Question 1 (1 point)
What are neural networks, what is their purpose, and what practical applications for these networks in medicine can you think of? Be brief in terms of the applications and examples, but you must be able to introduce and explain the concept of a neural network, clearly showing your understanding of the theory and briefly showing your practical vision.
Question 2 (1 point)
Explain the concept of inference in general terms and focus on the concept of Bayesian network model. Explain theory behind Bayesian algorithms and provide one or two examples of their practical use in clinical decision support systems. Draw from your professional experience in healthcare, or previous biomedical informatics courses, or literature (textbooks, journal articles, reputable online sources).
Question 3 (1 point)
Explain decision trees and draw a simple decision tree in a software program of your choice (Power Point, Vision, etc). Come up with a simple (or more complex, if you wish) tree describing a medical or operational decision you are familiar with. The target is 8-10 decision points/branches in the decision tree, but you are welcome to take this assignment as far as you wish.
Question 4 (1 point)
Pick a basic patient condition you are familiar with or research one using library resources. Draw a basic diagram showing inference engine based on Bayesian model for this patient condition, similar to the pneumonia example in Figure 2.2, page 35 in the Berner text.
Question 5 (1 point)
You continue working with the same patient condition you defined as part of your answer to Question 4 in this quiz. Define a set as a unique collection of related objects under Set Theory for this condition. Write the set using syntax defined in the textbook.
Note 1: While this is not a research assignment, please remember to cite any references you decide to use in proper APA style.
Note 2: Remember that wiki/user collaborative sites such as Wikipedia and blogs are examples of non-acceptable online sources, as these are not peer reviewed and cannot be verified for accuracy. Your best bets for web-based sources are online editions of professional and academic journals, as well as government and medical informatics professional association publications.