Ct physics in detail knowledge

Reading File

Finding Sources

Reading File

Finding Sources

Reading File

Finding Sources

Reading File

I now have comprehensive material from multiple textbooks. Let me compile the full detailed response on CT physics.

CT Physics - Detailed Knowledge

1. History and Overview

CT (Computed Tomography) was developed in the 1970s by British scientist Godfrey Hounsfield at EMI Corporation. He applied the physics of rotational tomography so that the x-ray tube and detector gantry rotate about a single axis in the body. X-ray detectors replaced film, taking continuous measurements during gantry rotation; a computer then reconstructed a digital image of an axial "slice" displayed on a CRT screen.

It became commercially available in 1973, initially as a head scanner. The term "tomography" comes from the Greek meaning "to slice."

Brogdon's Forensic Radiology, p. 10534-10554
Bradley and Daroff's Neurology in Clinical Practice, p. 3817

2. Basic X-ray Physics Relevant to CT

X-ray Nature

X-rays are electromagnetic waves with wavelengths of 10-0.01 nm, describable as high-energy photons with energies between 124 and 124,000 electron volts. "Hard" x-rays (higher energy end) are used in diagnostic imaging because they penetrate tissue yet are differentially absorbed.

X-rays are a form of ionizing radiation - health risks, while minimal, must always be accounted for.

X-ray Interactions with Tissue

Two main interactions:

Photoelectric effect - complete absorption of a photon by an inner shell electron; depends strongly on atomic number (Z³) and photon energy. This is why iodine and bone attenuate heavily.
Compton scattering - the incoming photon ejects an outer electron and scatters in a different direction, contributing to image noise and radiation dose.

3. Basic CT Scanner Components

X-ray Tube

Generates x-rays toward the patient
Key parameters: tube voltage (kVp) determines x-ray beam energy; tube current (mA) determines photon quantity

Collimator

A rectangular opening in a lead shield that shapes the x-ray beam to define slice thickness and limit scatter radiation

Detectors

Opposite the x-ray source; measure transmitted x-rays
Fourth-generation CT: detectors in a fixed 360° ring; x-ray source rotates around the patient
Modern multidetector CT (MDCT): multiple detector rows enable simultaneous multi-slice acquisition

Gantry

The rotating assembly housing the x-ray tube and detectors
Patient lies on a motorized table that moves through the gantry aperture
Bradley and Daroff's Neurology, p. 3819

4. CT Generations

Generation	Geometry
1st	Single pencil beam, single detector, translate-rotate
2nd	Fan beam, small detector array, translate-rotate
3rd	Rotating fan beam + rotating detector arc (rotate-rotate)
4th	Rotating fan beam + stationary 360° detector ring
5th (electron beam)	Electron beam deflected onto tungsten anode rings; used in cardiac CT

5. Helical/Spiral CT

Introduced over 20 years ago, helical (spiral) CT combines continuous gantry rotation with continuous table movement through the gantry. The x-ray path traces a helix around the patient.

Advantages:

Rapid large-volume acquisition (20-60 seconds)
Patients can hold breath, reducing motion artifact
Optimal contrast bolus timing
Allows multiplanar reformatting (MPR) and 3D reconstruction

Pitch

Pitch = table feed per gantry rotation / total beam collimation width

Pitch < 1: overlapping acquisition (more radiation, better image quality)
Pitch = 1: contiguous acquisition
Pitch > 1: gapped acquisition (less radiation, slightly lower quality)

Important: in most modern scanners, radiation dose is inversely proportional to pitch - higher pitch = lower dose.

Bradley and Daroff's Neurology, p. 3831

6. Multidetector CT (MDCT)

MDCT uses multiple parallel rows of detectors enabling simultaneous acquisition of multiple slices per gantry rotation.

4-slice MDCT → 4 simultaneous slices
64-slice MDCT → 64 simultaneous slices
256/320/640-slice scanners now exist for cardiac and whole-organ coverage

Geometric Efficiency

With 4-16 slice MDCT, penumbral dose wastage occurs at the beam edges - photons in the penumbral region do not contribute useful image data.

