---

Multisection (Multidetector) CT — Detailed Overview

---

1. Historical Development

"In 1998 several CT manufacturers introduced multidetector systems, which provided considerable improvement in acquisition speed, coverage, and temporal and spatial resolution." — Grainger & Allison's Diagnostic Radiology

---

2. Core Technical Principles

2.1 Detector Array and Simultaneous Section Acquisition

• A multirow detector array replaces the single detector row. Each row independently acquires a slice, so N rows → N simultaneous slices per rotation.
• Pitch = table speed ÷ gantry rotation speed. A higher pitch increases z-axis coverage per unit time but can reduce signal-to-noise ratio.
• Section collimation (beam width per detector element) and reconstruction thickness are now independently controllable — the raw data can be retrospectively reconstructed at any thickness from the archived dataset.

2.2 Isotropic Voxels

• Reformation in any arbitrary plane (axial, coronal, sagittal, oblique) with no loss of resolution
• Accurate 3D post-processing and volumetric measurements
• Computer-aided detection (CAD) of pulmonary nodules and emboli

2.3 Gantry Rotation Speed

• Single breath-hold chest coverage — feasible even in tachypnoeic patients
• Reduced motion artefact
• In paediatrics, reduced need for sedation

2.4 Novel Technologies for Further Improvement

---

3. Dose Profile and Geometric Efficiency

• Umbral (plateau) region: the entire focal spot illuminates the detector → uniform signal → used for reconstruction
• Penumbral regions (beam edges): partial focal spot illumination → non-uniform signal → discarded by post-patient collimator

• Decreases as section width increases (wider beam → penumbra is a smaller fraction)
• Decreases as the number of simultaneous sections increases (4→16 MDCT: significant waste; 64-MDCT: penumbra is minimal)

---

4. Image Reconstruction

4.1 Reconstruction Kernels (Algorithms/Filters)

4.2 Section Thickness Flexibility

• 0.6–1.25 mm thin sections — high spatial resolution, 3D post-processing, pulmonary nodule characterisation, interstitial lung disease, pulmonary embolism
• 2.5–5 mm thick sections — better contrast resolution, faster review; adequate for mediastinal masses, lung cancer staging

4.3 Iterative Reconstruction

---

5. Window Settings

• Window centre ≈ midpoint between the density of the structure of interest and surrounding tissue
• Wide windows (e.g., −600/1600 HU) → lung parenchyma
• Narrow windows (e.g., 40/400 HU) → mediastinum, soft tissue

---

6. Postprocessing Techniques

---

7. Contrast Medium Protocols

• Faster peak enhancement → need higher iodine delivery rate (faster injection rate + higher iodine concentration)
• Typical 64-slice thoracic MDCT: 60–120 mL of 320–400 mg/mL iodine at 3.5–5 mL/s, followed by 20–40 mL saline chaser
• Biphasic (dual-bolus) protocols are standard: the saline chaser dilutes contrast density in brachiocephalic veins, eliminating streak/beam-hardening artefacts and providing more homogeneous enhancement
• Triphasic protocols (contrast → contrast:saline mix → saline) used for "triple rule-out" CT (simultaneous coronary, pulmonary, and aortic evaluation)
• Bolus tracking / automated triggering (rather than fixed delays) is preferred given the narrow acquisition windows

---

8. Cardiac MDCT (ECG Gating)

• Heart rate ≤60 bpm (β-blockade preparation) — ensures R-R interval >1000 ms
• End-diastole selected for final reconstruction (period of least cardiac motion)

---

9. Dual-Energy CT (DECT)

1. Dual-source: two tubes operating simultaneously at different kVp (e.g., 80 and 140 kVp)
2. Rapid kVp switching: single tube alternates between energies
3. Dual-layer ("sandwich") detector: separate detector layers absorb different energy spectra

• Virtual unenhanced images — eliminates need for a separate pre-contrast scan → dose reduction
• Iodine maps → pulmonary blood volume maps (perfusion assessment in pulmonary embolism, comparable to scintigraphy)
• Nodule characterisation by differential iodine uptake

---

10. Dose Reduction Strategies

---

Summary

1. Simultaneous multi-section acquisition → dramatically reduced scan time
2. Isotropic voxels → unrestricted multiplanar and 3D reformatting
3. Flexible retrospective reconstruction → single acquisition, multiple slice thicknesses and kernels
4. ECG gating → non-invasive cardiac and coronary imaging
5. Dual-energy capability → material decomposition, virtual unenhanced, perfusion maps

---

---

MULTISECTION (MULTILAYER) TOMOGRAPHY — COMPREHENSIVE NOTES

---

1. DEFINITION

• Multisection (multilayer) tomography is a form of conventional X-ray tomography.
• Multiple body layers are imaged simultaneously in a single exposure.
• Each layer is recorded on a separate film within the same cassette.
• Contrast with standard tomography, which images only one layer per exposure.

