How to Read an MRI Image: The Basics Patients Should Know

MRI can look mysterious at first: grey slices, technical labels, unfamiliar planes, and dozens – sometimes hundreds – of images. But every MRI study is organised in a logical way. Once you understand the basics, you can start to appreciate why MRI is so powerful, why scan quality matters, and why two scans with the same body-part name may not provide the same diagnostic information.

This guide is not designed to teach patients how to diagnose themselves. MRI interpretation is complex and should be performed by a radiologist. But understanding the basics can help you understand the screen more confidently, ask better questions, and recognise why protocol design matters.

First: An MRI Is Not One Image

An MRI scan is usually made up of multiple series.

A series is a group of images acquired using a particular MRI setting, plane, and sequence. For example:

Sagittal T1
Sagittal T2
Axial T2
Coronal STIR
Axial DWI
Post-contrast T1 fat-saturated

Each series shows the body in a different way. Some sequences show anatomy clearly. Others highlight fluid, inflammation, fat, blood products, restricted diffusion, or contrast enhancement.

A radiologist does not usually diagnose from one image. They move through many series together, comparing the same structure across different planes and tissue weightings (MRI settings).

The Information Displayed Around the MRI Image

When you open an MRI image in a viewing program, you will usually see text around the edges of the image. This information is important.

1. Patient name and identifiers

The image should display the patient’s name and often other identifiers such as date of birth, medical record number, or accession number. This ensures the correct scan is being reviewed.

2. Date and time of scan

The scan date matters, especially when comparing with previous imaging. A lesion, disc protrusion, fracture, tumour, inflammatory change, or postoperative finding may look very different over time.

For surveillance imaging, the date is critical. The key question is often not simply “what is there?” but “has it changed?”

3. Referring doctor

The referring doctor may be listed in the imaging metadata or on the report. This helps connect the scan to the clinical question being asked.

A high-quality MRI is not just a technical exercise. It should be matched to the clinical problem. A scan for knee meniscus injury is not the same as a scan for bone tumour. A spine MRI for disc disease is not the same as one designed to assess infection, tumour, inflammation, or nerve involvement.

4. Sequence name

The sequence name tells you what type of MRI information you are looking at. For example:

T1
T2
PD
STIR
FLAIR
DWI
ADC
GRE
SWI
T1 post-contrast
Others

This is one of the most important labels on the screen. The same anatomy can look very different depending on the sequence.

5. Plane of imaging

MRI images are commonly acquired or reconstructed in three main planes:

Axial – Slices run horizontally through the body, like looking from the feet upward or from the head downward.
Sagittal – Slices run from side to side, dividing the body into left and right portions. This is commonly used in spine, brain, knee, ankle, and many musculoskeletal scans.
Coronal – Slices run from front to back, dividing the body into front and back portions. This is useful for shoulders, hips, pelvis, knees, abdomen, and whole-body imaging.

Good MRI interpretation often depends on seeing the same structure in more than one plane. A finding that looks suspicious on one plane may be clarified, confirmed, or dismissed when viewed in another.

Slice Thickness: Why It Matters

MRI does not usually image the body as a continuous block. It often captures a stack of slices.

Slice thickness refers to how thick each image slice is. For example, a sequence may have:

5 mm slices
4 mm slices
3 mm slices
1 mm slices

Thinner slices usually provide more detail and reduce the chance of small structures being blurred together. This matters for subtle lesions, small nerves, cartilage defects, small tumours, tiny fractures, ligament fibres, and small areas of inflammation.

However, thinner slices can take longer to acquire and may produce more noise if the scan is not properly optimised. High-quality MRI is a balance between resolution, signal, time, and clinical purpose.

Slice Gap: The Space Between Images

Slice gap is the space between one slice and the next.

For example:

Slice thickness: 4 mm
Gap: 1 mm

This means the scanner captures a 4 mm slab, then skips 1 mm before the next slab.

A large gap can create a problem: small abnormalities may fall partly or entirely between slices. In high-quality imaging, especially where small lesions matter, the gap should be minimised or eliminated where possible.

Modern 3D MRI often has no traditional interslice gap because the data is acquired as a volume.

Number of Slices: More Is Not Always Better, But It Matters

The image viewer may show something like:

Image 12/36

This means you are looking at image 12 out of 36 in that series.

