. We use the Fluid Resuscitation therapy of this CDSS as a case study to illustrate the development process.
UR - http://www.scopus.com/inward/record.url?scp=85187301057&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85187301057&partnerID=8YFLogxK
U2 - 10.1109/SMC53992.2023.10393931
DO - 10.1109/SMC53992.2023.10393931
M3 - Conference contribution
AN - SCOPUS:85187301057
T3 - Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics
SP - 4271
EP - 4276
BT - 2023 IEEE International Conference on Systems, Man, and Cybernetics
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2023 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2023
Y2 - 1 October 2023 through 4 October 2023
ER -