wk6 Imaging acquisitions

Nuclear Medicine Image Acquisition Factors

Prior to image acquisition, several parameters must be defined: - Radionuclide: Selection of the correct peak and energy window for the specific isotope. For example, for $^{99m}\text{Tc}$ , the peak is $140\text{\,keV}$ with a $20\%$ energy window. - Collimator: Common selections include the Low Energy High Resolution (LEHR) collimator. - Type of Acquisition: Includes choices such as dynamic, static, whole body, SPECT/CT, and gated acquisitions. Each scan protocol typically involves at least one, and potentially multiple, acquisition types. - Zoom Factor and Matrix Size: Parameters that determine the magnification and spatial sampling of the clinical images.
Post-acquisition procedures involves the display of diagnostic images and the rigorous analysis of gathered data as required by the clinical protocol.

Processing the Digital Image Matrix

Digital images are generated through the accumulation of valid events that fall within the pre-determined energy window.
Matrix Acquisition: The most frequent method for capturing gamma camera events. An image matrix is constructed in the computer memory. - Matrices are organized as square grids, commonly sized at $64 \times 64$ , $128 \times 128$ , $256 \times 256$ , or $512 \times 512$ .
Event Mapping: Every time a valid event is detected, two-dimensional coordinates (x and y) are identified. - 1. PM tube signals are digitized. - 2. The event x-y location is computed. - 3. The specific pixel location in the matrix is determined. - 4. That pixel is incremented by $1\text{\,count}$ .
Acquisition continues until the specific pre-determined stopping limit is reached.

Matrix Scaling, Spatial Resolution, and Counts

Pixels: These are the picture elements that comprise the matrix. Matrix size refers explicitly to the count of pixels (e.g., a $64 \times 64$ matrix contains $4,096\text{\,pixels}$ ).
Resolution vs. Pixel Size: The larger the matrix size, the smaller the individual pixel. Smaller pixels generally provide better image detail and spatial resolution. However, physical resolution limitations of the imaging device prevent further improvements beyond a certain threshold.
Whole Body Sweep: Utilizes a specific rectangular matrix, typically $512 \times 1024$ .
Pixel Value (p(x,y)): Corresponds to the total number of counts collected. - Data is stored as an array of count values. - Display relies on assigning a grey or color scale based on counts. - The pixel containing the highest count value is assigned the brightest intensity.

Pixel Depth and Data Storage

Pixel Depth: This is the maximum number of counts/events a pixel can record, determined by the bits allocated. - 8 bits (1 byte): Allows for a range of $0-255\text{\,counts}$ . - 16 bits (1 word): Allows for a range of $0-65,535\text{\,counts}$ .
Overflow: If an acquisition records more than $255\text{\,counts}$ in an 8-bit pixel, the value resets to $0$ , resulting in erroneous data and image artefacts.
Nuclear medicine acquisitions standardly utilize word mode ( $16\text{\,bits}$ ).

Spatial Resolution Determinants

Spatial resolution is defined as the ability of the gamma camera to resolve small objects or the minimum distance between two points such that they remain distinct.
It is governed by two primary factors: - 1. The resolution of the imaging hardware (collimator and detector). - 2. The size of the pixels used in the digital image.

List Mode versus Frame Mode Acquisition

Frame Mode: This is the most common acquisition method. Individual events are sorted immediately into the digital matrix grid upon digitization. Memory requirements are small, determined primarily by matrix size and the total number of frames.
List Mode: Incoming x and y signals are digitized but not immediately sorted. Instead, x and y coordinates are stored along with periodic markers (e.g., at millisecond intervals). - Advantage: Offers extreme flexibility for post-acquisition data analysis. - Disadvantage: Consumes a massive amount of memory. - Historical Context: While previously less common, it is seeing a resurgence with modern camera technology.

Image Magnification and Zoom Protocols

Zoom: Allows for the magnification of an object within the field of view without changing the matrix size.
It modifies the size of the pixels while maintaining a constant number of pixels.

