ABSTRACT
Soft-computing approaches are applied in industries such as finance, retail, construction, and agriculture to help with decision-making and problem-solving. Beyond these fields, they have also been applied in healthcare, particularly in the context of Clinical Decision Support Systems (CDSS). Accurate and timely medical diagnosis is critical in the healthcare industry because accuracy and timeliness directly impact the outcome of the treatment of a patient. Many diseases, like neurodegenerative diseases, some cancer types, etc., have multiple symptoms with varying intensities in patient groups due to the variation in the genetic makeup and the environment of the patient (Van den Broeck et al.; Wang et al.). The response of each patient to the same drug has been noted to be different, and this trend has been associated with single nucleotide polymorphisms (SNPs) (Shastry, 2007). But, it is not easy to determine the environment’s effect on the epigenetics of a patient and, hence, its effect on the disease outcome. Thus, it becomes very time and energy consuming for medical practitioners to come to a conclusive diagnosis. This issue also leads to a delay in the treatment initiation that may seriously affect the patient’s health.
The advantage of soft computing’s ability to adapt itself to a particular field makes it suitable for use in the healthcare domain. Soft computing’s applicability in CDSS has revolutionized the field of diagnosis and tracking of a patient’s treatment.
CDSS is used in diagnosis, prognosis, and prediction and acts like an overall aid to the medical professionals. Clinical Decision Support Systems are complex systems composed of various related and unrelated sub-systems. This chapter will summarize the most popular soft-computing techniques in medical DSS, such as Fuzzy Analytical Hierarchy Process (FAHP), Adaptive Neuro-Fuzzy Inference System (ANFIS), Artificial Neural Networks (ANN) and Fuzzy Cognitive Maps (FCM). A detailed literature review summarizing these models will be discussed in this chapter. Subsequently, the paper will also look into how these soft-computing techniques have resulted in the development of DSS and how these DSS are being used in aiding real-world decision-making in the healthcare domain. A detailed discussion showcasing their real-world applicability will be presented here. We have also highlighted the shortcomings of the current CDSSs being used around the world and how these shortcomings are being overcome by certain medical institutions. The chapter will also discuss potential applications of CDSS beyond diagnosis and patient health tracking. Lastly, a very important aspect that will be considered is how important it is to maintain high security in these CDSSs. Tinkering with the CDSS by hackers, no matter how seemingly insignificant, can adversely affect the patient’s health and treatment progression.
