Look-up in Google Scholar
Title: Automated on-screen eye movement analysis for the evaluation of medical device usability
Other Titles: Análisis automatizado del movimiento ocular en pantallas para la evaluación de la usabilidad en aparatos médicos
Advisor(s): Wang, Felix
OCDE field: https://purl.org/pe-repo/ocde/ford#2.03.01
Issue Date: 2021
Institution: Eidgenössische Technische Hochschule Zürich
Abstract: El objetivo de la tesis fue optimizar el algoritmo automated Dynamic AOI Mapping (aDAM). aDAM es un algoritmo de visión por ordenador para la detección de seguimiento de objetos mediante identificación de líneas e intersección de esquinas. El objetivo de la optimización fue rastrear objetos de dimensiones reducidas y realizar pruebas con ellos.

Usability Testing is indispensable in the development of medical devices. It validates intended functionality, guaranties patient safety, and provides a technological advantage. In this thesis, the performance of the automated Dynamic AOI Mapping algorithm (aDAM) is evaluated on two small medical devices. Its functionality is tested to see if it is suitable for its use in the growing field of eHealth applications. The performance showed very promising results. In the area of tracking, with the correspondent adjustments, a correct gaze point mapping can be achieved up to 98% of the cases. For Screen Matching purposes, a highly accurate matching can be acquired if the device plays a main role in the analyzed video. In case the images are from very low-resolution, the algorithm does not function correctly. Nevertheless, this algorithm performs well, and it can be used for the analysis of small medical devices.
Discipline: Ingeniería Mecánica
Grade or title grantor: Eidgenössische Technische Hochschule Zürich
Grade or title: Bachiller en Ciencias ETH en Ingeniería Mecánica
Juror: Wang, Felix; Meboldt, Mirko
Register date: 12-Jul-2022

Files in This Item:
File Description SizeFormat 
GuijaDezaCA.pdfTesis1.47 MBAdobe PDFThumbnail
View/Open
Autorizacion.pdf
  Restricted Access
Autorización del registro150.35 kBAdobe PDFView/Open Request a copy


This item is licensed under a Creative Commons License Creative Commons