With 64-channel MDCT, the incident beam width remains constant over both narrow and wide collimation - geometric efficiency is high, with minimal dose penalty. Gaps between detector elements also waste photons (scatter).

Grainger & Allison's Diagnostic Radiology, p. 256-258

7. Image Formation and Reconstruction

Attenuation

As the x-ray beam passes through tissues, it is attenuated (absorbed/scattered) to varying degrees based on:

Atomic composition (Z number)
Physical density
Tissue thickness

The computer collects detector readings at multiple angles (a full 360° sweep per slice), then uses back-projection algorithms (filtered back projection = FBP, or iterative reconstruction) to calculate x-ray attenuation for each individual tissue volume element (voxel).

Filtered Back Projection (FBP)

The classical reconstruction method. Fast but produces streak artifacts at low mA (high noise).

Iterative Reconstruction

Modern alternative to FBP. Starts with an estimate, compares to actual projections, corrects errors iteratively. Allows significant dose reduction (30-80%) with maintained or improved image quality. Subtypes include:

Adaptive statistical iterative reconstruction (ASIR)
Model-based iterative reconstruction (MBIR) - most dose-efficient

8. Hounsfield Units (HU)

The attenuation of each voxel is expressed as a Hounsfield unit (HU) on an arbitrary linear scale:

Formula:

CT number (HU) = [(μ_tissue - μ_water) / μ_water] × 1000

Where μ = linear attenuation coefficient.

Standard HU Reference Values:

Tissue/Material	Hounsfield Units
Dense air	-1024 (black)
Fat	-50 to -80
Water	0
Soft tissue/muscle	+20 to +80
Fresh blood	~+80
Brain parenchyma	+25 to +45
Bone (cortical)	+400 to +1000+
Cranial bone	up to +2000
Dense metal	up to +3071 (white)
Adrenal adenoma (lipid-rich)	<10 HU

Tissues with higher HU appear whiter (hyperdense)
Tissues with lower HU appear darker (hypodense)
Murray & Nadel's Respiratory Medicine, p. 1332
Bradley and Daroff's Neurology, p. 3822-3823
Scott-Brown's Otorhinolaryngology, p. 1918

9. Pixels, Voxels, and Spatial Resolution

Pixel

The smallest 2D picture element in the reconstructed image.

Current highest resolution: 512 × 512 pixels per image (some systems: 1024 × 1024)
In a 30 × 30 cm field of view (FOV), 512 × 512 gives high definition; larger FOV = lower resolution per cm

Voxel

A 3D volumetric element - essentially a pixel with depth (slice thickness).

Can be isotropic (cube-shaped) or anisotropic (cuboid)
Isotropic voxels (e.g., 0.5 mm³) allow high-quality multiplanar reformats in any plane without degradation

Partial Volume Averaging (Effect)

When a structure (e.g., bone = 1000 HU) fills only part of a voxel, the computed HU is an average of all tissues within that voxel - often underestimating true density. This reduces edge sharpness and is a significant limitation in delineating structural borders.

Scott-Brown's Otorhinolaryngology, p. 1922-1939

10. Windowing

The human eye cannot distinguish the full ~4000-level HU range. Windowing (also called "window level/width") narrows the displayed HU range to optimize contrast for specific tissues:

Window Level (WL) = center HU value of the displayed range
Window Width (WW) = range of HU values displayed (all outside appear pure black or white)

Clinical Window Settings:

Window	Level (HU)	Width (HU)	Use
Soft tissue	+40	+400	Abdomen, mediastinum
Lung	-600	+1500	Lung parenchyma, airways
Bone	+400	+2000	Fractures, cortical detail
Brain	+35	+80	Intracranial pathology
Subdural	+75	+200	Hemorrhage
Liver	+60	+150	Hepatic lesions

Bradley and Daroff's Neurology, p. 3825

11. Contrast Agents in CT

Iodinated Contrast

CT contrast agents contain iodine in an injectable water-soluble form. Iodine is a heavy atom whose inner electron shell absorbs x-rays via photoelectric capture - even small amounts block x-rays, appearing hyperdense.

