A Journey trough Clinical Applications of Multimethod Decision Trees

Determining risk factors for survival after LMCA stenosis with intelligent data analysis

“…Thus we decided to design a new approach that enables tight tangling of single methods. This approach is called a multimethod approach [4]. Opposed to the conventional hybrids our idea is to dynamically combine and apply different methods in not predefined order in the manner to solve a single problem or the decomposition of that problem.…”

“…Various methods such as neural networks, decision trees, genetic algorithms, hybrid approaches, etc. have been developed and evaluated on different medical databases [3][4][5][6]. However, from physician's point of view the ability to track and evaluate every step in the decision making process is the most important factor for trusting the decisions gained with machine learning methods.…”

Intelligent Data Analysis for the Diagnosis of Alcohol Dependence Syndrome

“…Genetic algorithms are based on the evolutionary ideas of natural selection and genetic processes of biological organisms. 26,29 They are often capable of finding optimal solutions even in the most complex search spaces, or at least they offer significant benefits over other search and optimization techniques. The first phase of the genetic process is the generation of an initial population.…”

“…Decision trees have been successfully used for years in many decision-making applications. 28,29 One of the main advantages of using decision trees, when D Dinevski, P Povalej, M Kravos Diagnosing alcohol dependence by intelligent data analysis compared with other methods of machine learning, is that they provide a very simple and clear representation of the path taken to reach the acquired decision. Inducing a decision tree is a form of machine learning, where knowledge is extracted from a set of examples (objects) and presented in the twodimensional form of a decision tree.…”