Which Lumbar, Cervical & Thoracic Spine Files Open Here

Archive formats

If your scan came as one big compressed file, drop it directly into the viewer — we unpack it in your browser.

• ZIP — the most common format on hospital CDs and patient-portal downloads
• TAR — common for research datasets
• .tar.gz and .tgz — TAR + GZIP compressed

DICOM transfer syntaxes

The technical format used to encode pixel data inside a DICOM file. Our viewer decodes every common transfer syntax used by clinical scanners.

• Implicit VR Little Endian (1.2.840.10008.1.2)
• Explicit VR Little Endian (1.2.840.10008.1.2.1)
• JPEG Lossless (1.2.840.10008.1.2.4.70)
• JPEG Extended 12-bit (used by i-CAT dental CBCT)
• RLE Lossless
• 8-bit, 12-bit, and 16-bit pixel data
• MONOCHROME1 and MONOCHROME2 photometric interpretations

NIfTI (research / converted scans)

NIfTI (Neuroimaging Informatics Technology Initiative) is the format used by most research and post-processing tools. If you ran your scan through a pipeline like FreeSurfer or fastMRI, you likely have a NIfTI file.

• .nii — uncompressed NIfTI volume
• .nii.gz — GZIP-compressed NIfTI (most common)

What a spine study usually contains

Lumbar, cervical, and thoracic MRI protocols share a similar skeleton: a sagittal T1 stack for marrow and disc height, a sagittal T2 stack for hydration and stenosis, and STIR when a fracture or oedema is suspected, since STIR suppresses fat signal and makes bone bruising or an acute compression fracture stand out clearly. Layered on top of that sagittal backbone is usually a separate axial stack per level or per pair of levels, angled to follow the tilt of each disc rather than running straight through the spine.

Spine X-ray series look different again. A standard lumbar or cervical series is a standing AP and lateral pair; an instability work-up adds flexion and extension laterals taken with the patient bending forward and arching back so a surgeon can see how much a segment slips between positions. Scoliosis clinics shoot full-length standing films that capture the entire spine in one exposure or as a stitched set of overlapping panels, so the whole curve can be measured on a single image.

Spine CT covers a third set of questions entirely: pars interarticularis defects, fusion hardware position, and fracture lines that MRI can under-call in dense bone. When a patient cannot have an MRI — a pacemaker or incompatible implant — a CT myelogram substitutes contrast injected into the spinal canal, imaged on CT instead, to show nerve root compression without magnetic fields.

How the files actually arrive

In practice a lumbar MRI folder from a hospital CD contains four or five short axial stacks — each covering only two or three disc levels — sitting beside the sagittal T1 and T2 series and a DICOMDIR index file at the disc root. A whole-spine screening study bundles that same pattern three times over, once per segment. Patient-portal downloads usually arrive as a single ZIP wrapping the same folder structure. EOS and other full-length radiograph systems export a single tall stitched image rather than a stack of small panels, and research exports converted through FreeSurfer-style pipelines typically land as .nii or .nii.gz volumes instead of raw DICOM.

Slice thickness and why spine 3D behaves differently

The angled per-disc axial stacks common in lumbar and cervical MRI are the reason spine volumes are harder to reconstruct than a knee or shoulder MRI: each stack only covers two or three levels and is tilted to a different plane than its neighbour, so the viewer cannot merge them into one continuous 3D volume — you review each angled stack on its own. Sagittal stacks acquired at 3–4 mm slice thickness limit volume rendering along the length of the spine even though they read perfectly well slice by slice. Spine CT is the exception: it is usually reconstructed as a thin, isotropic volume, which is exactly why surgeons request CT rather than MRI when they need a true 3D model for hardware planning. A single lateral or flexion-extension radiograph, meanwhile, stays a flat 2D image no matter how it was captured — there is no volume to build from one exposure.

Compatibility at a glance

Privacy and viewing a multi-stack spine study

Because a lumbar or cervical study can be split across half a dozen small stacks, it is tempting to upload each one somewhere different to check it — don't. Everything here decodes and renders locally in your browser, so dropping the entire folder at once is both the simplest and the most private way to review a segmented spine study; no stack is sent anywhere to be reassembled.

When a spine file won't open

The two spine-specific snags we see most: a study split across several angled axial stacks that a patient assumes should combine into one scrollable volume — they won't, because each stack was prescribed at its own angle, and that's expected behaviour, not a broken upload. The other is heavy streak artefact around pedicle screws and rods on post-fusion CT, which comes from the metal itself scattering the X-ray beam during the original scan and can't be removed after the fact. Outside of those two cases, an encrypted disc or a proprietary export outside the supported list is rare; ask your imaging centre for a standard DICOM export if either shows up. If a whole-spine screening study seems to be missing a segment after you drop the folder, check that every sub-folder from the cervical, thoracic, and lumbar acquisitions was copied off the disc — hospitals sometimes split a whole-spine order into three separate exam folders rather than one combined study.