Used for:

CT angiography (CTA) - vascular mapping
Contrast-enhanced CT - tumor detection, BBB disruption
CT myelography - intrathecal
CT perfusion - cerebral/organ perfusion assessment

Phases of Enhancement:

Arterial phase (~25-30 sec post-injection) - aorta, hepatic artery opacification
Portal venous phase (~60-70 sec) - portal vein, liver parenchyma
Delayed/equilibrium phase (~3-5 min) - renal cortex, biliary/urinary)

12. Radiation Dose in CT

Dose Metrics

CTDI (CT Dose Index) - measure of dose per rotation, expressed in mGy
DLP (Dose Length Product) = CTDI × scan length (mGy·cm)
Effective dose (mSv) - accounts for tissue radiosensitivity; derived from DLP × conversion factor

Parameters Directly Affecting Dose (Grainger & Allison):

Gantry geometry
Rotation time
Tube current (mA) and voltage (kVp)
Acquisition modes
Z-axis coverage
Pitch (higher pitch = lower dose)
Section collimation and overlap

Dose Reduction Strategies:

Strategy	Mechanism
Automatic exposure control (AEC)	Modulates mA to patient anatomy in real-time
Weight/size-based modulation	Reduces mA for smaller patients
Reduced tube current (mA)	50% mA reduction = 50% dose reduction
Reduced tube potential (kVp)	80-100 kVp instead of 120 kVp in thin patients
Higher pitch	More table advance = shorter scan time = less dose
Iterative reconstruction	Allows lower mA with acceptable image quality
Beam-shaping filters (bowtie)	Reduces dose to peripheral, lower-attenuation tissue
Prospective ECG gating	Limits x-ray exposure to specific cardiac phases
Restrict scan length	Only cover area of interest
Patient shielding	Thyroid, eye, breast shields

Reducing tube current by 50% halves the effective dose. A 120 → 100 kVp reduction yields large dose savings with minimal quality loss in thin patients.

Grainger & Allison's Diagnostic Radiology, p. 252-264

13. Image Quality Parameters

Spatial Resolution

Ability to distinguish two closely spaced objects
Determined by: detector size, focal spot size, reconstruction algorithm ("kernel"), pixel matrix, FOV
High-frequency (sharp) kernels: better edge detail but more noise - used for bone
Low-frequency (smooth) kernels: less noise but blurred edges - used for soft tissue

Contrast Resolution

Ability to distinguish tissues with small HU differences
Improves with higher mA (lower noise), larger voxels

Temporal Resolution

How fast a single image can be acquired
Critical in cardiac CT; improved with faster gantry rotation and multi-segment reconstruction

Noise

Random fluctuation in HU values; appears as "graininess"
Increases with: lower mA, smaller voxels, larger patient size
Decreases with: higher mA, larger voxels, iterative reconstruction

Artifacts:

Artifact	Cause
Beam hardening	Polychromatic beam - lower-energy photons absorbed preferentially; causes dark streaks between dense objects (e.g., posterior fossa "Hounsfield bar")
Metal artifact	High-density objects cause severe streaking
Motion artifact	Patient or cardiac/respiratory motion
Partial volume	Averaging of different densities in one voxel
Ring artifact	Faulty detector element
Stair-step artifact	Wide slice thickness in oblique/curved structures

14. Postprocessing Techniques

Technique	Description	Clinical Use
MPR (Multiplanar Reconstruction)	2D reformats in coronal, sagittal, oblique planes	Vascular, spinal, urinary tract
MIP (Maximum Intensity Projection)	Ray casting - only highest HU per ray displayed	Vascular imaging (CTA), dense nodules
MinIP (Minimum Intensity Projection)	Only lowest HU per ray	Emphysema, airways, bronchiectasis
SSD (Shaded Surface Display)	Threshold-based interface rendering	Airway abnormalities
Volume Rendering (VR)	Full HU mapping to opacity/color	3D vascular, bone, complex anatomy
Virtual Bronchoscopy	Endoscopic simulation from CT data	Airway lesions distal to obstruction
Dual-energy CT	Two simultaneous kVp acquisitions	Material decomposition, iodine maps, virtual non-contrast, uric acid stone characterization
Computer-aided detection (CAD)	Pattern recognition algorithms	Pulmonary nodule detection and measurement

