Sistem za beleženje zgodovine ročnega urejanja rezultatov dekompozicije večkanalnih površinskih elektromiogramov

diplomsko delo

Nina Murks (Author), Aleš Holobar (Mentor)

Abstract

Namen magisterska dela je izdelati sistem za shranjevanje zgodovine ročnega urejanja rezultatov dekompozicije večkanalnih površinskih elektromiogramov ter prikaz njegove uporabe. Področje obsega predstavljene strategije urejanja rezultatov dekompozicije, opis delovanja ter implementacije sistema za shranjevanje zgodovine in njegovo uporabo. Prva predstavljena uporaba sistema za beleženje zgodovine je izračun in prikaz statistike urejanja, s pomočjo katere je možno vrednotiti strategije urejanja. Drugi primer uporabe sistema za beleženje zgodovine je preprost primer delne avtomatizacije urejanja, pri čemer sta uporabljena dva različna modela nevronskih mrež. Prvi model vsebuje konvolucijske sloje, drugi pa sloje LSTM. Preučili smo uspešnost obeh modelov ter prikazali njune rezultate. Model s konvolucijskimi sloji je dosegel 71-% preciznost ob 83-% priklicu napovedi urejanja, model s sloji LSTM pa 100-% preciznost ob 75-% priklicu.

Keywords

beleženje zgodovine;rezultati dekompozicije;statistika urejanja;delna avtomatizacija urejanja;magistrske naloge;

Data

Language:	Slovenian
Year of publishing:	2022
Typology:	2.09 - Master's Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	[N. Murks]
UDC:	004.9:621.391(043.2)
COBISS:	113787395
Views:	120
Downloads:	27
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Logging of the manual editing of high-density surface electromyogram decomposition results
Secondary abstract:	The master's thesis aims to develop a logging system for the manual editing of high-density surface decomposition results and to demonstrate its use. The scope of the work includes the proposed strategies for editing the decomposition results, a description of how the logging system works, its implementation, and its use. In the first demonstration, the logging system is used to calculate and present different statistics of editing, which can be used to estimate editing strategies. A simple example of partial automation using two different neural network models is another example of using a logging system. The first model uses convolutional layers, while the second uses LSTM layers. Both models were evaluated, and their results were summarized. In the forecasting of editing actions, the model with convolutional layers demonstrated 71% precision with 83% recall, while the model with LSTM layers achieved 100% precision and 75% recall.
Secondary keywords:	logging system;decomposition results;editing statistics;partial editing automation;
Type (COBISS):	Master's thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Računalništvo in informacijske tehnologije
Pages:	1 spletni vir (1 datoteka PDF (X, 69 f.))
ID:	15496046