How a deck is put together

Before diving into the feature chapters, it helps to have a mental model of what a SPECT simulation deck contains and how the OpenTOPAS-SPECT pieces fit into a standard OpenTOPAS parameter file. Everything here is ordinary OpenTOPAS syntax; OpenTOPAS-SPECT only adds a handful of new components (geometry) and scorers (measurement).

The five ingredients

Every SPECT deck combines the same five parts. You mix and match them across the feature chapters.

Ingredient	OpenTOPAS prefix	What OpenTOPAS-SPECT provides
World and physics	Ge/World, Ph/	nothing new, use standard OpenTOPAS
Collimator	Ge/<name>	TsParallelHoleCollimator, TsPinholeCollimator, TsSlitSlatCollimator
Detector	Ge/<name>	a scintillator/semiconductor crystal volume (any OpenTOPAS solid)
Source	So/<name>	native OpenTOPAS sources, or one per phantom region
Scorer	Sc/<name>	TsEDepSpectrum, TsForcedDetectionProjection

A ready-made camera preset (see systems) bundles the collimator and detector so a single IncludeFile gives you a validated Symbia or GE geometry. You then add the world, physics, source, and scorer for your particular study.

The detector chain

A photon that reaches the camera passes through two stages, and you choose how much of each to simulate:

1. Collimation. A parallel-hole collimator only passes photons travelling nearly parallel to its holes; everything else is absorbed in the septa. This is what gives SPECT its direction information, and why it throws away ~99.98% of emitted photons. OpenTOPAS-SPECT models the holes and septa explicitly, so the transport captures septal penetration and scatter directly.

2. Detection. Photons that survive the collimator deposit energy in the crystal. The TsEDepSpectrum scorer records the pulse-height spectrum, from which you apply an energy window to select the photopeak.

Because full collimator transport is so wasteful, OpenTOPAS-SPECT also offers a fast path, the TsForcedDetectionProjection scorer, that replaces stages 1 and 2 with an analytic projection through the collimator. Use full transport when you care about scatter and penetration physics; use forced detection when you need many views quickly. See Variance reduction.

Coordinate conventions

The phantoms and acquisitions in this package follow one convention, and it is worth stating up front because it determines how detectors are placed and rotated:

• The patient long axis is z (the torso is long in z).
• A gamma camera images a transaxial plane; a single anterior detector sits beside the patient (offset in y) with its face looking back along the axis toward the patient.
• A rotating acquisition orbits the detector around z, or, equivalently and often more reliably, holds the detector fixed and rotates the phantom about z. The CT-driven motion chapter and examples/phantom/spect_acquisition.txt use both approaches and explain when each applies.

OpenTOPAS rotations are passive

A component's RotX/RotY/RotZ rotate its frame, not the object, and are applied in the order X→Y→Z about the parent axes. This matters when you aim a detector face at the patient: the surest way to confirm a face points where you intend is to compute the world direction of its local normal, instead of eyeballing a render. The tools in this package do that check for you.