Open a hand or finger MRI DICOM series in the browser to step through the flexor and extensor tendons, the annular pulleys, the thumb ulnar collateral ligament, and the small joints of the fingers without installing anything. Ask for an optional AI walkthrough of what a rendered pulley injury, a suspected Stener lesion, or a marrow signal change might mean before your appointment. The original files stay on your own device the whole time.
Upload Your Hand MRIThe axial plane cuts across the finger like rings on a tree trunk. Scrolling through it along the length of a digit is how you follow the flexor tendon from the palm to the tip, watch the A2 and A4 annular pulleys hold that tendon snug against the bone at each pulley window, and compare the radial and ulnar collateral ligaments on either side of a joint at the same slice level.
The coronal plane opens the hand out flat, side to side. It is the plane that shows the joint surfaces squarely — the metacarpal head against the proximal phalanx, or the base of the thumb metacarpal against the trapezium — and it is the plane a radiologist reaches for first when a thumb has been forced sideways, because the ulnar collateral ligament of the thumb runs across this view and a torn end curled back over the adductor aponeurosis is the classic sign of a Stener lesion.
The sagittal plane slices the finger front to back. It profiles the volar plate on the palm side of each joint and lays the extensor mechanism out along the back of the digit, from the central slip over the middle joint to the terminal tendon at the fingertip, which is why a sagittal series is usually the fastest way to spot a boutonnière or mallet pattern.
Fat-saturated, fluid-sensitive sequences suppress the bright fat signal from the marrow and subcutaneous tissue so that anything wet stands out instead: bone marrow oedema under a hairline fracture, fluid distending a tendon sheath in tenosynovitis, and the margins of a soft-tissue mass such as a ganglion or a giant cell tumor of the tendon sheath. Without fat suppression, that same fluid can hide against the naturally bright fat around it.
A dedicated single-digit acquisition uses a small surface coil wrapped around one finger, which sits close enough to pick up a much stronger signal from a small field of view. That buys the sub-millimeter in-plane resolution needed to separate a pulley from the tendon it restrains, or to size a partial collateral ligament tear at the PIP joint. A whole-hand study trades some of that per-digit detail for coverage — it captures every metacarpal, every joint row, and every digit in one series, which is the right call when the clinical question is broader, such as widespread synovitis or a suspected fracture whose exact location is not yet known. Reports for the two acquisitions read differently: a single-digit report is dense with pulley and tendon detail for that one finger, while a whole-hand report surveys row by row across all five digits.
Drop your MRI ZIP file or individual .dcm files. Supports all standard DICOM transfer syntaxes including JPEG Lossless.
Navigate through slices with keyboard arrows or the slider. Adjust windowing, zoom, pan, and measure distances.
4 AI models (GPT, Gemini, Claude, Grok) analyze independently. Claude synthesizes a unified report with findings and confidence levels.
A finger MRI stack is short compared to most joint studies — often a few dozen sub-millimeter slices instead of a few hundred — so scrolling through the whole sequence for one digit takes seconds rather than minutes. Use the keyboard arrows or the slider to step slice by slice, then window and level to swap between a tendon-and- pulley preset and a bone preset without leaving the image.
The measurement tool turns a visual impression into a number: drag it along the expected course of a flexor tendon to size a bowstringing gap where a pulley has ruptured, or across a fluid collection to record how large a suspected ganglion or soft-tissue mass really is.
When a study includes more than one digit, or paired pre- and post-contrast series, open several of those series side by side so the same joint on two different sequences sits in view at once. When the source acquisition has enough slices at a fine enough spacing, the viewer can also reconstruct additional planes from that single series locally in your browser, which is useful for confirming a pulley or ligament finding seen on one plane against another without re-scrolling from the start.
The metacarpophalangeal, or MCP, joints sit at the knuckles where the hand meets the fingers. A report on this row usually comments on the collateral ligaments on either side of each knuckle, the volar plate underneath, and — almost always singled out separately from the other four — the thumb metacarpophalangeal joint, where an ulnar collateral ligament tear is graded by whether the torn end still lines up with the joint or has flipped past the adductor aponeurosis into a Stener lesion.
