Medical diagnosis is accomplished by a set of complex cognitive processes requiring the iterative application of abduction, deduction, and induction. Previous research in computational modeling of medical diagnosis has had only limited success by defining sub-domains that offer a computationally tractable problem. However, the aspect of diagnostic reasoning requiring intelligence lies in the extraction of a well-structured problem from an ill-structured one. We propose an agent, based on the Learning Intelligent Distribution Agent (LIDA) model of cognition, which utilizes deliberation, learning, and a neurologically inspired cognitive cycle. The proposed agent, MAX (for "Medical Agent X") will be equipped to comprehend clinical data in the context of its perceptual ontology and learned associations, and to construct, evaluate, and refine by investigation a differential diagnosis that progressively reduces the dimensionality of its search space with each iteration. Furthermore, the agent will appropriately modify its own ontology with experience and supervised instruction.

A clinical decision support system (CDSS) can help physicians make clinical decisions, such as di ff erential diagnosis, therapy planning, or plan critiquing. To make such informed decisions, a physician may need to keep track of a large amount of medical data and literature, such as new research articles, pharmacological therapies, and updates in Clinical Practice Guidelines. Therefore, a CDSS can be designed to assist physicians by providing relevant evidence-based clinical recommendations, subsequently reducing the cognitive overhead required to stay up-to-date with an evolving body of literature. We designed a CDSS by leveraging Semantic Web technologies to create an AI system that reasons in a way similar to physicians. We base our abstraction of human reasoning on the Select and Test Model (ST-Model), which combines multiple forms of reasoning, such as abstraction, deduction, abduction, and induction, to arrive at and test hypotheses. Based on this framework, we perform ensemble reasoning, the integration and interaction of multiple types of reasoning. We apply our CDSS to the treatment of type 2 diabetes mellitus by designing a domain ontology, the Diabetes Pharmacology Ontology (DPO), that supports both deductive and abductive reasoning. DPO is also used to provide a schema for our knowledge representation of hypothetical patients, where each patient is encoded in RDF as a Personalized Health Knowledge Graph (PHKG). We use the Whyis knowledge graph framework to implement our CDSS. This is achieved by writing software agents to perform custom deductive reasoning and integrating abduction using an existing reasoning engine, the AAA Abduction Solver. We apply our approach to perform therapy planning on hypothetical patients.

This study explored the logical underpinnings of medical reasoning, focusing on the integration of abduction, deduction, and induction within clinical decision-making. It aimed to highlight the role of abduction in generating hypotheses, particularly in complex cases that defy standard protocols, and to examine the synergy between human expertise and AI-assisted tools in enhancing diagnostic accuracy. The research employed a qualitative approach, analyzing philosophical theories and integrating them with clinical case studies. The study examined the interplay of logical processes in medical diagnostics and the application of abduction in rare and novel cases. Additionally, the potential of AI-assisted tools to support clinical reasoning and reduce diagnostic noise was explored. Abduction was identified as a critical yet often underappreciated element in medical reasoning essential for hypothesis generation. Deduction refines hypotheses against established medical knowledge, while induction validates decisions through empirical data. AI-assisted tools were found to enhance diagnostic accuracy by reducing noise, although they did not engage in the musement or genuine abductions that characterize human clinical reasoning. The study concluded that a triadic approach to clinical reasoning, incorporating abduction, deduction, and induction, is essential for effective medical diagnostics. In particular, abduction plays a pivotal role in navigating the complexities of clinical decision-making. The integration of AI tools can reduce noise and improve diagnostic processes, but the essential human elements of insight and judgment remain irreplaceable in patient care.

NEOANEMIA: a knowledge-based system emulating diagnostic reasoning. Medical diagnosis can be modeled in terms of the classical notions of abduction, deduction, and induction. Abduction is making a preliminary guess that allows one to establish a set of plausible diagnostic hypotheses, followed by deduction for exploring their consequences and induction for testing the hypotheses with available patient data or for planning the acquisition of new data. Such a description of diagnostic reasoning at a knowledge level helps the construction of an expert system by fashioning the adopted expert system building tool to reflect the structure of the problem rather than by fitting the problem to the tool. To this aim, reasoning strategies need to be represented abstractly, separate from medical facts and relations, to make the design more transparent and explainable. Published in Computers and biomedical research, an international journal (1990)

Medical expert systems based on causal probabilistic networks. Causal probabilistic networks (CPNs) offer new methods by which you can build medical expert systems that can handle all types of medical reasoning within a uniform conceptual framework. Based on the experience from a commercially available system and a couple of large prototype systems, it appears that CPNs are now an attractive alternative to other methods. A CPN is an intensional model of a domain, and it is therefore conceptually much closer to qualitative reasoning systems and to simulation systems than to rule-based or logic-based systems. Recent progress in Bayesian inference in networks has yielded computationally efficient methods. The inference method used follows the fundamental axioms of probability theory, and gives a sound framework for causal and diagnostic (deductive and abductive) reasoning under uncertainty. Experience with the prototypes indicates that it may be possible to use decision theory as a rational approach to test planning and therapy planning.