Ensemble of Weak Spectral Total Variation Learners: a PET-CT Case Study

2025-07-11

AI Summary

The scarcity of paired PET-CT data in medical imaging hinders deep learning–based prediction of skeletal metastatic lesions exhibiting high PET uptake. Method: We propose an ensemble learning framework leveraging weakly correlated spectral total variation (STV)—a novel multiscale nonlinear texture descriptor—integrated into a shallow ensemble of weak learners. STV is applied to decomposed CT volumes to extract fine-grained, low-correlation texture features, enabling implicit association learning between CT morphology and PET uptake under limited supervision. Contribution/Results: Evaluated on 457 patients and 1,524 registered PET-CT slices, our method achieves an AUC of 0.87—significantly outperforming deep learning (0.75) and conventional radiomics (0.79). This demonstrates superior efficacy and generalizability in data-constrained settings, establishing STV as the first spectral texture representation incorporated into an ensemble architecture for PET-CT correlation modeling.

Technology Category

• Machine Learning: Ensemble Methods
• Computer Vision: Representation Learning for Vision
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Solving computer vision problems through machine learning, one often encounters lack of sufficient training data. To mitigate this we propose the use of ensembles of weak learners based on spectral total-variation (STV) features (Gilboa 2014). The features are related to nonlinear eigenfunctions of the total-variation subgradient and can characterize well textures at various scales. It was shown (Burger et-al 2016) that, in the one-dimensional case, orthogonal features are generated, whereas in two-dimensions the features are empirically lowly correlated. Ensemble learning theory advocates the use of lowly correlated weak learners. We thus propose here to design ensembles using learners based on STV features. To show the effectiveness of this paradigm we examine a hard real-world medical imaging problem: the predictive value of computed tomography (CT) data for high uptake in positron emission tomography (PET) for patients suspected of skeletal metastases. The database consists of 457 scans with 1524 unique pairs of registered CT and PET slices. Our approach is compared to deep-learning methods and to Radiomics features, showing STV learners perform best (AUC=0.87), compared to neural nets (AUC=0.75) and Radiomics (AUC=0.79). We observe that fine STV scales in CT images are especially indicative for the presence of high uptake in PET.

Problem

• Addresses lack of training data in computer vision using weak learners
• Proposes STV-based ensembles for improved medical imaging analysis
• Demonstrates superior performance in PET-CT prediction over deep learning

Innovation

• Ensembles of weak STV learners
• Spectral total-variation features for textures
• Outperforms deep-learning and Radiomics methods