Local Region Sparse Learning for Image-on-Scalar Regression

TL;DR

This paper introduces a novel image-on-scalar regression method that combines SCAD regularization and total variation to effectively identify contiguous regions of interest in images, scalable to large datasets.

Contribution

It proposes a new region-selection penalty integrating SCAD and TV regularization, along with an efficient ADMM-based algorithm and a divide-and-conquer approach for large images.

Findings

Outperforms existing methods in ROI detection accuracy

Effective in handling large-scale image data

Demonstrates superior spatial contiguity enforcement

Abstract

Identification of regions of interest (ROI) associated with certain disease has a great impact on public health. Imposing sparsity of pixel values and extracting active regions simultaneously greatly complicate the image analysis. We address these challenges by introducing a novel region-selection penalty in the framework of image-on-scalar regression. Our penalty combines the Smoothly Clipped Absolute Deviation (SCAD) regularization, enforcing sparsity, and the SCAD of total variation (TV) regularization, enforcing spatial contiguity, into one group, which segments contiguous spatial regions against zero-valued background. Efficient algorithm is based on the alternative direction method of multipliers (ADMM) which decomposes the non-convex problem into two iterative optimization problems with explicit solutions. Another virtue of the proposed method is that a divide and conquer learning algorithm is developed, thereby allowing scaling to large images. Several examples are presented and the experimental results are compared with other state-of-the-art approaches.