Joint CT-MRI reconstruction with cross-modal structural coupling

Technology
In development
University
Scouted & summarized by Halo AI

This joint image reconstruction framework optimizes CT and MRI datasets simultaneously. By using bidirectional estimation and cross-modal structural coupling, the system leverages shared anatomical features to improve image quality, contrast, and artifact suppression. It supports lower-dose CT and faster MRI acquisition, making it a powerful tool for enhancing multimodal medical imaging workflows.

Overview

The joint image reconstruction framework is an advanced technological solution designed to enhance the quality of medical imaging by simultaneously optimizing computed tomography (CT) and magnetic resonance imaging (MRI) datasets. By utilizing a bidirectional estimation process, the system incorporates structural information from one modality into the reconstruction of the other. This cross-modal approach creates mutually informed image updates that significantly improve anatomical fidelity, contrast, and edge definition, while effectively suppressing reconstruction artifacts. The framework is particularly valuable for hybrid or sequential imaging systems, enabling superior diagnostic performance from undersampled data.

Technical specifications
  • Bidirectional estimation: Uses structural information shared between CT and MRI to inform image updates.
  • Cross-modal coupling: Integrates structural features from different imaging modalities to enhance anatomical consistency.
  • Compressive sensing: Employs sparsity constraints to allow for high-quality image reconstruction from undersampled or sparse raw data.
  • Performance: Supports reduced radiation doses for CT scans and shorter acquisition times for MRI, enhancing overall scanning efficiency.
  • Versatility: Compatible with both hybrid and sequential multimodal imaging workflows and adaptable to various multi-contrast MRI applications.
Technology readiness level

The technology is protected by US Patent No. 11,020,077 B2. It represents a developed research framework that has been demonstrated through iterative algorithms and testing. While it offers significant potential for clinical imaging, it remains in a stage requiring ongoing validation across different anatomies and clinical protocols before widespread routine deployment, placing it at a TRL of 4-6 (lab validation/prototype demonstration).

Generated Oct 28, 2026. Please verify key information.

About Rensselaer Polytechnic Institute

Rensselaer Polytechnic Institute is a mid-sized private research university in Troy, New York, with a STEM‑intensive profile and applied research orientation. Industry collaborates through co‑located, industry‑accessible labs and a high‑performance computing environment for modeling, simulation, and data‑intensive research. A nearby research and technology park, incubation resources, and a well‑established co‑op pipeline link companies to faculty expertise and student talent. Research is supported by competitive federal funding, including NSF, NIH, DOE, and DoD. A dedicated technology transfer office streamlines IP, licensing, and startup formation with business‑friendly agreements.

Halo home
Partner smarter. Move faster.
Get new partnering requests
delivered to your inbox.