Fuzzy Computing in Healthcare

• Published in: 2024 International Visualization, Informatics and Technology Conference (IVIT) pp. 78 - 83
• Main Authors: Khushal, Rabia, Fatima, Ubaida
• Format: Conference Proceeding
• Language: English
• Published: IEEE 07.08.2024
• Subjects: Computational intelligence; Fuzzy Computing; Healthcare Dataset; Healthy lifestyle; Industries; Informatics; Input variables; Machine learning; Medical services; Nearest neighbor methods; Osteoporosis; Random forests; Uncertainty; Visualization
• DOI: 10.1109/IVIT62102.2024.10692652

Fuzzy computing, a computational intelligence technique can be employed in the healthcare industry for numerous purposes. The dataset available consists of several features that are equivalently important for the prediction or diagnosis however inclusion of all important features takes a large computation time. Application of fuzzy computing compares the binary variable features thus providing a vast set of values for the prediction of output moreover since two features are transformed into one the number of features is reduced hence taking less computation time for the computation of output. A healthy lifestyle is crucial to live a longer life. In this research work, fuzzy computing is applied to the healthcare dataset which is the healthy lifestyle dataset. Machine learning techniques which are random forest, k-nearest neighbors, and adaptive boosting are applied to the normal dataset and then these techniques are applied to the fuzzy modified dataset thus a fuzzy computing model is proposed in which machine learning techniques are utilized for inference and generation of fuzzy output that is defuzzified subsequently and provides crisp output. Machine learning techniques without fuzzy computing take more computation time for computation of the output moreover it doesn't consider the uncertainty of the binary input variables. However, machine learning techniques with fuzzy computing lead to a reduction of computation time and consideration of more features which provide a vast set of values for diagnosis of whether the individual is living a healthy lifestyle or unhealthy lifestyle.