Classifying Alzheimer's disease, Lewy body dementia, and normal controls using 3D texture analysis in magnetic resonance images

详细信息    查看全文

• 作者：Ketil Oppedal^a ; ^c ; ^{ketil.oppedal@gmail.com} ; Kjersti Engan^a ; ^{kjersti.engan@uis.no} ; Trygve Eftestø ; l^a ; ^{trygve.eftestol@uis.no} ; Mona Beyer^b ; ^{mona.beyer@lyse.net} ; Dag Aarsland^c ; ^d ; ^{daarsland@gmail.com}
• 关键词：Magnetic resonance imaging ; White matter hyperintensities ; Normal appearing white matter ; Alzheimer's disease ; Lewy body dementia ; Texture analysis ; Local binary pattern ; Classification ; Computer aided diagnosis
• 刊名：Biomedical Signal Processing and Control
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：33
• 期：Complete
• 页码：19-29
• 全文大小：4058 K
• 卷排序：33

文摘

3D texture anlaysis was used for classification of dementia and normal controls. Subjects with AD and LBD were discerned from NC with high performance. AD were discerned from LBD with higher performance than 2D texture anlaysis. Upsampling and cost-sensitive classification handled the imbalanced data problem. Features correlated with cognition in subjects with AD.