A knowledge-based method for temporal abstraction of clinical data

This dissertation describes a reasoning framework for knowledge-based systems, specific to the task of abstracting higher-level, interval-based concepts from time-stamped data, but independent of any domain. The framework includes a logical model of time, parameters, events and contexts: a knowledge-based temporal-abstraction theory. The knowledge required for the inference structure that I propose is well defined and can be acquired for particular domains. I have applied my framework to the domain of clinical medicine. My goal is to create, from time-stamped patient data, interval-based temporal abstractions, such as "severe anemia for 3 weeks in the context of administering AZT," and more complex patterns, involving several intervals. I define a knowledge-based temporal-abstraction method that decomposes the task of abstracting higher-level abstractions from input data into five subtasks. These subtasks are solved by five separate, domain-independent, temporal-abstraction mechanisms. The temporal-abstraction mechanisms depend on four domain-specific knowledge types. The knowledge types and the role they play in each mechanism are defined formally. The knowledge needed to instantiate the temporal-abstraction mechanisms in any particular domain and task can be parameterized and can be acquired from domain experts manually or with automated tools. I present a computer program implementing the knowledge-based temporal-abstraction method: RESUME. RESUME accepts input and returns output at all levels of abstraction; accepts input out of temporal order, modifying a view of the past or of the present, as necessary; generates context-sensitive, controlled output; and maintains several possible concurrent interpretations of the data. I have evaluated RESUME in the domains of protocol-based care, monitoring of children's growth, and therapy of diabetes. The knowledge required for instantiating the temporal-abstraction mechanisms was acquired, maintained, and reused for creating new application systems. A formal specification of a domain's temporal-abstraction knowledge supports the design of systems that perform temporal-reasoning tasks, the acquisition of that knowledge, the maintenance of that knowledge, the reuse of the temporal abstraction knowledge in other domains, and the sharing of the domain-specific temporal abstraction knowledge with other applications.