Implementation of Computer Vision and Image Processing in Technovigilance at the National Institute of Cardiology Ignacio Chávez: Adenosine Vial Case

This article addresses a case of application of computer vision along with image processing for the detection of elastomers inside syringes (Coring effect) used in the administration of adenosine to patients of the service 1 of the Ignacio National Institute of Cardiology Chávez. An experimental setup consisting of a 0.3 megapixel camera, computer and MATLAB software is used for the detection of an elastomers of approximately 1 mm, being almost imperceptible to the human eye, also, we analyze the puncture area in adenosine vial and the puncture technique. An experimental setup is developed with a 0.3 megapixel camera (resolution of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{640}\times \mathbf{480}$</tex>) and processing in MatLab which allows to determining the puncture area in the adenosine elastomer vial, as well as detecting the dimensions of the fragments present within the analyzed syringes at a distance of 0.05 m. This tool is characterized with a standard instrument (vernier), finding the following characteristics: resolution of 0.08 mm, precision of 98%, accuracy of 97% and an error of +/- 0.2 mm. These figures of merit are appropriate for application in the present case. Additionally, we use a similar setup to determine the puncture angle at a distance of 0.3 m. This tool is characterized with a standard instrument (goniometer), finding the following characteristics: resolution of 1°, precision of 98%, accuracy of 95% and an error of +/- 2°. These figures of merit are suitable for application in the present case. Also, note that accuracy and error vary depending on the caliber of the needle. Thanks to the use of these techniques, it was possible to detect the dimensions and shapes of the elastomer fragments detected inside the syringes, as well as the analysis of the puncture technique. The use of computer vision and image processing techniques turns out to be a quantitative, objective tool with minimal alteration of the evidence, which allows us to conclude that the use of an incorrect puncture technique and incorrect needle calibers could be the main causality of the Coring effect. The results are promising for their implementation in future cases in technovigilance, as well as in the clinical area of medicine to minimize errors in medical practice, safeguarding patient safety, as well as in other areas of science. Additionally, it is a technique that complies the current regulations in which technovigilance is established in Mexico.