MAGNETIC RESONANCE IMAGING

RATIONALE:

Magnetic resonance imaging (MRI) is a test that uses a magnetic field and pulses of radio wave energy to make pictures of organs and structures inside the body. In many cases, MRI gives different information about structures in the body than can be seen with an X-ray, ultrasound, or computed tomography (CT) scan. MRI also may show problems that cannot be seen with other imaging methods. This highlights the great importance of the study of this technology and understanding of the fundamentals upon which to build

1. Historical Development of MRI

Magnetic Resonance Imaging (MRI) is a sophisticated imaging modality derived from the principles of nuclear magnetic resonance (NMR)—a technique originally used in chemistry to study molecular structure.

Key Milestones

• 1946 – Felix Bloch and Edward Purcell independently discovered magnetic resonance → Nobel Prize (1952)

• 1950–1970 – NMR used for chemical analysis

• 1971 – Raymond Damadian discovered differences in relaxation times between normal and diseased tissues → foundation for medical MRI

• 1973

  • CT introduced (Hounsfield)

  • Paul Lauterbur produced the first MR images using gradients

• 1975 – Richard Ernst introduced Fourier Transform MRI → basis of modern MRI

• 1977

  • Damadian built the first full-body MRI scanner

  • Peter Mansfield developed Echo-Planar Imaging (EPI)

• 1980s–1990s

  • Faster imaging (minutes → seconds)

  • Development of MRA and fMRI

• 2003 – Nobel Prize awarded to Lauterbur and Mansfield

Key Insight

MRI evolved from chemistry → physics → clinical imaging, becoming one of the most powerful tools for soft tissue visualization.

2. Why MRI is Important

Compared to X-ray and CT:

Core Advantage of MRI

• Exceptional contrast resolution

• Can differentiate minute soft tissue differences

• Flexible imaging parameters (pulse sequences)

3. MRI Hardware

MRI systems consist of several critical components:

Main Components

1. Magnet (B₀ field)

2. Gradient system

3. RF (Radiofrequency) coils

4. Receiver system

5. Computer (image reconstruction)

4. Types of MRI Magnets

4.1 Permanent Magnets

• Made from ferromagnetic material

• Typically ≤ 0.4 Tesla

Advantages:

• Low cost

• No cryogens

• Open design → less claustrophobia

Disadvantages:

• Heavy

• Low field strength → lower image quality

4.2 Resistive Magnets

• Use an electric current through coils

Advantages:

• Lower cost

• Can be turned off

Disadvantages:

• High power consumption

• Heat production

• Limited field strength (~0.6 T)

4.3 Superconducting Magnets (MOST IMPORTANT)

• Use materials like niobium-titanium

• Cooled with liquid helium (~4 K)

Advantages:

• High field strength (1.5–3T clinically)

• High signal-to-noise ratio (SNR)

• Fast imaging

Disadvantages:

• Expensive

• Requires cryogens

• Technically complex

5. Shimming (Magnetic Field Correction)

MRI requires an extremely uniform magnetic field.

Types:

• Passive Shimming → metal pieces adjust the field

• Active Shimming → electric coils correct the field

Purpose:

To ensure image accuracy and resolution

6. RF Coils (Critical for Signal)

RF coils transmit and receive signals.

6.1 Volume Coils

• Large, uniform field

• Used for whole-body imaging

Limitation:

• Lower SNR (captures noise from a large area)

6.2 Surface Coils

• Placed directly over anatomy

Advantages:

• High SNR

• High resolution

Disadvantage:

• Limited depth penetration

6.3 Quadrature Coils

• Two coils at 90°

• Produce √2 more signals

6.4 Phased Array Coils

• Multiple small coils combined

Key Benefit:

• Large coverage + high SNR

7. Atomic Structure and MRI Basis

MRI is based on hydrogen atoms because:

1. Abundant in the body (water, fat)

2. High gyromagnetic ratio

Structure:

• Proton (positive)

• Neutron (neutral)

• Electron (negative)

8. Magnetization Concept

Hydrogen protons behave like tiny magnets.

Without a magnetic field:

• Random orientation → no net magnetization

With magnetic field (B₀):

• Align parallel (low energy) or anti-parallel (high energy)

Larmor Equation (VERY IMPORTANT)

ω0=γB0\omega_0 = \gamma B_0ω0​=γB0​

Where:

• ω₀ = Larmor frequency

• γ = gyromagnetic ratio

• B₀ = magnetic field strength

👉 This determines the exact frequency needed for resonance

9. RF Pulse and Excitation

When RF pulse = Larmor frequency:
→ Protons absorb energy
→ Magnetization flips (e.g., 90°)

Result:

• Longitudinal magnetization → Transverse plane

10. Relaxation Processes (CORE MRI PHYSICS)

After excitation, protons return to equilibrium.

