A Comparative Evaluation of Valine-Leucine Amino Acid Impact for Disease Identification using Fuzzy C-Means Clustering and KNN Classification

Amino acids are essential components that are necessitated for the human body, and contributes in balancing the health of an individual. Nonetheless, their proportion either higher or lesser can cause health gremlins that can be detrimental. The foremost health problems stems in the form of heart ailments, diabetes and bone relevant issues. While many studies have highlighted the rudimentary processing techniques, and subsequently identified the accuracy measures with respect to classification, the neoteric times have evinced that lack of exercise, and the levels of stress that can be cardinal factors in escalating the disease triggering cells in the human body. With the emphasis of Leucine and Valine amino acids which are specifically analyzed in this study, the importance to stay healthy, and the changes observed with the quantity change in the amino acids provide a wholistic picture of sustainable living. The previous research pertaining to health gremlins and medical data mining have always lacked in understanding the consumption of amino acids, but have focussed to render more priority to the stratification of diseases. Thus, circumventing to analyse the root-cause of any health issue that an individual can potently be challenged with. Thus, this paper proposes the analysis of disease data processing in relevance to heart attack and osteoporosis from valine and leucine approximations. This study takes into consideration an algorithmic approach of processing the valine-leucine data as signals, to explicitly filter and classify them based on the amino acid and corporeal value count of the individual. The results in procuring the classification accuracy of an individual depends on the muscle-vein thickness, Total Cholesterol count from the corporeal parameters and the valine-leucine levels to determine the triggering of blocks and fragility of muscles in the human body. The methodological approach of entailingPeak Signal-based saddle mitigation technique incorporated with fuzzy-C Means clustering, and KNN classification is used to identify the diseases in an individual. The simulations thus carried out, rendered successful results in MATLAB GUI, therebyaiding to provide an explicit comprehension of the occurrence of heart attack and osteoporosis in accordance to the consumption levels of the valine-leucine amino acids.