Medical Image Registration

Chimaera's Medical Image Registration is a high-performance OEM software component designed to reliably compensate for patient movements and misalignments between different image acquisitions and imaging modalities. It supports a wide range of clinical scenarios including follow-up imaging, image stitching, and multi-modal acquisitions involving CT, MR, PET, and SPECT. Delivered as a dynamic library with a C-API, it is purpose-built for life sciences software developers and medical device manufacturers looking to extend existing applications with accurate spatial mapping capabilities.

The solution is offered in two core registration modes — rigid and nonrigid — each addressing distinct clinical and technical needs. Both modes support 2D, 3D, and time series data, and are compatible with multi-modal imaging data. The component is designed for straightforward integration with minimal parameterization requirements, making it suitable for fast adoption within certified medical software pipelines.

Rigid Registration

• Compensates for rotations and translations only, leaving image content unmodified
• No manual parameterization required — fully automatic
• Supports 2D, 3D, and time series data
• Compatible with multi-modal data including CT, MR, PET, and SPECT
• Typical applications include retrospective image fusion (e.g. PET/CT, PET/MR), side-by-side viewing of follow-up studies, and image stitching
• Enables linking of different scanner coordinate systems and accounts for rigid body movements

Nonrigid Registration

• Computes complex deformations between images with far greater degrees of freedom than rigid registration
• Deformation field can compensate for respiratory motion, cardiac motion, patient positioning differences, and soft tissue misalignments
• Easy and flexible parameterization adaptable to different use cases
• Supports 2D, 3D, and time series data
• Compatible with multi-modal data (CT, MR, PET, SPECT, etc.)
• Cross-patient registration is supported, enabling atlas creation across different subjects
• Typical applications include perfusion imaging, difference imaging, atlas registration, and transfer of findings to follow-up studies

Clinical and Technical Applications

• Retrospective image fusion across modalities
• Follow-up study comparison and side-by-side viewing
• Difference and perfusion imaging
• 4D motion correction
• Model and atlas-to-image registration, including brain atlas segmentation transfer
• Post-market surveillance
• Industrial use cases such as position correction of workpieces, camera stitching, and calibration applications

Key Benefits

• Fully automatic rigid registration with no manual parameterization needed
• Minimal parameterization required for nonrigid registration
• CPU-based processing with low hardware requirements
• Simple integration via a single dynamic library with C-API and no additional dependencies
• Example interface code provided, including wrappers for ITK image types
• Output available as transformation or deformation vector field, in addition to resampled output images
• Quality documentation supplied to support fast integration into certified medical software products

Chimaera's Medical Image Registration is delivered as an OEM component with full quality documentation to facilitate regulatory compliance and product certification. The library requires no additional dependencies and includes example code to accelerate integration into customer applications. It is available as part of the broader Chimaera SDK, which supports projects from prototype through to finished medical software product.