Case Based Reasoning

Introduction

The field of healthcare involves decision-making in every sphere of its life cycle. Decision-making can pose a challenge in cases where there is less or negligible domain-specific knowledge. Although there exists ample amount of understanding of the way the healthcare domain works, it has its share of uncertainties and complex situations that call for an explicit understanding of the relation between various occurrences of events, likely causes and effects that govern the domain. In such cases, experience plays a crucial role in assisting the decision-making process, and one such approach to medical reasoning is the Case-based reasoning (CBR) approach, that uses previous experiences to solve new problems. This approach relies on

The central premise is that what solved a previous patient problem has a good chance of solving a new but similar patient problem (Hauan, 2004). The inference engine will retrieve similar cases and solutions, which presents the closest match to the new case symptoms and signs. The physician can reuse the treatment plan offered as a solution or may fill in missing information to revise the case for an updated solution. The updated problem and solution, once vetted can be retained in the case-based knowledge base for future case searches. There is no special training required to use this system, it is meant to complement and aid the physician, the case-based method is easy to interpret since it lends itself to previous experiences of the

The CDSS inference engine will accept the patient history, signs, symptoms, and test results from the EMR in real-time, and present the closest case and solution to the physician. The CDSS will match up current symptoms and signs and will place proper alerts and suggestions from within the current EMR. The alerts, informational messages, and diagnosis are based on a sophisticated knowledge base database loaded with evidence-based medical cases designed to work within a wide range of EMR domains. The case-based method allows the addition of revised problem-solution cases, and conversely allows for the soft removal of obsolete problem-solution cases by flagging them as inactive or “forgotten”.

Conclusion

A system developed using case-based reasoning would be able to provide more comprehensive solutions for users because it has access to knowledge acquired from a vast number of past cases. Clinicians must consider large amounts of information when searching for solutions and a case-based system is able to acknowledge the complex relationships between that data and examine how all of that information has worked when combined together before. The system would enable quick analysis and the use of knowledge acquired by many other clinicians from other healthcare