10.1 T1 Relaxation (Longitudinal)

• Recovery along the Z-axis

• Energy is released to the surroundings

Definition:

Time to recover 63% of the original magnetization

Key Idea:

• Fat → short T1 (fast recovery)

• Water → long T1 (slow recovery)

10.2 T2 Relaxation (Transverse)

• Occurs in the XY plane

• Due to the dephasing of spins

Definition:

Time to decay to 37% of the original signal

10.3 Key Differences

10.4 T2* Relaxation

Includes:

• True T2 decay

• Magnetic field inhomogeneities

Relationship:

T2* < T2

11. Signal Formation (FID)

After RF pulse:

• Spins emit an RF signal

• Signal decays → Free Induction Decay (FID)

👉 This is the raw MRI signal

12. Image Acquisition

• RF coils detect signal

• Must be perpendicular to B₀

• Poor positioning → noise

13. Fourier Transform (VERY IMPORTANT)

MRI signal initially:

• Time domain (FID)

To create an image:
→ Convert to frequency domain

Purpose:

• Separate signals by frequency

• Identify spatial information

Conceptual Meaning

• Time signal → raw data

• Frequency signal → image components

14. Why MRI Produces High Contrast

Different tissues have different:

• T1 times

• T2 times

• Proton densities

👉 This creates natural contrast without contrast agents

15. Key Concept Summary (MUST MASTER)

1. Hydrogen protons behave like tiny magnets

2. The external field (B₀) aligns them

3. RF pulse excites them

4. Relaxation produces a signal

5. Signal → processed via Fourier Transform → image

FINAL UNDERSTANDING (BIG PICTURE)

MRI is not just imaging—it is a controlled manipulation of atomic behavior:

• You align protons

• Excite them using RF energy

• Observe how they relax

• Convert signals into images

👉 In essence:
MRI translates microscopic atomic behavior into macroscopic medical images.

16 Why Gradient Coils Are Essential in MRI

In MRI, after hydrogen nuclei are excited by a radiofrequency (RF) pulse, they emit signals as they return to equilibrium (this is the Free Induction Decay, or FID).

However, a major problem arises:

• If the magnetic field (B₀) were perfectly uniform,

• All protons would precess at the same Larmor frequency,

• And they would emit identical signals.

👉 This means:

✅ Solution: Gradient Coils

Gradient coils slightly vary the magnetic field across space, causing:

• Different frequencies

• Different phases

• Different signal amplitudes

This allows MRI to localize signals spatially.

17. Role of Voxels in MRI

The body is divided into tiny 3D units called voxels.

Each voxel has:

• Frequency → determined by magnetic field strength

• Phase → determined by gradient timing

• Amplitude → depends on the number of protons

👉 This combination makes each voxel unique and identifiable.

18. Gradient Coils and Spatial Encoding

MRI uses three gradient coils, each corresponding to a spatial axis:

These gradients modify the magnetic field linearly, so:

19. Three Steps of MRI Signal Encoding

MRI localization occurs in three major steps:

4.1 Slice Selection (Z-gradient)

🔹 Principle:

• Apply the Gz gradient + RF pulse simultaneously

• The magnetic field varies along the body

• Therefore, frequency varies along the body

🔹 Key Idea:

👉 This selects a thin slice of tissue

🔹 Example:

• Feet: ~63.5 MHz

• Center: ~63.6 MHz

• Head: ~63.7 MHz

If RF = 63.7 MHz → only the head slice is excited

🔹 Slice Thickness Formula

Slice Thickness=BWtransγ⋅Gs\text{Slice Thickness} = \frac{\text{BW}_{trans}}{\gamma \cdot G_s}Slice Thickness=γ⋅Gs​BWtrans​​

Where:

• BWtrans = RF bandwidth

• γ (gamma) = gyromagnetic ratio

• Gs = gradient strength

🔹 Key Relationships:

• ↑ Gradient strength → ↓ Slice thickness

• ↓ RF bandwidth → ↓ Slice thickness

👉 Thin slice = better detail (high resolution) but lower SNR

4.2 Frequency Encoding (X-gradient)

After selecting a slice:

• All protons still have the same frequency & phase

• We still don’t know the left vs right position

🔹 Solution:

Apply Gx gradient during signal readout

👉 This causes:

• Left side → lower frequency

• Right side → higher frequency

🔹 Key Concept:

Using the Fourier Transform, MRI separates signals based on frequency.