Grainger & Allison's Diagnostic Radiology, p. 202-205

15. Dual-Energy CT (DECT)

Acquires data at two different x-ray energies (e.g., 80 kVp and 140 kVp) simultaneously. Because different materials (iodine, calcium, uric acid) have characteristic attenuation differences between energies, DECT can:

Generate virtual non-contrast images from contrast-enhanced studies
Characterize renal stones (uric acid vs. calcium)
Produce iodine distribution maps as perfusion surrogates
Reduce beam-hardening artifacts
Improve lesion characterization (e.g., adrenal adenoma vs. metastasis)

16. Cardiac CT Physics

Requires high temporal resolution (ideally <100 ms) to freeze cardiac motion at ~70 bpm
Prospective ECG triggering: x-ray on only during a defined cardiac phase (end-diastole); lowest radiation dose
Retrospective ECG gating: continuous x-ray acquisition; allows reconstruction at any cardiac phase; higher dose
High-pitch spiral CT (dual-source): can image entire heart in one heartbeat at very low dose
Calcium scoring uses Agatston score to quantify coronary artery calcium; unenhanced scan
CT coronary angiography (CCTA) requires HR <65 bpm, iodinated IV contrast, beta-blocker pre-medication
Grainger & Allison's CT Dose Considerations, p. 253-260

Summary Table: CT Physics at a Glance

Parameter	What it Affects	Clinical Relevance
kVp	Beam energy, penetration, contrast	Lower kVp → more iodine contrast but more noise
mAs	Photon quantity, noise, dose	Lower mAs → more noise; 50% reduction = 50% dose
Pitch	Scan speed, dose	Higher pitch = faster, lower dose
Slice thickness	Spatial/contrast resolution, noise	Thinner = better resolution, more noise
FOV	Pixel size, spatial resolution	Smaller FOV = smaller pixels = better resolution
Kernel	Spatial vs. contrast resolution trade-off	Bone (sharp) vs. soft tissue (smooth)
Reconstruction	Noise vs. speed	Iterative < FBP for dose
Gantry speed	Temporal resolution	Faster = better cardiac imaging

Sources:

Bradley and Daroff's Neurology in Clinical Practice - CT physics, HU values, windowing, helical CT
Grainger & Allison's Diagnostic Radiology - MDCT dose, pitch, iterative reconstruction, postprocessing
Brogdon's Forensic Radiology - CT history, Hounsfield development
Murray & Nadel's Textbook of Respiratory Medicine - CT numbers and lung applications
Scott-Brown's Otorhinolaryngology - Pixels, voxels, partial volume averaging

No answer yet

Here is the complete, detailed answer on CT Physics:

CT Physics — Detailed Knowledge

1. History and Development

CT was developed in the 1970s by British scientist Godfrey Hounsfield at EMI Corporation. He applied the physics of rotational tomography so that both the x-ray tube and detector gantry rotate about a single axis around the body. Detectors replaced film and took continuous measurements during rotation; a computer reconstructed these as a digital axial "slice."

Became commercially available in 1973, initially as a head-only scanner
The term tomography derives from the Greek "to slice"
First allowed radiologists to examine the brain without the skull obscuring soft tissue

2. X-ray Physics Fundamentals

Nature of X-rays

Electromagnetic waves, wavelengths 10 to 0.01 nm
Described as high-energy photons: 124 to 124,000 electron volts (eV)
"Hard" x-rays (higher energy) are used in CT - they penetrate tissue yet are differentially absorbed
X-rays are ionizing radiation - health risks must always be accounted for

X-ray Interactions with Matter (Two Key Processes)

1. Photoelectric Effect

The x-ray photon is completely absorbed by an inner-shell electron
The electron is ejected; the atom emits a characteristic x-ray
Dominant at lower photon energies and in high atomic number (Z) materials
Proportional to Z³ / E³ (Z = atomic number, E = photon energy)
Explains why bone, iodine, and barium strongly attenuate x-rays
Produces excellent image contrast but deposits all energy in tissue (more dose)

