A dense search challenge phantom fabricated with pixel-based 3D printing for precise detectability assessment

Area of Science:

• Medical Imaging Physics
• Radiological Sciences
• Image Quality Assessment

Background:

• Traditional CT scanner performance metrics (CNR, MTF, NPS) are inadequate for nonlinear reconstruction.
• Accurate measurement of detectability is crucial for optimizing CT imaging protocols and dose efficiency.
• Existing methods fail to precisely predict observer performance in complex reconstruction scenarios.

Purpose of the Study:

• To develop and validate a novel method for precisely measuring CT scanner detectability performance.
• To introduce a dense search challenge as a robust figure of merit for CT image quality.
• To assess the dose efficiency of CT scanners using a human or numerical observer-based task.

Main Methods:

• Designed a dense search challenge phantom with numerous, randomly placed target objects.
• Utilized human or numerical observers to analyze CT reconstructions and identify target locations.
• Calculated a figure of merit, such as Area Under the Curve (AUC), comparing reported to ground truth locations.

Main Results:

• Simulations demonstrated the feasibility of the dense search challenge for phantom design.
• Detectability measurements achieved a precision better than 5%.
• 3D printed phantoms using the PixelPrint technique confirmed the practical applicability of the method.

Conclusions:

• The dense search challenge provides a precise and sensitive measure of CT scanner detectability.
• This method overcomes limitations of traditional metrics in nonlinear reconstruction contexts.
• The technique enables accurate assessment of CT dose efficiency and scanner performance.