Redes neurais artificiais nomonitoramento de inibidores de tirosina quinase na leucemia mieloide crônica

• Autor(a) principal: Albuquerque, Patrícia Maria Simões de
• Data de Publicação: 2019
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UFPB
• Texto Completo: https://repositorio.ufpb.br/jspui/handle/123456789/19379
• topic: Leucemia mieloide crônica; Inibidores de tirosina quinase; Descontinuação do tratamento; Mapas auto-organizados; Eventos adversos; Chronic myeloid leukemia; Tyrosine kinase inhibitors; Discontinuation of treatment; Self-organized maps; Adverse events; CNPQ::CIENCIAS BIOLOGICAS::FARMACOLOGIA

The research aimed to build a computational intelligence model to monitor the treatment of patients with chronic myeloid leukemia (CML) with tyrosine kinase inhibitors (TKI). It is characterized as a clinical, observational and longitudinal study, based on institutional data whose results were analyzed, based on information obtained from the analysis of each participant's medical records, which was monitored for 12 months. The sample consisted of 105 patients from the outpatient chemotherapy sector of a hospital in the city of João Pessoa-PB, from September 2015 to September 2016. For data collection, three instruments were used: sociodemographic, clinical and therapeutic data; patient follow-up corresponding to adverse reactions; and the one containing the World Health Organization Quality of Life (WHOQOL-bref) questionnaire, to assess quality of life. The database was obtained, and a total of 689 variables submitted to self-organized mapping (SOMs) studies. From unsupervised machine learning techniques for CML patients in groups based on specific variables, non-serious adverse events were observed in patients treated with imatinib and dasatinib and probable causes responsible for unintentional treatment disruption: cutaneous hypopigmentation, degrees vomiting, degrees of orbital edema and degrees of tearing, vomiting, diarrhea, fatigue and hand and foot syndrome. The literature shows that adverse events (AEs) are significantly related to the patient's quality of life, interfering with their daily life and negatively affecting adherence to therapy as shown in this study. Thus, there is a need to update SOM models using new data to improve robustness in predicting treatment discontinuation due to adverse events and to identify key factors for treatment failure.