Free Shoulder MRI Viewer Built Around the Cuff and Labrum
Load the shoulder DICOM stack straight from the scanner disc and step through the oblique coronal, oblique sagittal, and axial runs the way a shoulder protocol actually acquires them, then ask for an AI read on the frames that show a cuff tear, a labral lesion, or fluid tracking through the subacromial bursa. Nothing leaves your device until you choose to send a rendered frame for review.
Opens on this device. Nothing is uploaded.
What Each Plane in a Shoulder MRI Series Is Actually For
A shoulder protocol is not one volume viewed from three angles — it is a set of purpose-built acquisitions, each angled to a different tendon or joint surface. The oblique coronal plane is prescribed parallel to the supraspinatus tendon itself, which is the only orientation that shows true tear thickness and how far a torn tendon has retracted toward the glenoid. The oblique sagittal plane cuts perpendicular to that, through the supraspinatus outlet, and is where a radiologist checks rotator cuff muscle bulk for fatty atrophy, assigns a Goutallier grade, and reads the acromion's shape for the hook that predicts impingement. Axial slices run through the joint from top to bottom and are the only plane that shows the subscapularis tendon end-on, the biceps tendon sitting in its groove, and both the anterior and posterior labrum in the same image. When the study includes an MR arthrogram, an additional ABER series — arm abducted and externally rotated — stretches the anteroinferior capsule and labrum so a subtle Bankart-type tear opens up and separates from the adjacent capsule. Fat-saturated T2 sequences run through several of these planes to null the bright fat signal so fluid, edema, and bursitis show up unambiguously bright against a dark background.
What Radiologists Flag on a Shoulder MRI
- Tear thickness and tendon retraction measured on the oblique coronal supraspinatus view
- Muscle atrophy graded on the oblique sagittal outlet view, plus acromion shape linked to impingement risk
- Subscapularis, biceps groove, and anterior or posterior labral abnormalities on axial slices
- Anteroinferior labral and capsular tears surfaced by ABER positioning on an arthrogram
- Subacromial-subdeltoid bursitis and joint fluid on fat-saturated T2
- Hill-Sachs and bony Bankart defects following a prior dislocation
Shoulder Conditions on MRI
How It Works
Upload DICOM Files
Drop your MRI ZIP file or individual .dcm files. Supports all standard DICOM transfer syntaxes including JPEG Lossless.
View in Browser
Navigate through slices with keyboard arrows or the slider. Adjust windowing, zoom, pan, and measure distances.
Get AI Analysis
4 AI models (GPT, Gemini, Claude, Grok) analyze independently. Claude synthesizes a unified report with findings and confidence levels.
Matching the Viewer to What a Shoulder Report Names
A shoulder report reads like a checklist tied to specific slices, and the viewer is built to follow it that way rather than as a single generic slider. Jump to the mid-substance of the oblique coronal stack to find the supraspinatus footprint the report is measuring in millimeters. Move a few slices anterior on the axial run to confirm whether a described subscapularis tear is partial or full-thickness against the lesser tuberosity. If the report references “anchor points” or a SLAP classification, the labrum sits at the joint margin on the same axial images, just above and below the biceps anchor. When an arthrogram was performed, toggle between the plain fluid-sensitive series and the contrast-filled ABER run side by side — a labral tear that is ambiguous on one often separates cleanly from the capsule on the other once joint fluid or injected contrast outlines it. A mention of the acromioclavicular joint usually points to the most superior coronal slices, where a hooked or downsloping acromion and any AC joint spurring sit directly above the supraspinatus footprint — worth checking alongside the cuff itself since impingement from above and a cuff tear often appear together on the same study.
3D Shoulder MRI and MPR Viewing
MPR reconstruction works best on volumes with thin, evenly spaced, isotropic slices — the kind a CT scanner produces by default. A shoulder MRI protocol is the opposite: the oblique coronal, oblique sagittal, and axial series are each acquired separately, at their own angle, often with gaps between slices, specifically so each plane frames a different tendon at its most informative angle. Reslicing one of those series into another orientation in the browser tends to blur or step the anatomy rather than reveal it, so the viewer reconstructs a volume only when a given series has enough closely spaced slices to support it, and otherwise keeps each acquired plane as its own stack.
Whether reconstruction is available depends on the specific series, its slice thickness and spacing, and your browser or device capability. AI analysis remains a support tool for clinician review, not a diagnosis, and does not replace clinician interpretation.
AI Explanation vs a Radiology Second Opinion
Asking the viewer to explain a rendered shoulder frame is not the same as getting a second radiology read — it summarizes what is visible in the image you sent and helps translate report terms like partial-thickness tear, Goutallier grade, or Bankart lesion into plain language you can bring to an appointment. Treat it as preparation for that conversation, particularly when the report describes a cuff tear size, a labral tear location, or biceps tendon instability that changes the surgical decision.
