| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font
First page > Show document

Show document Help

Title:Koherenčna resonanca v dvoplastnih mrežah ekscitatornih in inhibitornih nevronov
Authors:ID Ristič, David (Author)
ID Gosak, Marko (Mentor) More about this mentor... New window
Files:.pdf MAG_Ristic_David_2021.pdf (4,21 MB)
MD5: 21262AB5C976406B2FD0CE25267EF770
PID: 20.500.12556/dkum/95dbe619-2280-4bd8-9e2e-dd01c724847f
 
Language:Slovenian
Work type:Master's thesis/paper
Typology:2.09 - Master's Thesis
Organization:FNM - Faculty of Natural Sciences and Mathematics
Abstract:Sinhrono delovanje nevronov je ključnega pomena za mnoge možganske procese in se preučuje tudi v povezavi z nevrološkimi boleznimi. Eden izmed pojavov, ki se v kontekstu sinhronizacije pogosto proučuje, je koherenčna resonanca. Za ta nelinearen pojav je značilno, da dodan šum privede do izboljšane urejenosti impulzov v nevronih. V magistrski nalogi preučujemo pojav koherenčne resonance v populaciji ekscitatornih in inhibitornih nevronov. Za opis dinamike posameznega nevrona uporabimo modificirano Fitz-Hugh Nagumo enačbo z dodatnimi sklopitvenimi členi. Formalno naš model kolektivne nevronske dinamike temelji na konceptu dvoplastne mreže, kjer eno plast predstavljajo inhibitorni nevroni, drugo ekscitatorni, povezave med plastema pa simulirajo heterologne interakcije med njimi. Pare nevronov med seboj povežemo z dvema vrstama mrež. Nevrone istega tipa povežemo znotraj plasti z neusmerjenimi povezavami po načelu modela naključne geometrične mreže, medtem ko nevrone različnih tipov povezujemo z usmerjenimi povezavami, in sicer po načelu prostorsko vpetega modela mreže, ki temelji na pripisanih pomembnostih vozlov. Urejenost simulirane nevronske aktivnosti merimo z avtokorelacijsko funkcijo in korelacijskim časom. S tem modelom raziščemo območja parametrov, pri katerih se v sistemu manifestira koherenčna resonanca in raziščemo, kako različne kombinacije parametrov vplivajo na njeno stabilnost in naravo. Osredinimo se na parametre mreže medplastnih povezav med inhibitornimi in ekscitatornimi nevroni, na deleža inhibitornih nevronov in ekscitatornih aksonov v medplastni mreži ter na vlogo jakosti povezav med ekscitatornimi in inhibitornimi nevroni. Izkaže se, da pri večini parametrov obstajajo mejne vrednosti, ki omejujejo stabilno območje s koherenčno resonanco. Pokažemo, da lahko koherenco sistema povečamo z večjim dušenjem ekscitatorne plasti. To lahko dosežemo z večjim deležem inhibitornih nevronov, večjim deležem inhibitornih aksonov med plastema ali z večjo jakostjo sklopitve inhibitornih aksonov. Prav tako ugotovimo, da za največjo regularnost kolektivnega odziva nevronske mreže obstaja optimalna konfiguracija mreže, ki preferira povezave med bližnjimi nevroni s pravo mero daljnosežnih povezav. Tipično so optimalni parametri mreže blizu mejnih vrednosti parametrov, pri katerih je sistem še stabilen.
Keywords:ekscitatorni nevroni, inhibitorni nevroni, dvoplastna mreža, topologija, koherenčna resonanca, karakteristični korelacijski čas
Place of publishing:Maribor
Publisher:[D. Ristič]
Year of publishing:2021
PID:20.500.12556/DKUM-79136 New window
UDC:534:531.3(043.2)
COBISS.SI-ID:69171459 New window
Publication date in DKUM:22.07.2021
Views:1089
Downloads:114
Metadata:XML DC-XML DC-RDF
Categories:FNM
:
Copy citation
  
Average score:(0 votes)
Your score:Voting is allowed only for logged in users.
Share:Bookmark and 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:11.05.2021

Secondary language

Language:English
Title:Coherence resonance in two-layered networks of excitatory and inhibitory neurons
Abstract:Synchronous neuronal activity plays a vital role in many processes within the brain and is being studied in relation with neurological disorders as well. One of the many phenomena that is often studied in the context of neuronal synchronization is coherence resonance, where additional noise leads to improved regularity of spiking activity in neurons. In this master thesis, we investigate the phenomenon of coherence resonance in a population of excitatory and inhibitory neurons. We describe the dynamics of a single neuron using modified FitzHugh-Nagumo equations with additional coupling terms. Formally, our model of collective neuron dynamics is based on the concept of two-layered network, where one layer contains inhibitory neurons and the other excitatory neurons. Connections between layers represent heterologous interactions. Neurons of the same type are connected within a layer with undirected connections according to the principle of a random geometric network model, while neurons of different types are connected with directed connections according to the principle of spatially embedded network model based on the assigned vertex fitness model. We measure the regularity of simulated neural activity with the autocorrelation function and the correlation time. Using this model, we investigate the parameter range in which coherence resonance manifests itself in our system, and how different combinations of parameters affect its nature and stability. We focus on the network parameters of interlayer connections between inhibitory and excitatory neurons, the proportion of inhibitory neurons, the proportion of excitatory interlayer axons, and the importance of connection strength between layers. It turns out that for most parameters there exists a boundary value, which defines a stable parametric range with coherence resonance. We show that the coherence of the system can be increased by a stronger damping of the excitatory layer. This can be achieved with a higher proportion of inhibitory neurons, a higher proportion of inhibitory interlayer axons, or a stronger coupling of inhibitory axons. We also find that the maximum regularity of the collective response of a neural network is achieved at an optimal network configuration that prefers connections between nearby neurons, with the right amount of additional long-range connections. Typically, these optimal parameter values are close to the boundary values of a stable range.
Keywords:excitatory neurons, inhibitory neurons, two-layered network, topology, coherence resonance, characteristic correlation time


Comments

Leave comment

You must log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back
Logos of partners University of Maribor University of Ljubljana University of Primorska University of Nova Gorica