Back to search

National and Kapodistrian University of Athens

Εθνομεθοδολογική έρευνα σε εργαστήριο επεξεργασίας και ανάλυσης ιατρικής εικόνας για την ανάπτυξη συστημάτων υποβοηθούμενης διάγνωσης με τεχνητή νοημοσύνη: μια προσέγγιση βασισμένη στην αυτό-εκπαιδευτική πρακτική

Abstract

dc:description

This dissertation is an ethnomethodological laboratory study conducted in a research laboratory specializing in the development of computer-aided diagnosis (CAD) systems. The aim of the study is to describe a contemporary scientific field, that of Biomedical Engineering, through the practices that constitute it, as well as to articulate how these practices are involved in research work. The study focuses on describing my self-instructive practice based on established research practices of the laboratory members. Self-instructive practice has been developed within the ethnomethodological tradition, with its main characteristic and prerequisite being the social researcher’s own education in the subject she intends to study.The academic discipline of this dissertation is the field of Science, Technology, and Society (STS), as well as the social sciences more broadly. Specifically, as I present in the first chapter of the dissertation, I study and draw research tools from a body of literature concerning the study of practice in scientific work, with an emphasis on ethnographic and ethnomethodological approaches, such as multimodal video analysis and conversation analysis. In this vein, the primary methodology employed in this dissertation is the social researcher’s own practice of self-training in the practice under study. This approach, by explicitly highlighting the analyst’s practical experience, aims to capture the technical details of the practice under study through a description of the ways in which it is constituted. My research was conducted at the Medical Signal and Image Processing Laboratory of the Department of Biomedical Engineering at the University of West Attica. It was primarily based on my self-directed study in a series of laboratory exercises that are part of research projects aimed at developing algorithms to assist in medical diagnosis through so-called Decision Support Systems. As part of my self-directed learning, by following part of the practical training of biomedical engineering students, I was able to participate in the same practical activities. Specifically, I was trained in the observation and creation of histopathological images using a digital microscope, as well as in the Matlab programming language. I also participated in a microscopy project aimed at visualizing isolated nuclei in brain histopathological images using an image processing algorithm based on segmentation techniques.The first research chapter, which corresponds to the second chapter of the dissertation as a whole, concerns the analysis of a video recording from the laboratory work on the digitization of histopathology images and the creation of regions of interest (ROI). For this analysis, I relied on Conversation Analysis tools. Specifically, I examined how scientists, during collaborative work on creating regions of interest, understand, shape, and classify their experimental data through the engagement of their bodies with digital technology. In the following chapters, I shift methodologically from observing the practice under study to my embodied engagement with it. My practice at the microscope and the creation of digitized images from histopathology slides constituted the subject of the third chapter of my dissertation. Through my hands-on engagement and the difficulties, I encountered—which concerned the handling of the microscope (focusing, cleaning the lenses, and so on), a series of tactile techniques emerged that the dexterity of experienced researchers rendered invisible.In the fourth chapter, I describe my self-study process in the Matlab programming language. A significant portion of the chapter is devoted to my efforts to solve exercises in the Matlab programming language. Specifically, I describe the step-by-step process of solving them as a beginner user of the language, emphasizing my mistakes and omissions and thus bringing to light those technical practices that, while fundamental to writing code, experienced programmers tend not to mention them, taking them for granted; a typical example is copy/paste, commenting, running the code, and searching the internet. The fifth and final chapter of the thesis presents the programming component of the microscopy project. The ultimate goal of this project was the digital visualization of isolated nuclei in brain cancer cells using training code. My involvement in this type of project with clear objectives was a conscious choice. This is because my aim was to describe the effort to identify and implement the appropriate technical practices for carrying out a specific task. The aim of such a description is to present a research process in a way that makes it understandable—through the details of the practices themselves—how researchers carry out the individual practices from the perspective of experimental adequacy, rather than presenting a mere description of technical correctness. Through this engagement, I formulated a description based on my evolving understanding of the subject under study. The main objective was to make the practice under examination explicit in its own terms—that is, through its identifying details—while also studying the various ways in which its observation is achieved. During my practice—that is, when I carried out the processes for isolating the nuclei, and specifically during the introduction of the thresholds, a procedure I repeated several times— I described the segmentation practices through the evolution of my understanding of them. I was able to articulate the answer to why, how, and when I used the low-pass filter and simple thresholding with a threshold of 13. Thus, this dissertation has made possible a description that both explains and contributes to the practical implementation of nuclei isolation.

Degree

thesis:*
Grantor dc:publisher
National and Kapodistrian University of Athens
Year dc:date
2025

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Kandaraki, Aspasia
  • Κανδαράκη, Ασπασία

Subjects

dc:subject × 9

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/61511

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Kandaraki, Aspasia; Κανδαράκη, Ασπασία. Εθνομεθοδολογική έρευνα σε εργαστήριο επεξεργασίας και ανάλυσης ιατρικής εικόνας για την ανάπτυξη συστημάτων υποβοηθούμενης διάγνωσης με τεχνητή νοημοσύνη: μια προσέγγιση βασισμένη στην αυτό-εκπαιδευτική πρακτική. National and Kapodistrian University of Athens, 2025. http://hdl.handle.net/10442/hedi/61511