A Study on Medical Diagnosis Using Neutrosophic Over Complex Sets and Artificial Neural Networks Under Uncertain Environments

Vollständiger Abstract

Worum geht es in dieser Arbeit?

The present article introduces the idea of Neutrosophic Over Complex Sets, and in a way it kind of explores how they can be used for medical diagnosis tasks that live in uncertain and inconsistent environments. The proposed framework mixes the benefits of neutrosophic sets, over sets, and also complex-valued information into one single mathematical structure, so it can manage indeterminate, contradictory and over-valued complex data without too much fuss. In the proposed setup, the truth-membership, indeterminacy-membership and falsity-membership functions are written in a complex-valued form, and their amplitudes are greater than one, which is what basically makes the "over" condition hold. After that, a number of algebraic operations plus properties of Neutrosophic Over Complex Sets are defined and checked mathematically, like in a proper verification way. Then, there is an application: a medical diagnosis scheme is developed, for picking out the most critical patient from among a group of patients dealing with a serious viral disease. The evaluations coming from multiple experts are expressed through Neutrosophic Over Complex information, and they are aggregated using the set operations suggested in the framework. A score function is also used to arrange, and rank the patients based on their severity levels. Moreover, an Artificial Neural Network based approach gets added, mostly to validate the ranking results that the mathematical model outputs. The ANN model ends up producing the same ranking order as the proposed model, so it supports both reliability and computational efficiency of the whole framework. Overall, the results suggest that the Neutrosophic Over Complex Set model offers a flexible, realistic, and efficient mathematical tool, for tackling advanced decision-making and medical diagnosis problems where uncertain over complex information is involved.