Adaptive optimization of slit width for a slit-slat collimator

Although slit-slat is one of the few collimators that can be actively adapted during a scan, this adaptivity remains largely unutilized. Adaptively optimizing the slit width for different tasks has the potential to improve the performance of the instrument without increasing scan time. By automated adjustments of the slit width during the collimator´s orbit, lesion detectability can be enhanced. Here we simulated adjusting the slit width as a function of viewing angle for different phantom sizes and lesion locations. Scans over 20 viewing angles were simulated using 720 sets of dynamic and 9 sets of constant slit widths for collimator configurations and the performance was evaluated with different activity concentration at various locations in a uniform disk phantom. System performance was estimated based on calculating the average contrast at fixed relative noise ratio in reconstructions using a maximum likelihood expectation maximization (MLEM) algorithm over 500 noisy realizations for each configuration. To further test this idea, a prototype adaptive slit width collimator was built and a phantom was scanned with 15 sets of slit width configurations. Contrast at matched noise ratio was computed in reconstructed images for evaluation. Preliminary test results show that optimized sets of dynamic slit width again improved contrast. Both simulation and prototype phantom scans showed that adaptive slit width can improve image quality over constant slit width for off-center lesion.