Tools for Working Through a Shoulder Stack
Once a shoulder series is loaded, the same stack can be scrolled slice by slice with the keyboard or a slider, and the window and level can be pulled toward bone to check the greater tuberosity or toward soft tissue to separate tendon from fluid. A measurement tool lets you drag across the visible gap in a torn supraspinatus tendon and read the width in millimeters against the same scale the radiologist used, rather than eyeballing it. When both a plain fluid-sensitive series and a contrast arthrogram series exist for the same shoulder, they can be opened side by side so a labral finding can be checked against both before you decide it is worth flagging. Where the acquisition supports it, the viewer also builds multi-planar reconstruction views from the raw slices, entirely in your browser. Zooming into the axial slices at the biceps groove is useful on its own — that groove is narrow enough that a subluxed or dislocated biceps tendon can be easy to miss at the default zoom level, and a torn transverse ligament often shows up only once you have zoomed in past the point where the tendon should be centered.
Why Use Our Free MRI Viewer
Browser-Based Viewing
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.
No Installation Required
Works in any modern browser — Chrome, Firefox, Safari, Edge. No plugins, no Java, no desktop software to download.
AI-Powered Analysis
Four independent AI models can explain rendered MRI images. Each model provides its own findings, then Claude synthesizes a consensus report.
Advanced Viewing Tools
Multi-plane navigation, windowing presets (bone, soft tissue, brain), zoom up to 4x, pan, and distance measurement in millimeters.
All MRI Sequences Supported
T1, T2, PD, STIR, FLAIR, gradient echo, and fat-suppressed sequences. Sagittal, coronal, and axial planes.
Downloadable PDF Report
Download an educational AI summary with structured observations, suggested severity grades, and areas of model agreement to discuss with your doctor.
Supported MRI Formats
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
Frequently Asked Questions
Why does my shoulder MRI have separate oblique series instead of one volume?
The technologist angles each acquisition to a specific structure — the oblique coronal plane runs parallel to the supraspinatus tendon, and the oblique sagittal plane cuts through the cuff outlet at roughly a right angle to it. Neither angle lines up with a standard axial, coronal, or sagittal grid, which is why the viewer keeps them as separate stacks rather than forcing them into one shared volume.
Which plane shows a rotator cuff tear best?
The oblique coronal series, taken parallel to the supraspinatus, is where tear thickness and tendon retraction are usually measured. The oblique sagittal outlet view adds muscle bulk and acromion shape, and axial slices confirm whether the subscapularis anteriorly or the infraspinatus posteriorly are also involved.
What is an ABER series and why does my study include one?
ABER stands for abducted and externally rotated — a positioning used during a shoulder MR arthrogram that stretches the front-lower capsule and labrum. It is typically added when a prior dislocation is suspected, because that position separates a subtle anteroinferior labral tear from the surrounding capsule far more clearly than a standard axial view.
Can this viewer measure how retracted a torn tendon is?
Yes. The measurement tool works on the same oblique coronal slices a radiologist uses, so you can drag across the visible tendon gap or retraction distance and read the result in millimeters. It is a measurement aid, not a substitute for the radiologist's own reading of retraction relative to the glenoid.
Why can't the viewer build a full 3D model from my shoulder study?
Shoulder protocols favor thin slices angled precisely to the cuff and labrum over the thick, evenly spaced, isotropic slices that 3D reconstruction needs. When a specific series in your study has enough closely spaced slices, the viewer will reconstruct additional planes from it; when it does not, each acquired plane stays as its own stack rather than producing a distorted reconstruction.
Is my shoulder MRI data private?
Yes. DICOM parsing and rendering happen entirely in your browser using JavaScript and WebAssembly, and the raw files never leave your device. Only the specific frames you choose to submit are sent as rendered images for AI review.
Does the AI review distinguish a labral tear from normal anatomic variants?
Four independent models review the rendered frames and a synthesis pass flags where they agree or disagree, which is useful for catching an ambiguous labral finding worth asking about. It cannot reliably separate a true tear from a sublabral recess or a Buford complex the way a musculoskeletal radiologist can, so treat any labral finding as a discussion point for your surgeon rather than a conclusion.
See all supported file formats → for the complete list.
Shoulder Imaging Focus
On Analyze My Shoulder, the viewer helps you review shoulder MRI, X-ray, and CT images for common report terms before discussing care with a clinician.
Common shoulder findings
Review rotator cuff tears, shoulder labral tears, SLAP and Bankart patterns, impingement, clavicle fracture, AC joint separation, dislocation patterns, arthritis, and frozen shoulder.
Review the right structure
Shoulder reports often mix tendon, labrum, cartilage, fracture, and bony alignment language. These guides help separate the main structures and match CT, X-ray, or MRI findings to the right next page.
Trust and safety
Use the viewer and AI output as informational support. MRI can help with labrum and impingement questions, X-ray can help screen fracture or alignment, and CT can clarify bone detail, but diagnosis and treatment decisions need clinician review.
Shoulder MRI safety context
If a report mentions SLAP tear, Bankart lesion, dislocation, glenoid bone loss, or rotator cuff injury, compare the images with symptoms and exam findings. A clinician may choose radiographs, MRI, MR arthrography, CT, or CT arthrography based on the question.