---

2. BASIC TOMOGRAPHY PRINCIPLE (Foundation)

1. The X-ray tube (T) and film (F) are mechanically linked by a rigid lever arm.
2. During exposure, the tube moves in one direction while the film moves in the opposite direction — they are coupled and synchronised.
3. The lever rotates around a fixed pivot point called the fulcrum (Y).
4. The plane passing through the fulcrum remains in a fixed relationship to the film throughout movement → it appears sharp.
5. All structures above and below the fulcrum plane move relative to the film → their images are blurred (unsharpened) across the film.
6. The degree of blurring increases with distance from the fulcrum plane and with the amplitude of tube-film travel.

---

3. MULTISECTION EXTENSION OF THE PRINCIPLE

3.1 Multiple Films, Multiple Focal Planes

1. Several films are placed one above the other at different vertical levels within a single cassette.
2. Because each film sits at a different height, each film has a different geometric relationship to the lever system.
3. This means each film effectively has its own fulcrum level — i.e., a different anatomical plane stays sharp on each film.

1. A single tube-film sweep across the patient exposes all films simultaneously.
2. Result: N films = N tomographic slices from one exposure.

3.2 The Governing Principle Statement

"The layer recorded on a film is at the level of the fulcrum of the lever system, provided the film position coincides with the film-moving point on the lever."

• Films above the film-moving point on the lever → record higher planes
• Films below the film-moving point → record lower planes

3.3 How Multiple Fulcra Are Created

• Multiple fulcrum levels are created geometrically, not by physically shifting the pivot.
• Each film's vertical displacement from the nominal film position introduces a geometric offset → this shifts the effective focal plane for that film.
• The interval between focal planes = determined by the separator thickness between films and the geometric magnification factor of the system.

---

4. TECHNIQUE — STEP BY STEP

Step 1: Film Arrangement

• Films stacked vertically one above the other inside the multisection cassette.
• Separated by spacers (separators) of fixed, known thickness.
• Each film is accompanied by its pair of intensifying screens.

Step 2: Cassette Placement

• Cassette placed in the bucky tray (or special support tray for deep cassettes).
• Correct orientation: the side facing the X-ray tube must be identified and placed correctly — incorrect orientation causes radiographic chaos.

Step 3: Patient Positioning

• Patient positioned as for standard tomography.
• Fulcrum height set to the desired mid-plane of anatomy.

Step 4: Single Exposure

• One exposure is made.
• Tube and film assembly move simultaneously during exposure.
• All films exposed at exactly the same instant → same respiratory phase, same cardiac phase, same contrast bolus phase.

Step 5: Image Retrieval

• Films removed from cassette and processed individually.
• Each yields a tomographic slice of a different anatomical level.

---

5. ADVANTAGES

---

6. DISADVANTAGE

• Sequential tomography (one slice per exposure) produces better spatial detail (higher resolution) than multisection tomography.
• The multisection system has inherently lower image sharpness per slice because:
  • Each film sits at a slightly non-optimal position relative to the lever geometry
  • X-ray beam attenuation through multiple screen-film layers degrades lower films
• Rule: Multisection = faster, less detail. Sequential = slower, more detail.

---

7. EQUIPMENT

7A. Multisection Cassette

---

7B. Intensifying Screens and the Unequal Density Problem

Book Arrangement

• Each film is sandwiched between two intensifying screens.
• The screen-film "sandwiches" are assembled together with separators, like a book (screens = covers, film = page, separator = gap between pages).
• This arrangement:
  • Prevents incorrect film placement
  • Ensures correct film-screen alignment
  • Simplifies handling and loading

The Core Problem: Progressive X-ray Attenuation

X-ray beam enters from above
        ↓
Film 1 (top)     ← receives MOST radiation
        ↓  [attenuated]
Film 2            ← receives less
        ↓  [further attenuated]
Film 3            ← receives even less
        ↓  [further attenuated]
Film 4 (bottom)  ← receives LEAST radiation

• As the beam passes through each screen-film layer, it is progressively absorbed (attenuated).
• Without correction, lower films would be underexposed → insufficient density → non-diagnostic images.

