Análise de triagem para confirmação do estado de conservação de insulina humana NPH injetável

• Ano de defesa: 2020
• Autor(a) principal: Silva, Suelly Fernandes da
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Química; Programa de Pós-Graduação em Química; UFPB
• Palavras-chave em Português: Química; Controle de qualidade; Insulina; Espectroscopia NIR; Quimiometria; One class; Chemistry; Quality control; Insulin; NIR spectroscopy; Chemometrics; CNPQ::CIENCIAS EXATAS E DA TERRA::QUIMICA
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/123456789/20682

Insulin is a hormone produced by the pancreas and is necessary for the use of glucose (sugar) by the body as an energy source. When the pancreas does not produce enough insulin to control blood sugar, diabetes occurs. The treatment of diabetes can be using various types of insulin according to the patient's need. Insulins have stability, good stability and their biological action is preserved, as long as conservation and transport guidelines are followed. Therefore, the quantification and quality control of this medication is of great importance, so that it can be used effectively and safer since patients can be harmed by varying its effectiveness. Therefore, this work proposes the development of an analytical methodology for the quality control of NPH insulin, without the violation of the packaging, using spectrometry in the near infrared (NIR) region with a bench and a portable equipment, associated with chemometric methods of pattern recognition. Classifiers “one class” considered Data Driven-Soft Independent Modeling of Class Analogy (DD-SIMCA) and One-Class Partial Least Squares (OC-PLS) were used as pattern recognition tools. In terms of classification performance, the best result observed was using DD-SIMCA for Savitzky-Golay smoothing pre-processing (with a 41-point window) with baseline offset (BO) reaching maximum level in the discrimination of the samples in the bench NIR. For the portable NIR, both DD-SIMCA and OC-PLS reached maximum performance levels, correctly classifying of the samples using data pre-processing: multiplicative scatter correction (MSC) with BO correction and linear baseline correction (LBC) with MSC, respectively. Therefore, the proposed methodology represents a promising tool, non-destructive, fast, and low-cost for the screening of NPH human insulin.