2. Compton Scattering

The x-ray photon ejects an outer-shell electron and scatters in a new direction with reduced energy
Dominant at diagnostic x-ray energies in soft tissue
Does not depend on Z, only on electron density
Scattered photons contribute to image noise and radiation dose without useful information
A major source of image degradation in CT

3. Basic CT Scanner Components

Component	Function
X-ray tube	Generates x-rays using a rotating tungsten anode
Collimator (pre-patient)	Shapes the beam into a fan; limits slice thickness and scatter
Bow-tie filter	Compensates for patient's elliptical cross-section; reduces peripheral dose
Patient table	Motorized movement through the gantry aperture
Detectors	Solid-state (scintillator + photodiode) or gas ionization; measure transmitted x-rays
Gantry	Rotating frame housing tube and detectors; rotates 360° in 0.2-0.5 seconds
Data acquisition system (DAS)	Converts detector signals to digital data
Computer/reconstruction unit	Applies reconstruction algorithms to produce images

4. CT Scanner Generations

Generation	X-ray Beam	Detector	Motion	Era
1st	Single pencil beam	Single detector	Translate + rotate	Early 1970s
2nd	Narrow fan beam	Small array (3-30)	Translate + rotate	Mid 1970s
3rd	Wide fan beam	Curved detector arc	Rotate-rotate (both rotate together)	Late 1970s - present
4th	Wide fan beam	Fixed 360° ring	Tube rotates only	Late 1970s
5th (Electron Beam CT)	Electron beam deflected to tungsten anode rings	Fixed rings	No moving parts	1980s; cardiac CT

Modern clinical scanners are 3rd generation (rotate-rotate).

5. Image Acquisition - Scan Modes

Conventional (Axial/Sequential) CT

Gantry rotates while table is stationary for each slice
Table increments between rotations
Good for high-resolution, minimal motion situations
Slower; used for brain, inner ear, HR-CT of lungs

Helical/Spiral CT

Gantry rotates continuously while table moves continuously
The x-ray path traces a helix around the patient
Introduced ~early 1990s; now standard
Can scan entire chest or abdomen in 20-60 seconds

Advantages of helical CT:

Large volumes imaged rapidly
Patients can hold breath - minimizes respiratory motion artifacts
Optimized contrast bolus timing
Allows multiplanar reconstruction (MPR) and 3D imaging
Reduced contrast volume requirements

Pitch (Helical CT)

Pitch = Table feed per 360° gantry rotation ÷ Total beam collimation width

Pitch Value	Effect
< 1	Overlapping acquisition; higher dose, slightly better quality
= 1	Contiguous slices; standard
> 1	Gapped acquisition; lower dose, slightly lower quality

Higher pitch = faster scan, lower radiation dose
In most MDCT systems, mAs is automatically adjusted to maintain dose as pitch changes

6. Multidetector CT (MDCT)

MDCT uses multiple parallel rows of detectors allowing simultaneous acquisition of multiple slices per gantry rotation.

4-slice MDCT: 4 simultaneous slices
16-slice: 16 simultaneous slices
64-slice: 64 simultaneous slices in a fraction of a second
256/320-slice: entire organ (heart, brain) in one rotation

Benefits of MDCT:

Faster scan times
Thinner slice acquisition (sub-millimeter isotropic voxels)
Better 3D and MPR image quality
Reduced motion artifacts
Cardiac and perfusion imaging feasibility

Geometric Efficiency in MDCT:

Penumbral radiation waste occurs at beam edges in 4-16 slice systems
With 64-channel MDCT, beam width is constant across collimation settings → high geometric efficiency, minimal dose penalty
Gaps between detector elements also waste photons

7. Image Reconstruction

Data Collection

The x-ray beam sweeps 360° for each slice. The detector records transmitted x-ray intensity at each angle (called a projection). Each projection represents the sum of attenuation along the ray path.

Radon Transform / Sinogram

The collection of all projections forms a sinogram - a mathematical representation of the scanned object. Reconstruction reverses this process.

