667 0

Full metadata record

DC FieldValueLanguage
dc.contributor.advisor이종민-
dc.contributor.author황정민-
dc.date.accessioned2020-02-18T16:35:58Z-
dc.date.available2020-02-18T16:35:58Z-
dc.date.issued2016-02-
dc.identifier.urihttps://repository.hanyang.ac.kr/handle/20.500.11754/127211-
dc.identifier.urihttp://hanyang.dcollection.net/common/orgView/200000428755en_US
dc.description.abstractFunctional magnetic resonance imaging (FMRI), observing spontaneous fluctuation in blood oxygen level dependent (BOLD) signal have been studied for distinguishing diverse disease associated with brain function. Moreover, it is has been a great opportunity to examine functional connectivities of the brain to neuroscientist. Basically, task-positive network (TPN) was investigated from task FMRI and task negative network (TNN) was obtained from resting state FMRI. The results were useful for confirming which subject has disease or not as biomarker. However, recent studies have identified that TPN is also extracted in resting state FMRI. General ways of finding the functional connectivities are Seed Based Correlation Analysis (SBCA) and Independent Component Analysis (ICA). Even though these are helpful for finding intrinsic connectivity networks (ICN), the approaches are fraught with some problems notably among them is user bias. SOM was suggested to deal with this problem but the outcome was not enough to detect whole brain connecitivites. Furthermore, conventional SOM itself has some drawbacks which are noises effect and computation speed on training progress. In this study, we made a proposal that batch SOM was an appropriate method to solve the problems of conventional SOM and extract the whole brain functional connectivity which are the limitations of conventional SOM. As a result, our findings indicates that 8 networks were namely auditory network, basal ganglia network, default mode network (DMN), executive control network (ECN), salience network, sensory motor network, visual network and visuo-spatial network were detected. These result revealed that the 8 networks were competitive compared with previous results and it the batch SOM method which excludes seed selection as well as components number selection can be viewed as being objective hence making the entire process of extracting ICNs automated.-
dc.publisher한양대학교-
dc.title자기 조직화 지도를 이용한 휴지 상태 뇌의 내재 네트워크 예측-
dc.title.alternativePrediction of Intrinsic Networks in Resting State Brain using Self-Organizing Map-
dc.typeTheses-
dc.contributor.googleauthor황정민-
dc.contributor.alternativeauthorJung-Min, Hwang-
dc.sector.campusS-
dc.sector.daehak대학원-
dc.sector.department생체공학과-
dc.description.degreeMaster-
Appears in Collections:
GRADUATE SCHOOL[S](대학원) > BIOMEDICAL ENGINEERING(생체공학과) > Theses (Master)
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE