Cited 0 times in
Deep-learned time-signal intensity pattern analysis using an autoencoder captures magnetic resonance perfusion heterogeneity for brain tumor differentiation
DC Field | Value | Language |
---|---|---|
dc.contributor.author | 신일아 | - |
dc.contributor.author | 안성수 | - |
dc.date.accessioned | 2021-05-21T17:02:29Z | - |
dc.date.available | 2021-05-21T17:02:29Z | - |
dc.date.issued | 2020-12 | - |
dc.identifier.uri | https://ir.ymlib.yonsei.ac.kr/handle/22282913/182666 | - |
dc.description.abstract | Current image processing methods for dynamic susceptibility contrast (DSC) magnetic resonance imaging (MRI) do not capture complex dynamic information of time-signal intensity curves. We investigated whether an autoencoder-based pattern analysis of DSC MRI captured representative temporal features that improves tissue characterization and tumor diagnosis in a multicenter setting. The autoencoder was applied to the time-signal intensity curves to obtain representative temporal patterns, which were subsequently learned by a convolutional neural network. This network was trained with 216 preoperative DSC MRI acquisitions and validated using external data (n = 43) collected with different DSC acquisition protocols. The autoencoder applied to time-signal intensity curves and clustering obtained nine representative clusters of temporal patterns, which accurately identified tumor and non-tumoral tissues. The dominant clusters of temporal patterns distinguished primary central nervous system lymphoma (PCNSL) from glioblastoma (AUC 0.89) and metastasis from glioblastoma (AUC 0.95). The autoencoder captured DSC time-signal intensity patterns that improved identification of tumoral tissues and differentiation of tumor type and was generalizable across centers. | - |
dc.description.statementOfResponsibility | open | - |
dc.language | English | - |
dc.publisher | Nature Publishing Group | - |
dc.relation.isPartOf | SCIENTIFIC REPORTS | - |
dc.rights | CC BY-NC-ND 2.0 KR | - |
dc.subject.MESH | Adult | - |
dc.subject.MESH | Aged | - |
dc.subject.MESH | Brain Neoplasms / diagnosis | - |
dc.subject.MESH | Central Nervous System Neoplasms / diagnosis | - |
dc.subject.MESH | Contrast Media | - |
dc.subject.MESH | Databases, Factual | - |
dc.subject.MESH | Diagnosis, Differential | - |
dc.subject.MESH | Female | - |
dc.subject.MESH | Glioblastoma / diagnosis | - |
dc.subject.MESH | Humans | - |
dc.subject.MESH | Image Processing, Computer-Assisted / methods* | - |
dc.subject.MESH | Lymphoma / diagnosis | - |
dc.subject.MESH | Lymphoma, Non-Hodgkin / diagnosis | - |
dc.subject.MESH | Magnetic Resonance Imaging / methods* | - |
dc.subject.MESH | Male | - |
dc.subject.MESH | Middle Aged | - |
dc.subject.MESH | Neoplasms / diagnostic imaging* | - |
dc.subject.MESH | Perfusion | - |
dc.subject.MESH | Retrospective Studies | - |
dc.title | Deep-learned time-signal intensity pattern analysis using an autoencoder captures magnetic resonance perfusion heterogeneity for brain tumor differentiation | - |
dc.type | Article | - |
dc.contributor.college | College of Medicine (의과대학) | - |
dc.contributor.department | Dept. of Radiology (영상의학교실) | - |
dc.contributor.googleauthor | Ji Eun Park | - |
dc.contributor.googleauthor | Ho Sung Kim | - |
dc.contributor.googleauthor | Junkyu Lee | - |
dc.contributor.googleauthor | E-Nae Cheong | - |
dc.contributor.googleauthor | Ilah Shin | - |
dc.contributor.googleauthor | Sung Soo Ahn | - |
dc.contributor.googleauthor | Woo Hyun Shim | - |
dc.identifier.doi | 10.1038/s41598-020-78485-x | - |
dc.contributor.localId | A05848 | - |
dc.contributor.localId | A02234 | - |
dc.relation.journalcode | J02646 | - |
dc.identifier.eissn | 2045-2322 | - |
dc.identifier.pmid | 33293590 | - |
dc.contributor.alternativeName | Shin, Ilah | - |
dc.contributor.affiliatedAuthor | 신일아 | - |
dc.contributor.affiliatedAuthor | 안성수 | - |
dc.citation.volume | 10 | - |
dc.citation.number | 1 | - |
dc.citation.startPage | 21485 | - |
dc.identifier.bibliographicCitation | SCIENTIFIC REPORTS, Vol.10(1) : 21485, 2020-12 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.