Dynamic Multiscale Tree Learning Using Ensemble Strong Classifiers for Multi-label Segmentation of Medical Images with Lesions

TL;DR

This paper presents a dynamic multiscale tree architecture that combines strong classifiers like SRF and BN in a bi-directional, hierarchical structure to improve multi-label medical image segmentation, especially for brain lesions.

Contribution

The novel DMT model integrates multiple classifiers in a tree with bi-directional information flow, enhancing segmentation accuracy over existing methods.

Findings

Outperforms state-of-the-art segmentation methods

Effectively integrates multi-level image knowledge

Robustly handles irregular object boundaries

Abstract

We introduce a dynamic multiscale tree (DMT) architecture that learns how to leverage the strengths of different existing classifiers for supervised multi-label image segmentation. Unlike previous works that simply aggregate or cascade classifiers for addressing image segmentation and labeling tasks, we propose to embed strong classifiers into a tree structure that allows bi-directional flow of information between its classifier nodes to gradually improve their performances. Our DMT is a generic classification model that inherently embeds different cascades of classifiers while enhancing learning transfer between them to boost up their classification accuracies. Specifically, each node in our DMT can nest a Structured Random Forest (SRF) classifier or a Bayesian Network (BN) classifier. The proposed SRF-BN DMT architecture has several appealing properties. First, while SRF operates at a patch-level (regular image region), BN operates at the super-pixel level (irregular image region), thereby enabling the DMT to integrate multi-level image knowledge in the learning process. Second, although BN is powerful in modeling dependencies between image elements (superpixels, edges) and their features, the learning of its structure and parameters is challenging. On the other hand, SRF may fail to accurately detect very irregular object boundaries. The proposed DMT robustly overcomes these limitations for both classifiers through the ascending and descending flow of contextual information between each parent node and its children nodes. Third, we train DMT using different scales, where we progressively decrease the patch and superpixel sizes as we go deeper along the tree edges nearing its leaf nodes. Last, DMT demonstrates its outperformance in comparison to several state-of-the-art segmentation methods for multi-labeling of brain images with gliomas.