Multi-modal multiple kernel learning for accurate identification of Tourette syndrome children

详细信息    查看全文

• 作者：Hongwei Wen^a ; ^b ; ^c ; ¹Author Vitae ; Yue Liu^e ; ¹Author Vitae ; Islem Rekik^gAuthor Vitae ; Shengpei Wang^a ; ^b ; ^cAuthor Vitae ; Zhiqiang Chen^a ; ^b ; ^cAuthor Vitae ; Jishui Zhang^fAuthor Vitae ; Yue Zhang^eAuthor Vitae ; Yun Peng^e ; ^{ppengyun@yahoo.com}Author Vitae ; Huiguang He^a ; ^b ; ^c ; ^d ; ^{huiguang.he@ia.ac.cn}Author Vitae
• 关键词：TS ; Tourette syndrome ; DTI ; diffusion tensor imaging ; WM ; white matter ; SVM ; support vector machine ; MKL ; multiple kernel learning ; TBSS ; Tract-Based Spatial Statistics
• 刊名：Pattern Recognition
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：63
• 期：Complete
• 页码：601-611
• 全文大小：1101 K
• 卷排序：63

文摘

We combine VBM and TBSS analysis to investigate GM/WM changes in TS children. We apply most-representative-subject TBSS procedure suitable for young children. We integrate multi-modal image features using multiple kernel learning. We achieved an excellent accuracy of 94.24%. We identify the most discriminative ROIs and features for classification.