Image Acquisition, Processing, and Quality Control
Core concepts
Acquisition modes include static, dynamic, gated, whole-body, and SPECT. Matrix size and zoom determine pixel size; pixel size = field of view / matrix size (adjust for zoom). Word mode stores each event; frame mode bins counts into frames. Energy window settings and scatter correction (e.g., dual-energy or triple-energy window) affect contrast. Processing includes filtering (Butterworth, Hann; cutoff and order), uniformity correction, motion correction, and region-of-interest (ROI) analysis for time-activity curves. Quality control includes daily flood/uniformity, energy peaking, extrinsic vs intrinsic floods, center of rotation (COR) for SPECT, and bar phantom resolution checks. Photopeak drift and PMT gain changes are common QC failures.
Traps
- Cutoff frequency and order: higher cutoff preserves resolution but passes more noise; higher order sharpens the roll-off.
- Confusing intrinsic (collimator removed) and extrinsic (collimator in place) uniformity.
- Forgetting that zoom changes pixel size - recalculate when comparing matrices.
- COR misalignment causes ring artifacts or blurring in SPECT.
Rule of thumb
For a 40 cm field of view with a 128×128 matrix, pixel size ≈ 3.1 mm; doubling zoom halves pixel size.
Worked example: FOV 400 mm, matrix 128 → 400/128 ≈ 3.1 mm/pixel. With zoom 2, pixel ≈ 1.6 mm.