Low complexity system architecture design for medical Cyber-Physical-Human Systems (CPHS)

Cyber-Physical-Human Systems (CHPS) are safety-critical systems, where the interaction between cyber components and physical components can be influenced by the human operator. Guaranteeing correctness and safety in these highly interactive computations is challenging. In particular, the interaction between these three components needs to be coordinated collectively in order to conduct safe and effective operations. The interaction nevertheless increases by orders of magnitude the levels of complexity and prevents formal verification techniques, such as model checking, from thoroughly verifying the safety and correctness properties of systems. In addition, the interactions could also significantly increase human operators' cognitive load and lead to human errors. In this thesis, we focus on medical CPHS and examine the complexity from a safety angle. Medical CPHS are both safety-critical and highly complex, because medical staff need to coordinate with distributed medical devices and supervisory controllers to monitor and control multiple aspects of the patient's physiology. Our goal is to reduce and control the complexity by introducing novel architectural patterns, coordination protocols and user-centric guidance system. This thesis makes three major contributions for improving safety of medical CPHS. Reducing verification complexity: Formal verification is a promising technique to guarantee correctness and safety, but the high complexity significantly increases the verification cost, which is known as state space explosion problems. We propose two architectural patterns: Interruptible Remote Procedure Call (RPC) and Consistent View Generation and Coordination (CVGC) protocol to properly handle asynchronous communication and exceptions with low complexity. Reducing cyber-medical treatment complexity: Cyber medical treatment complexity is defined as the number of steps and time to perform a treatment and monitor the corresponding physiological responses. We propose treatment and workflow adaptation and validation protocols to semi-autonomously validate the preconditions and adapt the workflows to patient conditions, which reduces the complexity of performing treatments and following best practice workflows. Reducing human cognitive load complexity: Cognitive load (also called mental workload) complexity measures human memory and mental computation demand for performing tasks. We first model individual medical staff's responsibility and team interactions in cardiac arrest resuscitation and decomposed their overall task into a set of distinct cognitive tasks that must be specifically supported to achieve successful human-centered system design. We then prototype a medical Best Practice Guidance (BPG) system to reduce medical staff's cognitive load and foster adherence to best practice workflows. Our BPG system transforms the implementation of best practice medical workflow.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2016

Author dc:creator

• Wu, Po-Liang

Contributors dc:contributor

• Sha, Lui R.
• Abdelzaher, Tarek F.
• Kirlik, Alex
• Mangharam, Rahul

• Cyber-Physical-Human Systems
• Healthcare
• Complexity

Statement dc:rights

• Copyright 2016 Po-Liang Wu

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/90509

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/90509