The number of slices tells you how much coverage was obtained and how finely the area was sampled. A spine MRI may have multiple sagittal and axial stacks. A knee MRI may include 20–40 images per series. A whole-body MRI may include hundreds or thousands of images across multiple body stations.

More slices can mean better coverage or thinner imaging, but only if the sequence is well designed. A large number of poor-quality images is not better than a smaller number of high-quality, clinically targeted images.

Isotropic Imaging: Why 3D MRI Can Be Powerful

Isotropic imaging means the MRI data is acquired with equal resolution in all directions.

For example, a 3D isotropic MRI may have voxels of approximately:

1 mm × 1 mm × 1 mm

This allows the scan to be reconstructed in different planes without major loss of detail. In practical terms, a high-quality 3D isotropic acquisition can allow the radiologist to review the anatomy in axial, sagittal, coronal, and oblique planes from the same dataset.

This can be very useful for:

Cartilage assessment
Small joints
Nerve imaging
Brain volumetric imaging
Tumour assessment
Spine and foraminal evaluation
Complex anatomy
Surgical planning

Not all MRI is isotropic. Many routine 2D sequences have thicker slices in one direction, which means they cannot be reformatted as cleanly.

The Main MRI Sequences Explained

T1-weighted imaging

T1 is excellent for anatomy.

Fat usually appears bright on T1. Fluid is usually darker. Bone marrow, fat planes, muscles, and anatomical boundaries are often well seen.

T1 is useful for:

Anatomy
Bone marrow assessment
Fat-containing lesions
Chronic structural change
Pre-contrast comparison
Post-contrast imaging when gadolinium is used

After contrast, T1 fat-saturated images are often used to show enhancement from inflammation, tumour vascularity, infection, or postoperative change.

T2-weighted imaging

T2 makes fluid bright.

This makes it very useful for detecting swelling, inflammation, oedema, cysts, joint fluid, disc hydration, and many pathological processes.

T2 is useful for:

Fluid
Oedema
Inflammation
Cysts
Disc and joint assessment
Tumour water content
Infection-related change

A simple way to remember it: T2 often makes pathology more visible when pathology contains water.

Proton density imaging

Proton density, often called PD, is commonly used in musculoskeletal MRI.

It provides high anatomical detail and is particularly useful in joints. PD sequences are often combined with fat suppression to make fluid, oedema, and subtle injury more conspicuous.

PD is useful for:

Menisci
Ligaments
Tendons
Cartilage
Labrum
Joint surfaces
Small musculoskeletal injuries

Many knee, shoulder, ankle, wrist, and elbow MRI protocols rely heavily on PD sequences.

Fat-suppressed sequences

Fat suppression is not a sequence by itself, but a technique applied to sequences such as T2, PD, or T1 post-contrast.

Fat normally produces high signal on many MRI images. Suppressing fat can make fluid, inflammation, oedema, and enhancement easier to see.

Fat-suppressed imaging is useful for:

Bone marrow oedema
Soft tissue injury
Synovitis
Tumour visibility
Inflammation
Post-contrast enhancement

STIR imaging

STIR stands for Short Tau Inversion Recovery.

It is a robust fat-suppression technique that makes fluid and oedema bright while suppressing fat. STIR is widely used in spine, musculoskeletal, trauma, infection, inflammatory, and whole-body MRI protocols.

STIR is useful for:

Bone marrow oedema
Fracture stress response
Infection
Inflammatory change
Tumour screening
Whole-body marrow and soft-tissue assessment

STIR is sensitive, but not always specific. It can show that something is abnormal, but other sequences are often needed to determine exactly what it is.

FLAIR imaging

FLAIR stands for Fluid-Attenuated Inversion Recovery.

It is mainly used in brain MRI. FLAIR suppresses normal cerebrospinal fluid, making many brain lesions easier to see.

FLAIR is useful for:

White matter lesions
Multiple sclerosis assessment
Stroke-related change
Brain inflammation
Tumour-related oedema
Subtle abnormalities near ventricles or sulci

DWI/DWIBS and ADC

DWI stands for Diffusion-Weighted Imaging, DWIBS stands for Diffusion-Weighted Imaging with Background Suppression. ADC stands for Apparent Diffusion Coefficient.