Solution: Graded Intensifying Screen Speed

• Different-speed intensifying screens are used for each level:

• Fast screens at the bottom compensate for lower beam intensity → all films receive equivalent effective exposure → uniform radiographic density across all films.

"The last intensifying screens must be greater in speed than those nearer to the X-ray tube."

---

7C. Separators

• Separator thickness directly determines the interval between focal planes.
• Thinner separators → planes are closer together (higher z-resolution, more slices in same anatomical region).
• Thicker separators → planes are further apart (wider anatomical coverage per cassette depth).

---

8. EXPOSURE FACTORS

8.1 Why the Multisection System Is Photographically Slower

1. The X-ray beam must penetrate multiple screen-film layers — significant total attenuation.
2. Energy is absorbed by each screen layer sequentially.
3. The combination of multiple screens, films, and separators creates a more absorbing assembly.

8.2 Exposure Adjustments Required

• These are approximate ranges — no universal fixed formula exists.
• Exact values depend on:
  • Number of films in the cassette
  • Screen-film combination used
  • Separator material and thickness
  • Specific cassette manufacturer design

8.3 How to Determine Correct Exposure

1. Trial and error — expose test films, assess density, adjust until uniform density achieved across all layers.
2. Step wedge experiments — expose a calibrated aluminium step wedge under the cassette; analyse the density gradient across films to calculate required exposure adjustments.

---

9. PRACTICAL ERRORS AND PRECAUTIONS

---

10. MULTISECTION vs. SEQUENTIAL TOMOGRAPHY — COMPARISON

---

11. CLINICAL APPLICATIONS

• Pulmonary tomography — assess lung lesions at multiple levels simultaneously
• Angiographic tomography — capture contrast filling at multiple planes in a single bolus pass
• Renal tomography (nephrotomography) — assess renal parenchyma and collecting system
• Skeletal tomography — especially for fractures where multiple planes aid characterisation
• Any situation where the physiological state must be identical across all images

---

12. QUICK REVISION SUMMARY

---

SCAN CONVERTER, IMAGE FORMATION, STORAGE & DISPLAY IN ULTRASOUND

MSc Radiology — Comprehensive Notes

---

SECTION 1: BACKGROUND — THE FUNDAMENTAL PROBLEM

• Ultrasound data is acquired in polar coordinates (radial scan lines radiating from the transducer face — like spokes of a wheel).
• A standard video monitor displays data in rectangular (Cartesian) coordinates — rows and columns of pixels arranged in a square matrix.
• These two coordinate systems are geometrically incompatible.
• The scan converter is the hardware/software system that bridges this mismatch — converting polar ultrasound data into a rectangular pixel matrix for display.

---

SECTION 2: SCAN CONVERTER — DEFINITION AND ROLE

1. Receives the processed echo signal data (in polar/scan-line format)
2. Converts it into a standard rectangular video matrix
3. Stores the resulting image as a digital matrix
4. Outputs it to the display monitor as a real-time video image

---

SECTION 3: ANALOG SCAN CONVERTER vs. DIGITAL SCAN CONVERTER (SCAN CONVERTER MEMORY)

---

3A. ANALOG SCAN CONVERTER

Definition

Construction

• A storage tube — a glass vacuum tube with an electron gun and a target screen coated with a charge-retaining material (e.g., silicon or mica)
• The electron gun writes image data as a pattern of electrical charges directly onto the storage screen
• A separate read beam scans the screen to output the stored charge pattern to the display

How It Worked (Mechanism)

1. Echo signals from each scan line were converted into a voltage-modulated electron beam
2. The beam struck the storage screen in a pattern corresponding to the ultrasound scan geometry
3. Positive charges accumulated at points struck by stronger echoes → encoded spatial and amplitude information as a charge map

1. A low-current read beam continuously scanned the storage screen
2. The charge pattern was read as a varying current → converted to a video signal → sent to monitor

Characteristics

Key Limitation

• Charge instability → image brightness drifts, fades, produces flicker
• Limited grey scale → reduced diagnostic information
• Cannot re-process stored image — what was written is what you get
• Geometric distortion — charge distribution was non-uniform across the storage screen

---

3B. DIGITAL SCAN CONVERTER (SCAN CONVERTER MEMORY)

Definition

Structure

• A rectangular matrix of memory locations (pixels)
• Typical matrix size: 512 × 512 or 1024 × 1024 elements
• Each memory location (pixel) stores a digital value representing echo amplitude
• Standard grey scale: 256 levels (8-bit: 2⁸ = 256) — far superior to analogue

