(Tuesday, 7/31/2018) 1:15 PM - 1:45 PM
Room: Exhibit Hall | Forum 1
Purpose: Quantitative imaging performance analysis has recently been the focus in medical imaging fields. It would not only provide objective information but also it could aid a patient diagnosis by giving optimized system parameters for various imaging task. However, the previous studies on task-based metric in breast tomosynthesis usually take into account a cascaded system modeling for generalized noise equivalent quanta. In this study, we focused on the experimental study for calculating task-based detectability index (d') in the prototype digital breast tomosynthesis (DBT) system.
Methods: We used the prototype DBT system developed by Korea Electrotechnology Research Institute for image acquisition. For d’, the non-prewhitening matched filter observer were calculated. We analyzed task function, local resolution and local noise to calculate task-based d’ with angular range of ±10.5°, ±14.5°, ±21° and ±24.5°.
Results: The task functions showed the same patterns, and mostly same levels with each other because we implemented the same circular object function (same shape in task function). On the other hand, the local MTF and NPS for each angular range presented different shapes and intensities with each other. These could demonstrate that the local MTF and NPS were highly dependent on the acquisition angular distribution even if we measured the identical spatial location. Moreover, resulted local NPS showed the particular decrease of local noise properties with reduction of total angular range. The resulted d’ showed a significant decrease by the angular ranges from ±10.5° to ±24.5° with different size of masses with ∅6.3 mm, ∅4.7 mm, and ∅3.9 mm.
Conclusion: In this study, we have focused on the experimental study on calculating task-based detectability in the prototype DBT system. Our study may present more convincing performance assessment for clinical applications. The future works will be directed to generalize the results using a variety of imaging geometry and tasks.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
