Prognostic system for clinical classification of liver disease using fuzzy cluster means

Clinical management of liver diseases and severe disorders in liver tissue involves analysis of blood enzymes, usually from the blood count, as well as a functional test of the liver (LFT). The functional precision of the existing system is quite limited due to the complexity of clinical data and the lack of multi-class elements in decision support. In this study, quantitative and experimental designs were employed, incorporating fuzzy cluster means (Fuzzy-CM), for predictive segmentation of liver disorders as a data mining model. One thousand three hundred forty-five (1,345) instances of liver data were collected from the clinical repository. Experimental analysis of classifier (Fuzzy-CM) was done in matrix laboratory (MATLAB) and revealed that the classified clusters of over ninety percent (90%) of concerned patients (test cases) conform to fuzzy inference, which yielded improved precision and classification accuracy based on multi-class parameters of the clinical domain. Predictive projection was achieved through an intelligent classification model that handled the ambiguities of clinical data for decision support with 91.3% accuracy and 93.3% precision. The prognostic system, as a technological artifact, is recommended for adoption in the clinical management of liver diseases to facilitate the timely detection and implementation of therapeutic measures.