Tehnike umetnega povečanja testnih podatkov za klasifikacijo EKG signalov

Lahovnik, Zala

| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

First page > Show document

Show document

Title:	Tehnike umetnega povečanja testnih podatkov za klasifikacijo EKG signalov
Authors:	ID Lahovnik, Zala (Author) ID Podgorelec, Vili (Mentor) More about this mentor...
Files:	MAG_Lahovnik_Zala_2025.pdf (3,49 MB) MD5: C33EBF520663FF6BC3A829BF32B8ACBA
Language:	Slovenian
Work type:	Master's thesis/paper
Typology:	2.09 - Master's Thesis
Organization:	FERI - Faculty of Electrical Engineering and Computer Science
Abstract:	Atrijska fibrilacija je ena najpogostejših in najnevarnejših srčnih aritmij, njena pravočasna identifikacija pa je ključna za preprečevanje resnih zapletov. Kljub razvoju metod strojnega učenja ostaja pomanjkanje kakovostnih in strokovno označenih EKG podatkov, ki jih je težko pridobiti brez kardioloških ekspertov, velik izziv. V zaključnem delu smo zato preučili tehnike umetnega povečanja testnih podatkov, pri katerih model med napovedovanjem obravnava več spremenjenih verzij istega signala. Z eksperimentom, ki je vključeval različne kompleksnosti modelov, pristope učenja in načine odločanja, smo pokazali, da tehnike umetnega povečanja testnih podatkov delno prispevajo k boljši napovedi.
Keywords:	umetno povečanje testnih podatkov, EKG signali, klasifikacija
Place of publishing:	Maribor
Place of performance:	Maribor
Publisher:	[Z. Lahovnik]
Year of publishing:	2025
Number of pages:	1 spletni vir (1 datoteka PDF (XIII, 81 str.))
PID:	20.500.12556/DKUM-96080
UDC:	004.8:616.12-037.7
COBISS.SI-ID:	266983939
Publication date in DKUM:	22.12.2025
Views:	113
Downloads:	81
Metadata:
Categories:	KTFMB - FERI
:	Copy citation

Average score:	(0 votes)
Your score:	Voting is allowed only for logged in users.
Share:

Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:	CC BY-NC-ND 4.0, Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International

Link:	http://creativecommons.org/licenses/by-nc-nd/4.0/
Description:	The most restrictive Creative Commons license. This only allows people to download and share the work for no commercial gain and for no other purposes.
Licensing start date:	01.12.2025

Secondary language

Language:	English
Title:	Test time augmentation for classification of ECG signals
Abstract:	Atrial fibrillation is one of the most common and dangerous cardiac arrhythmias, and its timely identification is key to protecting against serious complications. Despite the development of machine learning methods, the lack of high-quality and professionally labeled ECG data, which is difficult to obtain without cardiology experts, remains a major challenge. As part of our final work, we examined artificial augmentation techniques for test data, in which the model considers multiple modified versions of the same signal during prediction. With an experiment that included different model complexities, learning approaches, and decision-making methods, we showed that artificial augmentation techniques for test data partially contribute to better predictions.
Keywords:	test time augmentation, ECG signals, classification

Comments

Leave comment

You must log in to leave a comment.

Comments (0)

0 - 0 / 0

There are no comments!

Back