Filtered Back Projection (FBP)

Classical reconstruction method
Each projection is "smeared back" across the image matrix at the angle it was acquired
A ramp/high-pass filter applied in Fourier domain to correct the inherent blurring from back-projection
Fast and reliable
Produces streak artifacts at low mA (quantum noise amplification)

Iterative Reconstruction (IR)

Modern alternative to FBP
Starts with an initial image estimate; compares forward projection to actual measured projections; corrects discrepancies iteratively
Converges on the most statistically likely image
Allows 30-80% dose reduction while maintaining image quality
Types include:
- ASIR (Adaptive Statistical Iterative Reconstruction) - GE
- IRIS/SAFIRE - Siemens
- MBIR (Model-Based Iterative Reconstruction) - most dose-efficient, computationally intensive

8. Hounsfield Units (HU)

The attenuation of each voxel is expressed on the Hounsfield scale (also called CT numbers):

Formula:

HU = [(μ_tissue - μ_water) / μ_water] × 1000

Where μ = linear attenuation coefficient of the material

Standard HU Values:

Material / Tissue	HU Range
Dense air	-1024 (black)
Lung parenchyma	-700 to -900
Fat	-50 to -80
Water	0
Soft tissue / muscle	+20 to +80
Brain white matter	+20 to +30
Brain grey matter	+35 to +45
Fresh blood (normal Hct)	~+80
Subacute blood (clotted)	+50 to +80
Iodinated contrast	+100 to +400+
Bone (trabecular)	+100 to +400
Bone (cortical)	+400 to +1000+
Cranial bone	up to +2000
Dense metallic objects	up to +3071 (white)
Adrenal adenoma (lipid-rich)	< +10
Uric acid stone	lower HU than calcium stones

Hyperdense = higher HU, appears bright/white
Hypodense = lower HU, appears dark/black
Values are relative - always compared to surrounding structures

9. Pixels, Voxels, and Spatial Resolution

Pixel (Picture Element)

Smallest 2D unit in the reconstructed image
Current standard: 512 × 512 matrix (some: 1024 × 1024)
Pixel size = FOV / matrix size (e.g., 30 cm FOV ÷ 512 = 0.59 mm pixels)

Voxel (Volumetric Element)

Pixel with depth (= slice thickness)
Can be isotropic (cube, equal dimensions in all axes) or anisotropic (cuboid)
Isotropic voxels (e.g., 0.5 × 0.5 × 0.5 mm) allow equal-quality reformats in any plane

Partial Volume Averaging

When a voxel contains two different tissue types, the computed HU is an average of both
Example: a voxel spanning cortical bone (1000 HU) and brain (40 HU) might compute as ~520 HU
Causes blurring of tissue borders and underestimates true density
Minimized by using thinner slices
A significant limitation in edge delineation for surgery planning

10. Windowing

The human eye distinguishes only ~30-40 shades of gray, yet the CT HU scale spans ~4000 values. Windowing maps a selected HU range onto the full grayscale:

Window Level (WL) = center HU value of the displayed range
Window Width (WW) = total HU range displayed; outside values appear pure black or white
Narrower window = higher contrast; wider window = more tissue types visible simultaneously

Standard Window Settings:

Window	Level (HU)	Width (HU)	Use
Brain/soft tissue	+35	+80	Intracranial pathology
Subdural	+75	+200	Hemorrhage detection
Lung parenchyma	-600	+1500	Airways, nodules
Mediastinum/soft tissue	+40	+400	Chest/abdomen
Bone	+400	+2000	Fractures, cortex
Liver	+60	+150	Hepatic lesions

11. Contrast Agents in CT

Iodinated Contrast Media

Iodine is a heavy atom (Z = 53) - its inner electron shell absorbs x-rays via photoelectric capture
Even small amounts create marked hyperdensity on CT
Water-soluble, injectable; cleared by kidneys

Routes and Uses:

IV contrast - most common; for vessel opacification and lesion detection
Oral contrast - bowel delineation
Rectal contrast - colorectal evaluation
Intrathecal (CT myelography) - spinal canal

Enhancement Phases (IV contrast):

