Stitched Exams, Real Detectors: What Long-Leg and Whole-Spine Imaging Actually Demands from a Flat Panel Detector

stitching — combining two or more exposures into one long image of a spine or a full leg — is mostly a software job, but the detector quietly decides whether the seam is invisible or the exam gets repeated. Consistent gain between exposures, stable calibration, reliable timing and enough panel coverage per shot matter more than any stitching algorithm. This article walks through what the detector side contributes to long-length imaging, and what to check before you promise a scoliosis or leg-alignment exam to your orthopedic referrals.

Long-length imaging comes up the moment a clinic starts seeing orthopedic work: full-spine films for scoliosis follow-ups, standing leg-alignment studies before knee surgery, occasionally a tall patient whose thoracic and lumbar spine simply do not fit one 17×17 field. The inquiries from orthopedic clinics usually start exactly here — “can your detector do the whole spine in one shot?” The honest answer has two parts, and the second part is the one buyers forget.

Two ways to cover a tall anatomy — and where the detector fits in each

• One large panel, one exposure. An oversized detector such as a 43×43 format can capture a whole spine or a long leg in a single shot. No stitching software is involved; the “seam” problem simply does not exist. The trade-offs — panel availability, physical handling, generator output — are covered in the dedicated article on the large-format flat panel detector;
• Multiple exposures, stitched in software. A standard 14×17 or 17×17 panel shoots two or three overlapping fields while the patient holds still, and the workstation joins them into one long image. This is the route most rooms take, because it uses the detector they already own.

Everything below is about the second route, because that is where detector behavior quietly decides image quality.

What the detector contributes to a clean stitch

Stitching software can align edges, match brightness on paper and blend transition zones. What it cannot do is invent information that the panel captured inconsistently. Four detector-side properties do the real work:

• Gain consistency between exposures. If the second exposure’s response drifts relative to the first — from panel warming, residual image or a calibration state that changed between shots — the two halves of the stitched image will not match in texture, and the seam shows as a visible step in soft tissue;
• Current calibration. Stitched exams amplify any gain-map defect: a field-uniformity flaw that hides inside one ordinary exposure becomes a visible discontinuity the moment two exposures are joined along it. This is the same calibration discipline described in the article on calibration procedures for image accuracy — stitched workflows just raise the stakes;
• Exposure timing the room can control. Successive exposures need consistent triggering — same geometry, same technique factors, and a detector response that does not vary with the pace of exposures. If the panel’s ready-state behavior changes on the second shot, positioning errors creep in between fields;
• Coverage per field. The larger the useful area per shot, the fewer the seams. A 17×17 panel covers a full lumbar segment in one field where a 14×17 may need two; fewer exposures also means fewer chances for the patient to move — which, as below, is the biggest failure mode of all.

The failure mode software cannot fix: the patient moved

Between two exposures there is a gap — a few seconds, sometimes longer if the technique changes. Patients sway. Weight shifts. Children, in particular, do not hold a position twice. When the anatomy shifts between fields, the stitch either misaligns, or the software blends over a real position change, and the orthopedic measurement built on that image inherits the error.

This is why standing long-length exams favor bars, calipers and patient-support aids as much as detector specs — and why the practical guidance in the article on adjusting lumbar spine images reads as positioning advice as much as processing advice. The detector’s contribution here is speed and preview: an immediate low-dose-style preview after each field lets the operator confirm overlap and position before the patient leaves the wall stand, instead of discovering the problem at the review station.

Wired or wireless for a stitching workflow?

Stitched exams are a fixed-position workflow: the patient is against the wall stand, the panel sits in the bucky tray and does not move between fields. That makes a wired panel the natural fit for this exam type — no battery or connection variables between exposures, and one less thing that can differ between shot one and shot two. The broader trade-offs are laid out in the wired vs wireless comparison; for stitched work specifically, the argument tilts further toward the tethered option than everyday department work would suggest.

What to check before promising stitched exams

1. Confirm the stitching function exists in the software you are actually buying — the article on buying a detector without software explains why this line item is not automatic. Ask which workstation package includes long-length imaging, and get the answer in writing;
2. Ask how many fields the software supports and what overlap it expects — two- and three-field workflows behave differently in positioning and in exam time;
3. Test the repeat behavior — if one field fails, can the software replace a single field, or does the whole exam restart? For a scoliosis follow-up with a tired teenager, this decides whether the exam takes four minutes or fourteen;
4. Check calibration currency on your panel before your first stitched case — the uniformity reference the software joins against is only as good as the last gain calibration;
5. Decide whether you need it at all — if your orthopedic referrals are mostly localized (single spine segments, knees, hips), a standard panel with a dedicated large-format unit occasionally rented or referred out may be the more honest answer than promising a stitched workflow you will rarely run.

FAQ

Do I need a special flat panel detector for whole-spine stitching?

No special panel is required — stitching runs on standard 14×17 or 17×17 detectors, using two or three overlapping exposures joined by software. What the panel must provide is consistency between exposures: stable calibration, repeatable triggering and uniform response. A dedicated large-format panel is the alternative when you want the whole anatomy in one exposure without stitching software.

Why does my stitched image show a visible seam?

The most common causes are, in order: patient movement between exposures; detector response that drifted between the two shots (panel warming or an outdated gain calibration); and inconsistent exposure technique between fields. Fix the positioning and calibration items first — stitching software cannot compensate for inconsistent input data.

Is one large detector better than stitching two exposures?

It removes the seam problem entirely and shortens exam time, but large-format panels cost more, weigh more and may exceed your generator’s comfortable output. For clinics that run long-length exams occasionally, stitching on a standard panel is usually the pragmatic route; high-volume scoliosis and leg-alignment services justify the large-format investment.

Can stitched exams be done with a wireless detector?

Technically yes, but a wired panel suits the workflow better: stitched exams happen at the wall stand with the panel locked in the bucky tray, so the mobility advantage of wireless is unused, while the wired connection removes battery and link variables between exposures.

Author：Newheek-Detector