These sequences assess the movement of water molecules within tissue. Areas where water movement is restricted may appear bright on DWI and dark on ADC.

DWI is important in:

Acute stroke
Tumour detection and characterisation
Infection and abscess assessment
Whole-body oncology MRI
Lymph node and marrow assessment
Some abdominal and pelvic imaging

DWI is powerful, but it must be technically well performed. Poor diffusion imaging can create distortion, false positives, or false reassurance.

GRE and SWI

GRE stands for Gradient Echo. SWI stands for Susceptibility-Weighted Imaging.

These sequences are sensitive to magnetic susceptibility effects, which makes them useful for detecting blood products, microbleeds, calcification, and some metal-related effects.

They are useful for:

Brain microbleeds
Haemorrhage
Prior trauma
Cavernomas
Calcification versus blood product assessment
Some cartilage and joint applications

T1 post-contrast imaging

Some MRI scans use gadolinium contrast. After contrast is injected, T1-weighted images are obtained, often with fat suppression.

Contrast can help assess:

Tumour enhancement
Infection
Inflammation
Synovitis
Postoperative scar versus recurrent disease
Vascularity
Breakdown of normal tissue barriers

Not every MRI requires contrast. Whether contrast is useful depends on the clinical question.

What the Patient Usually Does Not See

An MRI image contains more information than is usually obvious to the patient.

Behind the visible image is a large amount of technical data, including:

Sequence parameters
Repetition time
Echo time
Inversion time
Flip angle
Field of view
Matrix size
Pixel spacing
Slice thickness
Slice gap
Coil used
Acquisition time
Diffusion b-values
Reconstruction method
Scanner strength
DICOM metadata

Some of this information may be visible in advanced DICOM viewers. Some may be hidden in the image metadata. Some raw acquisition information is generally not available through standard patient portals.

This matters because image quality is not determined only by what the image looks like at first glance. The technical design of the scan affects what can and cannot be reliably seen.

Why MRI Quality Varies So Much

Two MRI scans can have the same name but very different diagnostic capability.

For example, “MRI knee” may mean a short, basic protocol with thicker slices and limited sequences. Or it may mean a high-resolution protocol with carefully selected planes, thin slices, appropriate fat suppression, cartilage-sensitive imaging, and targeted sequences matched to the clinical question.

The same applies to spine, brain, prostate, breast, abdomen, nerve imaging, and whole-body MRI.

Important quality factors include:

Magnet strength
Coil quality
Patient positioning
Sequence selection
Slice thickness
Slice gap
Motion control
Fat suppression quality
Field of view
Resolution
Diffusion quality
Scan time
Radiologist expertise
Whether the protocol matches the clinical question

A scan can be technically “completed” but still be limited for the diagnosis being considered.

A Simple Way to Read an MRI Screen

When looking at an MRI image, start with these questions:

1- Is this the correct patient?
Check the name and identifiers.

2- When was the scan performed?
The date matters, especially for comparison.

3- What body part and side am I looking at?
Look for labels and orientation markers.

4- What plane is this?
Axial, sagittal, or coronal?

5- What sequence is this?
T1, T2, PD, STIR, FLAIR, DWI, ADC, GRE, SWI, post-contrast or other?

6- How many images are in the series?
Look for image number and total slice count.

7- How thick are the slices? Is there a gap?
Thicker slices and larger gaps can reduce sensitivity for small findings.

8- Is the imaging 2D or 3D isotropic?
This affects whether the data can be reconstructed in other planes without major loss of detail.

9- Does the scan answer the clinical question?
This is the most important question – and one that often requires radiologist input.

The Bottom Line

Reading an MRI image begins with understanding that MRI is not one picture. It is a structured dataset made from multiple sequences, planes, slice thicknesses, and technical choices.

The labels around the image – sequence name, plane, slice number, slice thickness, gap, date, patient name, and referring doctor – are not just administrative details. They help explain what information has been captured and how reliable that information may be.

For patients, the key message is simple: MRI quality matters. A scan is only as useful as the protocol behind it, the technical quality of the acquisition, and the expertise used to interpret it. A high-quality MRI does not just show the body – It is designed to answer a specific clinical question with the highest possible confidence.

Frequently Asked Questions About Reading an MRI Scan

If you have any further questions please have a look at our FAQs page

Can I read my own MRI images?

