Self-normalized Classification of Parkinson's Disease DaTscan Images

TL;DR

This paper introduces a self-normalized classification method for DaTscan images that removes the need for normalization regions, based on a geometric analysis of image equivalence, and demonstrates its effectiveness on Parkinson's disease data.

Contribution

The paper proposes a novel self-normalized classification approach that eliminates the dependence on normalization regions, supported by a mathematical analysis of image geometry.

Findings

Effective classification of Parkinson's DaTscan images

Elimination of normalization region dependence

Insights into PD progression from baseline to year 4

Abstract

Classifying SPECT images requires a preprocessing step which normalizes the images using a normalization region. The choice of the normalization region is not standard, and using different normalization regions introduces normalization region-dependent variability. This paper mathematically analyzes the effect of the normalization region to show that normalized-classification is exactly equivalent to a subspace separation of the half rays of the images under multiplicative equivalence. Using this geometry, a new self-normalized classification strategy is proposed. This strategy eliminates the normalizing region altogether. The theory is used to classify DaTscan images of 365 Parkinson's disease (PD) subjects and 208 healthy control (HC) subjects from the Parkinson's Progression Marker Initiative (PPMI). The theory is also used to understand PD progression from baseline to year 4.