Abstract:
Diabetic is a chronic human disease. It affects blood glucose in the human body.
Imidazole (C3N2H4) is an organic compound. Imidazole has many properties
such as antituberculosis, antimicrobial anti-inflammatory, antimycotic, antitumor,
and antiviral in the human body. By utilizing Fourier transform infrared spectroscopy
(FTIR), recent review represent antidiabetic action of inadvance synthesized
Metronidazole (MTZ) complexes (MTZ-Benz, MTZ-Benz-Cu, MTZ-Benz-
Cu-Cl2CHCOOH, MTZ , MTZ - Cu , MTZ - Cu - Cl2CHCOOH , MTZ - Benz -
Ag , MTZ - Benz - Ag - Cl2CHCOOH , MTZ - Ag and MTZ - Ag - Cl2CHCOOH)
opposed to consequence of IC50 (μg/mL) Values of substances for α-glucosodase
and α-amylase inhibition assays. Two methods were used in this research named
Artificial Neural Network and Partial Least Square. For correcting the spectrum
baseline Asymmetric Least Squares were used. In fitted models, Artificial Neural
Networks outperform cross-validated root mean square error. Influential wavelengths
that explain variance in the antidiabetic activity of Metronidazole compounds
were found and mapped against functional groupings using Artificial neural
networks.