You can learn to recognise basic features such as the body part, imaging plane, sequence name, slice number and scan date. However, diagnosing abnormalities on MRI is complex and should be performed by a radiologist who interprets multiple sequences together and considers your clinical history and previous imaging.

Why does an MRI have so many different images?

An MRI is not a single picture. It is made up of multiple series, with each series designed to show different tissue characteristics.

Some sequences demonstrate anatomy, while others highlight fluid, inflammation, fat, blood products, restricted diffusion or contrast enhancement. The radiologist combines information from all of these sequences when interpreting the examination.

What do T1, T2, PD and STIR mean on an MRI?

These are different types of MRI sequences.

T1 is particularly useful for anatomy, fat and bone marrow.
T2 makes fluid more conspicuous and is useful for oedema and inflammation.
PD, or proton density, is commonly used in joint and musculoskeletal imaging.
STIR suppresses fat and makes oedema and many inflammatory abnormalities appear bright.

No single sequence provides all the information required for diagnosis.

What do axial, sagittal and coronal mean?

These describe the direction in which the body is being viewed.

Axial – horizontal slices through the body.
Sagittal – divides the body into left and right portions.
Coronal – divides the body into front and back portions.

Seeing a structure in several planes can help confirm whether a finding is real and determine its extent.

Why does MRI slice thickness matter?

Slice thickness affects the amount of anatomical detail that can be seen.

Thinner slices can improve assessment of small structures such as nerves, cartilage, ligaments, small tumours and subtle fractures. Larger slice gaps can potentially allow very small abnormalities to fall partly between images.

However, MRI quality is always a balance between resolution, signal, scanning time and the clinical question.

Is 3D MRI better than normal MRI?

3D isotropic MRI can be particularly powerful because it acquires data at similar resolution in all directions. This allows the radiologist to reconstruct the anatomy in different planes without substantial loss of detail.

It can be useful for cartilage, nerve imaging, brain volumetric assessment, complex spinal anatomy, tumours and surgical planning. However, 3D imaging is not automatically better for every clinical problem.

What are DWI and ADC on an MRI?

DWI stands for Diffusion-Weighted Imaging, while ADC stands for Apparent Diffusion Coefficient.

These sequences assess the movement of water molecules within tissues and can provide important information in conditions such as acute stroke, infection and some tumours. Diffusion imaging is also an important component of high-quality whole-body oncology MRI.

DWI should not usually be interpreted in isolation. The radiologist compares it with ADC and the other MRI sequences.

Why can two MRI scans of the same body part be very different?

Because the name of the examination does not define the quality of the protocol.

Two examinations both called an “MRI knee” or “MRI lumbar spine” may use very different sequences, slice thicknesses, resolution, scan times and imaging techniques.

MRI protocols should ideally be designed around the clinical question being investigated. A routine spine MRI, for example, may be different from one designed to assess tumour, infection, inflammation or nerve pathology.

Can AI review my MRI with confidence?

Not with the same level of confidence as a specialist radiologist reviewing the complete examination.

AI can increasingly assist with specific imaging tasks, but an MRI consists of multiple sequences, planes and technical parameters that must be interpreted together and in the context of the patient’s symptoms, medical history and previous imaging.

There is also an important difference between AI being given a few screenshots and AI having access to the complete diagnostic-quality DICOM examination and clinical context. Individual screenshots may omit sequences, slices, technical information or subtle findings that materially change the interpretation.

AI can be useful as an additional tool, but patients should not rely on a consumer AI interpretation of selected MRI images to exclude disease or replace a formal radiology report.

Can an MRI be technically completed but still miss something?

Potentially, yes.

Completing an MRI examination does not automatically mean that every possible abnormality can be assessed confidently. Diagnostic sensitivity can be affected by motion, thick slices, large slice gaps, inadequate coverage, poor fat suppression, low spatial resolution or a protocol that was not designed for the condition being investigated.

This is why the scanner, protocol, technical quality and radiologist interpretation all matter.

What is the most important thing to understand when looking at an MRI?

The most important point is that an MRI is not one image.

It is a structured dataset made up of multiple sequences, planes and technical acquisitions. The diagnostic value of MRI depends on what was acquired, how well it was acquired and how the complete examination is interpreted.

Author: Dr Kirralee Sherif (PhD)