Multi-Agent Model for Clinical Decision Support System

Multi-Agent Model for Clinical Decision Support System

Jirapun Daengdej, Kitikorn Dowpiset, Kitti Phothikitti, Vechayan Choychoowong
Copyright: © 2024 |Pages: 14
DOI: 10.4018/979-8-3693-2667-1.ch009
OnDemand:
(Individual Chapters)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Clinical process contains a highly complex and sensitive set of activities. To minimize errors that may occur during the process, a system called clinical decision support system, or CDSS, has been widely suggested and researched. Recently, artificial intelligence (AI) has been introduced as part of the engine for the CDSS. Unfortunately, due to complexing of the clinical process, to the authors' knowledge, most of works propose that AI is used as small individual process for each of the task required in CDSS. This clearly does not mimic how human works in the real-world clinical process. This chapter discusses a novel AI-based CDSS model, which has been designed based on five senses of human. This results in a model that allows the CDSS system to be incrementally developed and maintained, according to the increasing complexity of human reasoning/decision making activities in the clinical process.
Chapter Preview
Top

Clinical Decision Support System

Healthcare is one of the most important issues for every country. According to a report by OECD released in 2023, between 2013 and 2022, most of the OECD countries spent approximately 8% to almost 10% of their GDP on healthcare (OECD, 2023). According to the budget spent, one of the main purpose of spending is on the issue of “quality care”. This is because maintaining quality care is not at all easy. Given any task in real-world situations, in general, regardless of how many experiences one has and how careful they are in performing the task, error can still occur. This also applies to healthcare, particularly “clinical process”.

Due to its nature, which is a highly complex and sensitive process, it has been well known that there are various issues in clinical process, specially related to safety and quality of care (Balogh, et.al., 2015)(Suutari, 2023). Poter and Lee (2013) also argue that regardless of the number of qualified personnel who are willing to take care of all patients, issues such as rising costs and uneven quality of services will remain. One of the main issues that may occur during the process is missing of important information about the patients, which can easily lead to many concerns during clinical process (Toussaint and Correia, 2018). In addition, examples of the problems that can occur during the clinical process also include improperly diagnosis the patients and give the wrong dosage of a medication (Awati, et.al., 2024).

According to the clinical process, in general, there are 6 steps in the process. Figure 1 depicts the clinic process described by Custiss (2010).

Figure 1.

Clinical Process

979-8-3693-2667-1.ch009.f01
Source: (Curtiss, 2010)

According to the 6 steps in Figure 1, every step of the clinical process can lead to issues in quality of care for patients. To reduce the amount of these errors, various approaches have been introduced and implemented, Clinical Decision Support System (CDSS) is one of them. By using data which naturally occurred during the clinical process, CDSS can gather, prepare, analyze, and visualize its findings to all personnel involved in making decisions during the process. Based on the said functions of the CDSS, several computerized clinical processes have been proposed. The following process which is referred to as “Clinical Intelligence” is proposed by THINKMD.

Figure 2.

Clinical Process with Clinical Intelligence by THINKMD

979-8-3693-2667-1.ch009.f02
Source: (THINKMD)

In Figure 2, a computerised CDSS can cover end-to-end process, from inputting data regarding current symptoms of the patients, integrating all information related to the patients including their previous visits, analyse all integrated data, and propose and visualise all possible treatments and care plan.

Top

Ai In Healthcare

In today’s world, we are now surrounded (sometime even embedded) with technologies. Changing in these technologies is getting faster and faster. In fact, recently, Intel's CEO Pat Gelsinger believes that speed of development of the technology may be faster than what Moore’s law has stated (Luke, 2023). The important result of this change is that it influences directly to our behavior. We are now becoming less patience; we require things happen in a short period time. This is the main reason why automation is coming into every part of our life. In fact, changing of our behavior even makes Google to change their vision from Mobile-First to AI-First in 2017 (Karpińska-Nowak, 2017). This change also occurred with most of the world-leading IT companies including Microsoft, IBM and Amazon. As far as healthcare business is concerned, Holley and Becker (2021) discuss their ideas on how AI-First concept can be applied in clinical process and healthcare in general.

Key Terms in this Chapter

Generative AI: A type of Artificial Intelligence (AI) systems that particularly focus on creating new content, such as text, images, music, or other data, by learning patterns from existing data.

Clinical Decision Support System (CDSS): computer-based programs that analyze clinical-related information and, based on its data and result of analysis, used in decision-making processes of healthcare practitioners.

Multi-Agent System (MAS): A multi-agent system is a computational system in which multiple autonomous entities, called agents, interact and collaborate to solve complex problems or achieve specific goals.

Complete Chapter List

Search this Book:
Reset