Phase	Timing (post-injection)	Structures Opacified
Non-contrast	Before injection	Baseline density; calcium, hemorrhage, fat
Arterial	~25-35 sec	Aorta, hepatic artery, pancreas
Portal venous	~60-70 sec	Portal vein, liver parenchyma
Delayed/nephrographic	~3-5 min	Renal cortex, ureters, biliary system

Contrast Safety:

Risk of contrast-induced nephropathy in renal impairment
Allergic reactions: mild (urticaria) to severe (anaphylaxis)
Contraindicated with metformin (risk of lactic acidosis if acute kidney injury develops)
Non-ionic, low-osmolar/iso-osmolar agents are now standard - fewer reactions, better tolerated

12. Radiation Dose

Dose Metrics:

Metric	Definition	Units
CTDI (CT Dose Index)	Dose per rotation at the center and periphery of a standard phantom	mGy
CTDIvol	Volume-averaged CTDI; accounts for pitch	mGy
DLP (Dose-Length Product)	CTDIvol × scan length	mGy·cm
Effective dose	DLP × tissue-specific conversion factor; accounts for organ radiosensitivity	mSv

Typical Effective Doses:

Chest CT: ~5-7 mSv
Abdomen/pelvis CT: ~8-10 mSv
Head CT: ~1-2 mSv
CT coronary angiography (modern low-dose): ~1-3 mSv

Parameters Directly Affecting Dose:

Gantry geometry, rotation time
Tube current (mA) - 50% reduction in mA = 50% dose reduction
Tube voltage (kVp) - reducing from 120 to 80-100 kVp = large dose reduction (dose ∝ kVp²)
Pitch, Z-axis coverage, section collimation

Dose Reduction Strategies:

Strategy	How It Reduces Dose
Automatic Exposure Control (AEC)	Real-time mA modulation based on patient anatomy thickness/density
Weight/size-based mA	Pediatric and thin patients scanned at lower mA
Reduced tube current	Most direct; 50% mA = 50% dose
Reduced kVp	100 kVp in thin adults; 80 kVp not recommended even in small patients due to beam hardening
Higher pitch	Faster table = less exposure time
Iterative reconstruction	Allows lower mA with acceptable noise
Bow-tie filters	Reduces dose to low-attenuation peripheral tissue
Prospective ECG gating	X-rays only during specific cardiac phases
Restrict scan length	Cover only region of interest
Patient shielding	Thyroid, breast, eye shields

13. Image Quality Parameters

1. Spatial Resolution

Ability to distinguish two closely spaced structures
Determined by: detector size, focal spot, reconstruction kernel, pixel matrix, FOV
High-frequency (sharp) kernels: better edge definition, more noise (bone window)
Low-frequency (smooth) kernels: less noise, blurrier edges (soft tissue window)

2. Contrast Resolution

Ability to distinguish tissues with small HU differences
Improved by: higher mA (less noise), larger voxels, lower kVp (more contrast for iodine)

3. Temporal Resolution

Speed of image acquisition
Critical for cardiac CT
Improved by: faster gantry rotation, multi-segment reconstruction, dual-source CT

4. Noise

Random HU fluctuations; appears as graininess
Increases with: lower mA, higher kVp, smaller voxels, larger patient
Decreases with: higher mA, iterative reconstruction, larger slice thickness

5. Artifacts

Artifact	Cause	Appearance
Beam hardening	Polychromatic beam; low-energy photons preferentially absorbed; remaining beam "hardens"	Dark bands between dense objects (posterior fossa "Hounsfield bar")
Metal artifact	Extreme attenuation by metallic implants	Severe star/streak pattern
Motion artifact	Patient movement during acquisition	Blurring, ghosting, streaks
Partial volume	Multiple densities averaged in one voxel	Blurred, inaccurate edges
Ring artifact	Malfunctioning detector element	Concentric rings in image
Stair-step artifact	Thick slices in oblique/curved structures	Stepped edges in reformats
Windmill artifact	High-pitch spiral acquisition	Rotating pattern around high-contrast objects

