A publicly accessible Monte Carlo database for validation purposes in emission tomography

Area of Science:

• Medical Imaging
• Computational Physics
• Nuclear Medicine

Background:

• Monte Carlo (MC) methods are crucial for generating high-quality data for emission tomography (ET) research.
• Existing MC codes for ET are computationally intensive and require user adaptation, limiting widespread use.
• A standardized, accessible resource for MC-simulated ET data is needed to facilitate research.

Purpose of the Study:

• To create and publish a publicly accessible database of MC-simulated ET data.
• To provide researchers with ready-to-use datasets for evaluating reconstruction and correction techniques.
• To overcome the computational burden and customization challenges of existing MC simulation tools.

Main Methods:

• Development of the MC-ET database, a web-accessible repository of MC-simulated ET data.
• Inclusion of data generated by multiple MC codes for diverse radioactive sources and phantoms.
• Data derived from various single-photon emission computed tomography (SPECT) and positron emission tomography (PET) scanners.
• Ensuring data validation against measured data and classification by key characteristics.

Main Results:

• The MC-ET database offers direct access to raw MC-simulated data, including unscattered, scattered, and total events.
• Data sets are available for a range of scenarios, from simple phantoms to complex patient studies.
• The database features validated, classified data in a common file format for ease of use.
• The resource is freely accessible online for download and utilization.

Conclusions:

• The MC-ET database significantly lowers the barrier to entry for researchers utilizing MC simulations in emission tomography.
• It promotes standardized evaluation of imaging techniques across different institutions and research groups.
• This resource facilitates advancements in emission tomography reconstruction and correction methodologies.