SubCMap: Subject and Condition Specific Effect Maps

By Ender Konukoglu, Ben Glocker. 2017

Current methods for statistical analysis of neuroimaging data identify condition related structural alterations in the human brain by detecting group differences. They construct detailed maps showing population-wide changes due to a condition of interest. Although extremely useful, methods do not provide information on the subject-specific structural alterations and they have limited diagnostic value because group assignments for each subject are required for the analysis. In this article, we propose SubCMap, a novel method to detect subject and condition specific structural alterations. SubCMap is designed to work without the group assignment information in order to provide diagnostic value. Unlike outlier detection methods, SubCMap detections are condition-specific and can be used to study the effects of various conditions or for diagnosing diseases. The method combines techniques from classification, generalization error estimation and image restoration to the identify the condition-related alterations. Experimental evaluation is performed on synthetically generated data as well as data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. Results on synthetic data demonstrate the advantages of SubCMap compared to population-wide techniques and higher detection accuracy compared to outlier detection. Analysis with the ADNI dataset show that SubCMap detections on cortical thickness data well correlate with non-imaging markers of Alzheimer's Disease (AD), the Mini Mental State Examination Score and Cerebrospinal Fluid amyloid-$\beta$ levels, suggesting the proposed method well captures the inter-subject variation of AD effects.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Visual Feature Attribution Using Wasserstein GANs (2017, methodology): “Pinpointing all evidence of a class is important for a variety of tasks such as weakly supervised localisation or segmentation of structures [41, 43, 65], and better understanding disease effects, and physiological or pathological processes in medical images [67, 18, 12, 19, 13, 27, 54, 30, 31, 63].”