|One area where human intelligence is limited is in the area of medicine. It often takes a doctor to determine a medical condition, and doctors cannot be everywhere. However, if the knowledge and processes that doctors use can be transmitted to machines, the knowledge and diagnosis of the medical field can be greatly expanded.
Discuss how expert systems, neural networks, and genetic algorithms can help scientists meet medical challenges. Provide an example of each system. Select one of your three examples and describe the system more in-depth. Give an example of a system currently being used and describe how and where it is in use today.
Use the class lectures and textbook as sources of reference.
Title: Essentials of Management Information Systems
Author: Kenneth Laudon; Jane Laudon
Edition / Copyright: 11
Publisher: Pearson Education
Meeting Medical Challenges with Artificial Intelligence
The use of artificial intelligence in medicine remains widely unexploited topic. The biggest constraint has been the process of identifying areas in medicine that are best suited for the application of artificial intelligence concepts. However, system developers have identified this gap and are now concentrating on looking for functional systems that are effective in terms of long-term application within specific medical areas. The three main systems that have been developed with the aim of aiding in the clinical diagnosis and monitoring process include expert systems, neural networks and genetic algorithms (Laudon & Laudon, 2013). With proper technological focus, these systems can be deployed successfully to address most of the medical challenges that persist in the present world.
Expert systems are increasingly being applied in clinical laboratories, where they have made the interpretation and organization of medical results easier and more efficient (Wagman, 2000). One example of this application is the PUFF system, which has been used on a commercial level for the interpretation of pulmonary tests results. It is one of the first intelligent systems to be developed and which continues to gain popularity that is being spurred by global acceptability, simplicity and accuracy. Other examples of this approach include Germwatcher and PEIRS systems. Expert systems are typically function independently of other medical structures in the sense that they do not interfere with the tasks being undertaken by healthcare personnel. Therefore, the existing health practitioners still exercise great control over information collection, examinations and treatment.
Similarly, neural networks are extremely helpful in the medical field particularly in situations where numerous challenges arise due to the absence of algorithm solutions or in situations where the existing ones are too complex. So far, numerous neural networks have been successfully implemented in various areas of diagnosis and image analysis (Quinlan, 2003). For example, the NeuroXL Predictor and Clusterizer models are being widely used in combination with Excel platforms for clustering tasks. They simplify processes and project networks, making them easy to comprehend and use in solving algorithms. More specifically, they lead to a major cut-down of the entire duration it would otherwise take to complete processes, meaning that the entire process becomes extremely easy to execute with a high level of accuracy without necessarily compromising on quality. Moreover, these networks can be used to find a solution even in conditions of incomplete data. They mirror the human brain and are therefore able to solve complex problems including some of the ones in which humans tend to excel more than computers. At the same time, they are exempted from human shortcomings such as fatigue and emotional factors.
Elsewhere, genetic algorithms are being developed for use in the areas of pediatrics, surgery, neurology, radiology and health management. They offer extremely precise solutions to complex challenges that would otherwise have remained unsolved. Hardware devices with software counterparts made up of algorithms are the main forms of genetic algorithms being used today (Mitchell, 1998). However, genetics, remains a greatly complex area of research. Nevertheless, genetic algorithms are increasingly being used to break down the process of studying genetic factors as well as altering, improving and further monitoring related medical processes.
In terms of application, genetic algorithms are useful for identifying solution points and then evaluating each of them based on accurate methods of probability and random generators (Mitchell, 1998). Their application in oncology has led to the emergence of a new method of early cancer detection. An example is a non-invasive cervical cancer screening in which information generated is analyzed under a genetic algorithm system to differentiate normal tissue cells from cancerous ones. Besides, scientific advancements in DNA profiling has streamlined the process of genetic algorithms through accurate analysis. Furthermore, the approach has made it more possible to identify other facilitating factors in human body thereby creating an in-depth understanding of the rationale for the development of various genes and cancer cells. Using the genetic algorithm approach, medical practitioners are able to selects the most crucial variables based on which outcome of various treatments can be predicted.
Of the three aspects of artificial intelligence being used in medicine, genetic algorithm seems to have the most valuable application because of its role in early cancer detection. It is an extremely powerful tool that touches on the basic foundation of medicine: genetic composition and structure. Moreover, it is a cost-effective and sustainable manifestation of artificial intelligence that can be suitably integrated in medicine without interfering in other human functions. Meanwhile, further research is needed to strengthen these technological aspects of artificial intelligence for the benefit of humankind. Similarly, the ongoing application of artificial intelligence through expert systems and neural networks has also played a crucial role in addressing contemporary medical challenges.
Laudon, K. & Laudon, J. (2013). Essentials of Management Information Systems. Boston, MA: Pearson.
Mitchell, M. (1998). An Introduction to Genetic Algorithms. Cambridge, MA: MIT Press.
Quinlan, P. (2003). Connectionist Models of Development. Hove: Psychology Press.
Wagman, M. (2000). Scientific Discovery Processes in Humans and Computers: Theory and Research in Psychology and Artificial Intelligence. Westport, CT: Praeger.