MedSR-Impact: Transformer-Based Super-Resolution for Lung CT Segmentation, Radiomics, Classification, and Prognosis

2025-07-21

AI Summary

High-resolution pulmonary CT is critical for precise diagnosis and treatment but is constrained by radiation exposure and hardware costs. To address this, we propose TVSRN-V2—a novel 3D super-resolution framework that integrates Through-Plane Attention with Swin Transformer V2 and incorporates a pseudo-low-resolution data augmentation strategy, significantly enhancing generalizability and robustness for low-dose CT reconstruction. Evaluated on a multi-center clinical cohort, our method improves lobar segmentation Dice score by 4%, substantially enhances radiomic feature reproducibility, and boosts prognostic prediction performance—increasing the C-index and AUC by 0.06 each. This work establishes a generalizable, transformer-based solution for low-dose, high-fidelity pulmonary CT reconstruction.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Feature Construction/Reformulation
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

High-resolution volumetric computed tomography (CT) is essential for accurate diagnosis and treatment planning in thoracic diseases; however, it is limited by radiation dose and hardware costs. We present the Transformer Volumetric Super-Resolution Network ( extbf{TVSRN-V2}), a transformer-based super-resolution (SR) framework designed for practical deployment in clinical lung CT analysis. Built from scalable components, including Through-Plane Attention Blocks (TAB) and Swin Transformer V2 -- our model effectively reconstructs fine anatomical details in low-dose CT volumes and integrates seamlessly with downstream analysis pipelines. We evaluate its effectiveness on three critical lung cancer tasks -- lobe segmentation, radiomics, and prognosis -- across multiple clinical cohorts. To enhance robustness across variable acquisition protocols, we introduce pseudo-low-resolution augmentation, simulating scanner diversity without requiring private data. TVSRN-V2 demonstrates a significant improvement in segmentation accuracy (+4% Dice), higher radiomic feature reproducibility, and enhanced predictive performance (+0.06 C-index and AUC). These results indicate that SR-driven recovery of structural detail significantly enhances clinical decision support, positioning TVSRN-V2 as a well-engineered, clinically viable system for dose-efficient imaging and quantitative analysis in real-world CT workflows.

Problem

• Enhancing lung CT resolution for better diagnosis and treatment
• Improving segmentation, radiomics, and prognosis accuracy in low-dose CT
• Ensuring robustness across diverse CT acquisition protocols

Innovation

• Transformer-based super-resolution for lung CT
• Through-Plane Attention Blocks enhance detail
• Pseudo-low-resolution augmentation improves robustness