Sistema de software para análise de parâmetros sanguíneos de forma não invasiva
Clinical analysis laboratories can diagnose, track, and monitor diabetes mellitus through tests that use patient collection methods that are uncomfortable for the patient because they are invasive and expose the analyst and the environment to contamination risks. Therefore, the development of spectr...
| Autor principal: | Carvalho, Steven Kariston Louback de |
|---|---|
| Formato: | Dissertação |
| Idioma: | Português |
| Publicado em: |
Universidade Tecnológica Federal do Paraná
2020
|
| Assuntos: | |
| Acesso em linha: |
http://repositorio.utfpr.edu.br/jspui/handle/1/4987 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: |
Clinical analysis laboratories can diagnose, track, and monitor diabetes mellitus through tests that use patient collection methods that are uncomfortable for the patient because they are invasive and expose the analyst and the environment to contamination risks. Therefore, the development of spectroscopy-based forms of clinical analysis for less invasive analyzes could reduce the number of conventional blood tests and bring advantages such as low cost, operational simplicity, minimal sample preparation (when necessary), no waste generation, as well as reducing analyst exposure to biological material. To promote this development, preliminary tests for spectroscopic blood tests were performed directly on the oral mucosa of patients, and the results obtained were very encouraging. However, despite the excellent expectation, it was observed the dependence of human interventions in the process that goes from the acquisition of spectra until the answers of interest are provided, delaying their application. Considering the possibility of optimizing the process, this work proposes a software system to integrate the process from the spectral measurement to the quantitative response on the glycemic content. The proposed analysis process through the software begins with the automated integration of the equipment responsible for obtaining near-infrared spectral measurements with a smartphone or personal computer. A client module of this software will be responsible for transmitting the information collected by the equipment to a computational cloud, in which a responsible web service performs multivariate calibration (by Partial Least Squares). This web service enables the use and updating of a multivariate calibration model, responsible for establishing the correlation between the obtained spectra and conventional measurements, allowing
real-time monitoring of patients and, consequently, more assertive interventions. Thus, the diagnosis of diabetes mellitus can be made by a less uncomfortable and extremely agile methodology, making it accessible to a larger number of people and being even more representative by allowing daily measurements. It is also emphasized the sustainable aspect of the methodology that, by not generating biological waste and hospital waste such as needles and syringes, directly benefits the environment, besides the possibility of expansion to measure other blood parameters. |
|---|