Visual Analytics and Interactive Machine Learning for Human Brain Data

Huang Li, Purdue University

Abstract

This study mainly focuses on applying visualization techniques on human brain data for data exploration, quality control, and hypothesis discovery. It mainly consists of two parts: multi-modal data visualization and interactive machine learning. For multi-modal data visualization, a major challenge is how to integrate structural, functional and connectivity data to form a comprehensive visual context. We develop a new integrated visualization solution for brain imaging data by combining scientific and information visualization techniques within the context of the same anatomic structure. In this study, new surface texture techniques are developed to map non-spatial attributes onto both 3D brain surfaces and a planar volume map which is generated by the proposed volume rendering technique, Spherical Volume Rendering. Two types of non-spatial information are represented: (1) time-series data from resting-state functional MRI measuring brain activation; (2) network properties derived from structural connectivity data for different groups of subjects, which may help guide the detection of differentiation features. Through visual exploration, this integrated solution can help identify brain regions with highly correlated functional activations as well as their activation patterns. Visual detection of differentiation features can also potentially discover image based phenotypic biomarkers for brain diseases. For interactive machine learning, nowadays machine learning algorithms usually require a large volume of data to train the algorithm-specific models, with little or no user feedback during the model building process. Such a big data based automatic learning strategy is sometimes unrealistic for applications where data collection or processing is very expensive or difficult. Furthermore, expert knowledge can be very valuable in the model building process in some fields such as biomedical sciences. In this study, we propose a new visual analytics approach to interactive machine learning. In this approach, multi-dimensional data visualization techniques are employed to facilitate user interactions with the machine learning process. This allows dynamic user feedback in different forms, such as data selection, data labeling, and data correction, to enhance the efficiency of model building. In particular, this approach can significantly reduce the amount of data required for training an accurate model, and therefore can be highly impactful for applications where large amount of data is hard to obtain. The proposed approach is tested on two application problems: the handwriting recognition (classification) problem and the human cognitive score prediction (regression) problem. Both experiments show that visualization supported interactive machine learning can achieve the same accuracy as an automatic process can with much smaller training data sets.

Degree

Ph.D.

Advisors

Fang, Purdue University.

Subject Area

Medical imaging|Aging|Artificial intelligence|Neurosciences

COinS