4.3 Phase Encoding (Y-gradient)

We still need to determine the front vs. the back position.

🔹 Solution:

Apply the Gy gradient briefly before readout

👉 This causes:

• Protons to gain different phase shifts

• Phase depends on position

🔹 Important:

• The gradient is applied multiple times

• Each time with a different strength

👉 This builds spatial information gradually

🔹 Summary of Encoding:

20. Why Not Use Frequency Encoding in Both Directions?

It may seem logical to use frequency encoding for both X and Y directions.

❌ But this fails because:

• Frequencies from multiple voxels overlap

• Fourier Transform cannot distinguish them uniquely

👉 Result: Ambiguous image

21. K-Space (Raw Data Storage)

Before forming an image, MRI stores signals in k-space.

🔹 What is K-space?

• A data matrix of spatial frequencies

• Not an image yet

• Contains raw MRI signal data

🔹 Key Concept:

22. Gradient Echo Pulse Sequence

MRI uses pulse sequences to control the timing of: the

• RF pulses

• Gradients

• Signal acquisition

🔹 In Gradient Echo (GRE):

• Small flip angle (<90°)

• Faster imaging

• Shorter TR (Repetition Time)

🔹 Key Parameters:

23. Gradient System Performance

Important characteristics:

🔹 1. Maximum Strength

• Stronger gradients → better resolution

🔹 2. Rise Time

• Time to reach max strength

• Shorter = better

🔹 3. Slew Rate

• Speed of gradient change

Slew Rate=Max StrengthRise Time\text{Slew Rate} = \frac{\text{Max Strength}}{\text{Rise Time}}Slew Rate=Rise TimeMax Strength​

👉 Higher slew rate = faster imaging

24. MRI Image Quality Factors

9.1 Signal-to-Noise Ratio (SNR)

SNR=SignalNoise\text{SNR} = \frac{\text{Signal}}{\text{Noise}}SNR=NoiseSignal​

🔹 Increased by:

• Thick slices

• Large voxel size

• Long TR

• Short TE

• High field strength

9.2 Spatial Resolution

Ability to distinguish small structures.

🔹 Improved by:

• Thin slices

• Small FOV

• Fine matrix (e.g., 512×512)

9.3 Contrast Resolution

Ability to distinguish tissue differences.

Depends on:

• T1, T2 properties

• TR and TE

25. TR and TE Effects

TR (Repetition Time)

• Long TR → ↑ SNR, ↓ T1 contrast

• Short TR → ↓ SNR, ↑ T1 contrast

TE (Echo Time)

• Long TE → ↑ T2 contrast, ↓ SNR

• Short TE → ↑ SNR

26. Voxel, Pixel, and Matrix

• Pixel = 2D image unit

• Voxel = 3D tissue volume

• Matrix = grid of pixels

👉 Smaller voxel = better resolution but lower SNR

27. Inter-Slice Gap and Cross-Talk

• RF pulses affect neighboring slices → cross-talk

• Solution: leave a 25–50% gap

28. Field of View (FOV)

• Large FOV → ↑ SNR, ↓ resolution

• Small FOV → ↓ SNR, ↑ resolution

29. Number of Acquisitions (NEX)

• More averages → ↑ SNR

• But → ↑ scan time

30. MRI Contrast Agents

Used when natural contrast is insufficient.

Types:

🔹 T1 Agents (e.g., Gadolinium)

• Shorten T1

• Appear bright

🔹 T2 Agents (e.g., Iron oxide)

• Shorten T2

• Appear dark

31. Special MRI Applications

MRA (Magnetic Resonance Angiography)

• Visualizes blood vessels

• Flowing blood appears bright

MRCP

• Visualizes bile and pancreatic ducts

• No contrast needed

DWI (Diffusion Weighted Imaging)

• Measures water movement

• Used for stroke detection

FINAL CORE IDEA (IMPORTANT FOR EXAMS)

MRI image formation depends on:

1. Exciting protons (RF pulse)

2. Encoding location (gradients: Gz, Gx, Gy)

3. Storing data (k-space)

4. Reconstructing image (Fourier Transform)