14. Postprocessing Techniques

Technique	Description	Clinical Application
MPR (Multiplanar Reconstruction)	Axial data reformatted into coronal, sagittal, oblique planes	Spine, aorta, urinary tract
Curved MPR	Along a curved axis (e.g., vessel course)	Coronary arteries, colon
MIP (Maximum Intensity Projection)	Only the highest HU value along each ray is displayed	CT angiography, pulmonary nodules
MinIP (Minimum Intensity Projection)	Only the lowest HU along each ray	Airways, emphysema mapping
SSD (Shaded Surface Display)	Threshold-based surface rendering	Airway assessment
Volume Rendering (VR)	All HU values mapped to opacity and color	3D vascular, complex bone anatomy
Virtual Bronchoscopy/Colonoscopy	Endoscopic simulation from CT	Airway stenosis, polyp detection
Dual-Energy CT	Simultaneous dual kVp acquisition; material decomposition	Stone characterization, iodine maps, virtual non-contrast
Computer-Aided Detection (CAD)	Automated pattern recognition	Pulmonary nodule detection and volumetry
CT Perfusion	Serial rapid acquisitions with IV contrast; time-density curves	Cerebral/myocardial/hepatic perfusion

15. Dual-Energy CT (DECT)

Acquires data at two different x-ray energies simultaneously (e.g., 80 kVp and 140 kVp).

Principle:

Different materials have characteristic spectral signatures - their attenuation changes differently between low and high energies. This allows material decomposition.

Applications:

Renal stone characterization - uric acid (lower HU at both energies) vs. calcium oxalate
Virtual non-contrast images - iodine subtracted from contrast-enhanced scan
Iodine distribution maps - surrogate for perfusion (lung, liver, myocardium)
Gout - monosodium urate crystal detection in joints
Reduced beam-hardening artifacts
Improved lesion characterization - adrenal adenoma vs. metastasis without washout CT

16. Cardiac CT Physics

Requires high temporal resolution (<100 ms ideally) to freeze cardiac motion at 70 bpm
Modern dual-source CT achieves 66 ms temporal resolution

Acquisition Modes:

Mode	X-ray Exposure	Dose	Advantage
Retrospective ECG gating	Continuous throughout cardiac cycle	Higher (~12-15 mSv)	Reconstruct at any phase; functional data
Prospective ECG triggering	Only during pre-specified phase (end-diastole)	Lower (~1-5 mSv)	Most common; dose-efficient
High-pitch spiral (dual-source)	Single heartbeat acquisition	Very low (~0.5-2 mSv)	Needs low, regular HR

Calcium Scoring:

Unenhanced scan; Agatston score quantifies coronary artery calcium
Score 0 = very low risk; >400 = high risk of events

CT Coronary Angiography (CCTA):

HR <65 bpm required (beta-blockers pre-scan)
IV iodinated contrast essential
Sublingual nitrates for coronary vasodilation

17. Summary - CT Parameters at a Glance

Parameter	Primary Effect	Clinical Trade-off
kVp	Beam energy, tissue contrast	Lower kVp → more iodine contrast, more noise; higher dose per mAs
mAs	Photon quantity, noise, dose	Lower mAs → more noise; 50% reduction = 50% dose saved
Pitch	Scan speed, dose	Higher pitch = faster, lower dose, very slightly lower quality
Slice thickness	Spatial resolution, noise	Thinner = better resolution, more noise
FOV	Pixel size, spatial resolution	Smaller FOV = smaller pixels = finer detail
Kernel	Resolution vs. noise trade-off	Sharp (bone) vs. smooth (soft tissue)
Reconstruction algorithm	Noise, dose efficiency	Iterative < FBP for noise at same dose
Gantry rotation speed	Temporal resolution	Faster = better for cardiac/dynamic imaging

Sources consulted:

Bradley and Daroff's Neurology in Clinical Practice - CT physics, HU values, helical CT, contrast agents
Grainger & Allison's Diagnostic Radiology - MDCT, dose reduction, pitch, postprocessing
Brogdon's Forensic Radiology - CT history, Hounsfield development
Murray & Nadel's Respiratory Medicine - CT numbers, lung CT applications
Scott-Brown's Otorhinolaryngology - Pixels, voxels, partial volume effect

This is a shared conversation. Sign in to Orris to start your own chat.