Core Components

---

3C. KEY DIFFERENCES — ANALOG vs. DIGITAL SCAN CONVERTER

---

SECTION 4: IMAGE FORMATION IN ULTRASOUND — COMPLETE STEP-BY-STEP

---

STEP 1 — ULTRASOUND PULSE GENERATION

1. The pulser/transmitter sends a short electrical voltage pulse to the transducer.
2. The piezoelectric crystal in the transducer deforms in response → generates a brief ultrasound pulse (~1–3 cycles).
3. Typical frequencies used: 2.5 MHz (adult TTE deep structures) → 7.5–20 MHz (superficial/intravascular).
4. The pulse travels through tissue as a longitudinal pressure wave (compressions and rarefactions).
5. Speed of sound in soft tissue ≈ 1540 m/s (this constant is used for all depth calculations).

• Pulse Repetition Frequency (PRF) = number of pulses emitted per second
• Duty factor = percentage of time ultrasound is actively being transmitted (typically <1% for 2D imaging)

---

STEP 2 — BEAM-TISSUE INTERACTION

1. Reflection — part of the energy returns to the transducer (specular reflection from smooth surfaces)
2. Scattering — part is scattered in all directions (from rough surfaces and small structures; important for parenchymal imaging)
3. Transmission — attenuated beam continues forward
4. Refraction — beam direction changes at oblique interfaces

• Echo amplitude decreases with depth: attenuation ~0.5 dB/cm/MHz in soft tissue
• Lower frequencies penetrate deeper; higher frequencies give better resolution

---

STEP 3 — ECHO RECEPTION AND SIGNAL DETECTION

1. The transducer switches to receive mode after transmitting (the crystal is a transmitter and receiver simultaneously).
2. Returning echoes strike the piezoelectric crystal → generate a tiny electrical signal.
3. Signal amplitude is proportional to:
   • Incident angle between beam and interface (perpendicular → maximum reflection)
   • Acoustic impedance mismatch at the interface
4. Signal timing encodes depth: depth = (speed of sound × time of flight) ÷ 2

---

STEP 4 — SIGNAL PROCESSING CHAIN

a) Pre-Amplification

• The received echo signal is extremely weak (microvolt range)
• A pre-amplifier increases signal strength while minimising added noise

b) Time-Gain Compensation (TGC)

• Echoes from deeper structures are weaker due to attenuation
• TGC applies progressively increasing gain with depth
• Near-field gain set low (strong echoes); far-field gain set high (weak echoes)
• Goal: uniform brightness across all depths for similar tissue types
• Operator-adjustable via TGC sliders on the machine

c) Filtering

• Band-pass filter removes frequencies outside the transducer bandwidth
• Reduces electronic noise and artefacts

d) Rectification and Envelope Detection

• The received radio-frequency (RF) signal is a rapidly oscillating waveform
• Rectification converts it to a positive-only signal
• Envelope detection extracts the smooth amplitude profile (the outline/envelope of the oscillating signal)
• This amplitude profile represents echo strength at each depth point

e) Compression (Dynamic Range Reduction)

• The amplitude range of all echoes spans 100+ dB — far more than a monitor can display (typically 30–40 dB)
• Logarithmic compression maps the wide dynamic range of echoes to the available grey-scale range
• Adjustable by the operator as "dynamic range" or "compression" control
• Higher compression → more grey levels used → softer, more textured image
• Lower compression → fewer grey levels → higher contrast, starker image

---

STEP 5 — ANALOG-TO-DIGITAL CONVERSION (ADC)

1. The processed analogue echo amplitude signal along each scan line is sampled at regular intervals by the ADC.
2. Each sample is converted to a digital number (8-bit → 0–255; representing echo amplitude at that depth point).
3. Sampling must satisfy the Nyquist criterion: sampling rate ≥ 2× highest signal frequency.
4. This step generates a string of digital values for each scan line, representing echo amplitude vs. depth.