The proximal interphalangeal, or PIP, joints are the middle knuckle of each finger and the joint most often injured by a jammed or hyperextended finger. Expect comments on the volar plate, which can avulse a small bone fragment, and on the central slip of the extensor tendon just above the joint, because a missed central slip injury here is what leads to a boutonnière deformity months later.
The distal interphalangeal, or DIP, joints are the fingertip knuckles. Reports on this row focus on the terminal extensor tendon insertion, since an avulsion here — with or without a bone chip — is the mallet finger pattern, and on the flexor digitorum profundus insertion on the volar side, since a jersey finger avulses that tendon from the same joint level on the opposite surface.
Raw MRI files stay on your device for DICOM parsing, rendering, and viewing. If you request AI analysis, the app uses rendered images and minimal context.
Works in any modern browser — Chrome, Firefox, Safari, Edge. No plugins, no Java, no desktop software to download.
Four independent AI models can explain rendered MRI images. Each model provides its own findings, then Claude synthesizes a consensus report.
Multi-plane navigation, windowing presets (bone, soft tissue, brain), zoom up to 4x, pan, and distance measurement in millimeters.
T1, T2, PD, STIR, FLAIR, gradient echo, and fat-suppressed sequences. Sagittal, coronal, and axial planes.
Generate a professional report with structured findings, severity grading, model agreement, and recommendations for your physician.
A dozen sub-millimeter slices over one finger deliver outstanding in-plane detail for that digit, but the slice spacing along the length of the finger is rarely small enough, and the field of view is rarely wide enough, to build a smooth volume the way a full isotropic dataset can. A hand CT, by contrast, is typically acquired as one continuous isotropic block covering every bone from the wrist to the fingertips, with voxels the same size in every direction — exactly the shape 3D and volume-rendering algorithms need to sculpt a rotatable bone model. That is why a finger MRI series is read plane by plane for soft-tissue and marrow detail, while a hand CT is the study reached for when a surgeon wants a 3D model of a comminuted fracture or a complex bony deformity.
DICOM (.dcm) — individual slice files
ZIP archives — containing DICOM series folders
JPEG Lossless (1.2.840.10008.1.2.4.70) transfer syntax
Explicit VR Little Endian, Implicit VR Little Endian
8-bit, 12-bit, and 16-bit pixel depth
MONOCHROME1 and MONOCHROME2 photometric interpretation
Fluid-sensitive fat-suppressed sequences highlight marrow oedema, tenosynovitis, and soft-tissue masses against the surrounding fat. T1 sequences carry the anatomic detail — tendon outlines, cortical margins, and marrow fat pattern — that the fat-suppressed series alone does not show.
Yes. Axial images through the digit can show pulley thickening, oedema, or an outright gap, and the flexor tendon bowstringing away from the bone underneath the torn pulley. Dynamic ultrasound during active finger flexion can add information MRI cannot, since MRI is a static acquisition.
Coronal images through the thumb metacarpophalangeal joint show whether the torn ulnar collateral ligament end still lies against the joint or has flipped superficial to the adductor aponeurosis. That flipped position is the Stener lesion, and it changes management from bracing to surgical repair because the ligament cannot heal in place once displaced.
A single-digit acquisition on a small surface coil gives the sharpest detail for one finger and is typically ordered when the clinical exam already points to that digit. A whole-hand study trades some per-digit resolution for coverage of every metacarpal and joint row, which suits a broader or less localized complaint. That choice is made by the ordering clinician and radiographer, not by the viewer.
Each joint row fails in different, characteristic ways — thumb ulnar collateral ligament tears at the MCP row, central-slip and volar-plate injuries at the PIP row, and terminal-tendon or flexor-profundus avulsions at the DIP row — so grouping findings by joint level keeps the report matched to how these injuries are actually diagnosed and treated.
When a series has enough slices at a fine enough spacing, the viewer can reconstruct additional planes from that data locally in your browser. A short single-finger stack usually lacks the through-plane resolution for a smooth volume render — that is the strength of an isotropic hand CT instead. Reconstruction quality always depends on how the original series was acquired.
No. AI explanation reviews the rendered images you choose to share and describes what it observes in plain language to help you prepare questions for your clinician. It does not replace a hand surgeon or radiologist's interpretation of your case.