| Grant number: | 18/08692-2 |
| Support type: | Research Grants - Innovative Research in Small Business - PIPE |
| Duration: | February 01, 2019 - December 31, 2019 |
| Field of knowledge: | Engineering - Biomedical Engineering - Medical Engineering |
| Principal Investigator: | Pietro Teruya Domingues |
| Grantee: | Pietro Teruya Domingues |
| Company: | Indigo Labs Ltda |
| CNAE: |
Fabricação de aparelhos eletromédicos e eletroterapêuticos e equipamentos de irradiação Manutenção e reparação de equipamentos eletrônicos e ópticos Serviços de engenharia |
| City: | São Paulo |
| Co-Principal Investigators: | Lincoln Makoto Kawakami |
| Assoc. researchers: | Alessandra Soubhia Gil Maldonado ; Eduardo de Senzi Zancul ; Flávia Helena da Silva |
| Associated scholarship(s): | 19/05008-6 - Photometric and computer vision techniques applied in pre-analytical phase of blood samples testing, BP.PIPE |
Abstract
One of the main goals of laboratory diagnostics is to provide health professionals and patients with reliable medical reports, based on safe and efficient procedures. It is estimated that approximately 70% of all diagnoses are made based on laboratory tests. The results of these tests are responsible for 60 to 70% of all the decisions about patient admission, discharge, and therapeutic regimen. In this context, errors caused by misdiagnoses can have serious consequences that threaten patient's health and are one of the leading causes of legal disputes in the USA. Clinical errors reduction has vital importance in the operation of health service providers. Laboratory diagnostics is responsible for the whole chain of activities within the laboratory and the tools and devices used for the reduction of these errors. Among laboratory errors, at least 50% occur in the pre-analytical phase. This phase consists of receiving and evaluating rejection/acceptance criteria for the samples, among other activities. This project focuses on a solution that automates pre-analytical phase activities that are traditionally manual. Research in photometric technologies (cameras, light sensors, spectrometers, etc.) will be conducted, alongside techniques in computer vision, machine learning, and pattern recognition algorithms. This project aims to provide an innovative solution capable of analyzing blood samples with higher reliability and precision than the manual methods currently used, as well as to provide greater accessibility compared to the traditional devices. To the moment of the submission of this proposal, initial research and technological prospection has been done with the support of a laboratory chain, called Grupo Fleury. The result of the initial research is a prototype that calculates the Hemolysis Index (HI) of blood samples (HI is an essential rejection criteria for blood samples) in an automated, non-invasive way (i.e., does not have direct blood contact). This prototype has shown adequate performance. It was considered that it has potential to be applied in laboratory workflow. Further research and development to be conducted in this project include improving the precision and consistency in the evaluation of Hemolytic Index, and the detection of other parameters of interest, such as icterus and lipemic turbidity, resulting in a broader and more reliable technological solution. (AU)