论文信息 - Unsupervised 3-D Feature Learning for Mild Traumatic Brain Injury

Unsupervised 3-D Feature Learning for Mild Traumatic Brain Injury

We present an unsupervised three-dimensional feature clustering algorithm to gather the mTOP2016 challenge data into 3 groups. We use the brain MR-T1, diffusion tensor fractional anisotropy, and diffusion tensor mean diffusivity images provided by the mTOP2016 competition. A distance-based size constraint method for data clustering is used. The proposed approach achieves 0.267 adjusted rand index and 0.3556 homogeneity score within the 15 labeled subjects, corresponding to 10 correctly classified data items. Based on visual exploration of the data, we believe that a localized analysis of the lesion regions, using the computed tractography data, is a promising direction to pursue.

B. S. Manjunath | Po-Yu Kao | Angela Zhang | Jefferson W. Chen | Eduardo Rojas

[1] M. Cugmas,et al. On comparing partitions , 2015 .

[2] Andrew Y. Ng,et al. Learning Feature Representations with K-Means , 2012, Neural Networks: Tricks of the Trade.

[3] David J. Field,et al. Emergence of simple-cell receptive field properties by learning a sparse code for natural images , 1996, Nature.

[4] Honglak Lee,et al. An Analysis of Single-Layer Networks in Unsupervised Feature Learning , 2011, AISTATS.

[5] Yee Whye Teh,et al. A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[6] Alex Krizhevsky,et al. Learning Multiple Layers of Features from Tiny Images , 2009 .

[7] Julia Hirschberg,et al. V-Measure: A Conditional Entropy-Based External Cluster Evaluation Measure , 2007, EMNLP.

[8] Shunzhi Zhu,et al. Data clustering with size constraints , 2010, Knowl. Based Syst..