---

STEP 6 — SCAN CONVERSION (THE CRITICAL STEP)

The Problem

• Each scan line is a radius emanating from the transducer face at a defined angle
• Scan lines are in polar format: (r, θ) where r = depth, θ = beam angle
• The display monitor uses Cartesian format: (x, y) pixel addresses

The Solution — Scan Conversion Algorithm

1. Calculate the corresponding (x, y) pixel address:
   • x = r × sin θ
   • y = r × cos θ
2. Write the digital amplitude value to that pixel address in the frame buffer (memory matrix)

Interpolation

• Scan lines are not evenly distributed across the pixel matrix — gaps exist between scan lines at greater depths (sector format diverges with depth)
• Interpolation algorithms fill in pixels between scan lines using:
  • Nearest neighbour — quick but produces blocky image
  • Bilinear interpolation — weighted average from adjacent scan lines → smoother, more accurate

---

STEP 7 — FRAME BUFFER STORAGE (IMAGE MEMORY)

1. The complete rectangular pixel matrix for one image frame is stored in the frame buffer — a block of RAM in the digital scan converter.
2. Typical frame: 512 × 512 pixels × 8 bits = 2 MB per frame
3. The frame buffer is continuously written as new data arrives → real-time imaging
4. A cine loop buffer stores the last several seconds (typically 256–1024 frames) of imaging in RAM → allows review of recent frames
5. When image freeze is activated: writing to the frame buffer stops; the last complete frame remains stored; it is read out continuously to the display

---

STEP 8 — POST-PROCESSING (DIGITAL IMAGE MANIPULATION)

---

STEP 9 — DISPLAY (DIGITAL-TO-ANALOG CONVERSION AND MONITOR OUTPUT)

1. The digital-to-analog converter (DAC) reads pixel values from the frame buffer and converts them back to an analogue voltage signal.
2. The analogue video signal drives the display monitor (historically CRT; now LCD/LED flat panel).
3. The monitor maps signal voltage → screen brightness at each pixel location:
   • 0 (digital) → black (no echo / anechoic)
   • 255 (digital) → white (maximum reflection)
   • Intermediate values → 254 shades of grey
4. The display refreshes at standard video frame rates (25 Hz PAL / 30 Hz NTSC or higher) — updates continuously with new incoming data in real time.
5. Additional overlay information is added at this stage: depth markers, callipers, patient data, ECG trace, Doppler spectral display.

---

SECTION 5: IMAGE STORAGE — COMPLETE PATHWAY

5A. Short-Term (Real-Time) Storage

• Frame buffer: holds the current frame being displayed
• Cine loop RAM buffer: holds 256–1024 frames (last ~10–30 seconds of scanning)
• Volatile — lost when machine is switched off unless archived

5B. Long-Term Storage

5C. DICOM in Ultrasound

• Ultrasound stores both still frames and cine loops as DICOM objects
• DICOM header contains: patient ID, date/time, transducer frequency, frame rate, depth setting, institution
• Enables structured reporting and measurement annotation to be stored with images

---

SECTION 6: THE COMPLETE CHAIN — SUMMARY DIAGRAM

TRANSDUCER
    │
    ▼ Electrical pulse → Ultrasound pulse
TISSUE INTERACTION
    │
    ▼ Reflected echo → Electrical signal
PRE-AMPLIFIER
    │
    ▼ Amplified RF signal
TIME-GAIN COMPENSATION (TGC)
    │
    ▼ Depth-corrected signal
FILTERING → RECTIFICATION → ENVELOPE DETECTION
    │
    ▼ Amplitude profile per scan line
COMPRESSION (Dynamic range reduction)
    │
    ▼ Compressed analogue signal
ADC (Analog-to-Digital Converter)
    │
    ▼ Digital values per scan line (polar coordinates)
DIGITAL SCAN CONVERTER
    │
    ├─ Scan conversion (polar → rectangular)
    ├─ Interpolation (fill matrix gaps)
    ▼
FRAME BUFFER (Digital Image Matrix: 512×512 or 1024×1024 pixels)
    │
    ├─ Post-processing (grey map, zoom, averaging, measurements)
    ▼
DAC (Digital-to-Analog Converter)
    │
    ▼
DISPLAY MONITOR (Grey-scale real-time image)
    │
    ▼
ARCHIVING (PACS / DICOM / Cine loop)

---

SECTION 7: GREY SCALE AND BIT DEPTH

---

SECTION 8: FRAME RATE AND ITS TRADE-OFFS

Frame rate = PRF ÷ (number of scan lines per frame)

PRF(max) = 1540 m/s ÷ (2 × imaging depth in metres)

• Greater depth → lower PRF → lower frame rate (or fewer scan lines)
• More scan lines → better lateral resolution → but lower frame rate
• Narrower sector angle → fewer scan lines for same depth → higher frame rate

---

SECTION 9: QUICK REVISION TABLE

---