The National Cancer Informatics Program (NCIP) Annotation and Image Markup (AIM) Foundation model

Area of Science:

• Biomedical Informatics
• Medical Imaging
• Data Standards

Background:

• In vivo imaging data is often unstructured text, separate from other critical information.
• Existing annotation models require evolution to meet the demands of the imaging community.
• The National Cancer Informatics Program (NCIP) sought to improve data integration and usability.

Purpose of the Study:

• Introduce the NCIP AIM Foundation information model as an advancement over previous versions.
• Describe the model's comprehensive structure for capturing diverse imaging-related data.
• Explain how to utilize and extend the AIM model for various scientific disciplines.

Main Methods:

• Developed the AIM Foundation model based on feedback from the NCI imaging community.
• Defined a model encompassing physical entities, observations, markups, and ancillary data.
• Established AIM instances as DICOM SR objects or XML documents for processing.

Main Results:

• The AIM Foundation model captures a wide array of imaging information, including annotations, markups, and metadata.
• AIM instances can be stored in standard formats like DICOM SR and XML.
• The model is extensible for use across different imaging disciplines.

Conclusions:

• The AIM Foundation model provides a robust framework for organizing and analyzing in vivo imaging data.
• Standardized data improves interoperability and facilitates advanced computational analysis.
• The model supports diverse imaging applications and future extensions.