Dynamic Acquisition and Temporal Evaluation

Consists of a series of images acquired sequentially to measure the rate of radiopharmaceutical accumulation or excretion within organs or tissues.
Framing: Each frame is acquired for a fixed amount of pre-set time.
Matrix: Usually acquired in $64 \times 64$ or $128 \times 128$ \text{\,matrices}.
Frame Rates: Can range from as fast as $50\text{\,frames/sec}$ to as slow as $1\text{\,frame/hr}$ .
Dynamic Phases: Acquisitions can be split; within each specific phase, the frame rate stays fixed.
Clinical Examples: - Bone flow: $2-3\text{\,sec/frame}$ for $60\text{\,seconds}$ . - Biliary scan: $1\text{\,min/frame}$ for $60\text{\,frames}$ . - Renal scan: Phase 1 at $1\text{\,sec/frame}$ for $40\text{\,frames}$ ; Phase 2 at $20\text{\,sec/frame}$ for $57\text{\,frames}$ .
Images are often summed together for display or reviewed as a cine.

Static Imaging and Whole-Body Scanning

Static Acquisition: A single 2D image of a specific body area. - Matrix: $128 \times 128$ or $256 \times 256$ . - Stopping Limits: Can be set by counts (e.g., $800\text{k\,counts}$ ) or time (e.g., $5\text{\,mins}$ ). - Clinical Specifics: Thoracic bone views ( $1000\text{k\,counts}$ ); Feet views ( $300\text{k\,counts}$ ). Pelvic views should be time-acquired rather than count-acquired to prevent bladder activity from prematurely ending the scan.
Whole Body Acquisition: A 2D image covering the entire patient. - The scanning bed moves automatically through detectors. - Matrix: $512 \times 1024$ or $512 \times 256$ . - Speed: Typically $10-15\text{\,cm/min}$ for bone scans, resulting in a total scan time of $15-20\text{\,minutes}$ .

SPECT (Single Photon Emission Computed Tomography) Fundamentals

SPECT involves 3D images reconstructed from multiple planar projection images taken at various angles (typically $3^{\circ}$ intervals).
Dual Detectors: Each head typically rotates $180^{\circ}$ .
Acquisition Type: Standardly uses "step and shoot," where the camera moves, stops to acquire, and then moves again.
Configuration: Often acquired as $64\text{\,views}$ at $10\text{\,sec/view}$ .
Matrix: $64 \times 64$ or $128 \times 128$ .
Planes: Images are reconstructed into Sagittal, Coronal, and Transverse planes, and can be viewed as Maximum Intensity Projections (MIP).
Advantages: Improved anatomical localization, increased contrast, and availability of quantitative analysis.
Disadvantages: Long acquisition times ( $10-20\text{\,minutes}$ ), low counts per projection, and susceptibility to artifacts from patient motion or high-intensity activity zones.

Gated Cardiac Acquisition Techniques

Used to image beating heart chambers and left ventricular function using a Gated Blood Pool Study (GHPS) with $^{99m}\text{Tc}$ labelled Red Blood Cells (RBCs).
ECG Integration: Uses the R-wave of the patient's ECG (R-R interval).
Frame Division: The cardiac cycle is divided into frames (usually $8$ or $16$ ).
Process: Computer records the first portion of the cycle into frame 1, second into frame 2, etc. The subsequent R-wave reset starts the process back at frame 1.
Viewed as a cine and used for quantitative functional analysis.

DICOM, PACS, and Information Management

DICOM: Digital Imaging and Communications in Medicine. This is the standardized format for medical images (X-ray, CT, MRI, NM, RT) to allow data exchange between systems.
PACS: Picture Archiving and Communication System. Used for the storage and network-based display of DICOM images.
RIS: Radiological Information System. Used for the electronic management of imaging departments.
NMT Responsibilities: Ensure correct details are selected from the worklist or entered manually; errors must be identified immediately.

The Rationale for Hybrid Imaging (SPECT/CT)

Modern practice requires detailed anatomical info for surgery and radiation therapy.
Nuclear medicine has limitations in anatomical detail because radiopharmaceuticals do not concentrate in all tissues.
Functional vs. Anatomical: targeted agents visualize metabolic changes at the cellular level before anatomy changes. Hybrid imaging (SPECT/CT, PET/CT, PET/MRI) combines functional and anatomical data.
Quote: Dr. Stanley Goldsmith stated: "SPECT without hybrid imaging is like reading with one eye closed. Hybrid imaging opens both eyes."

Historical Milestones in Hybrid Modalities

CT: Developed in the 1970s by Sir Godfrey Hounsfield (Nobel Prize 1979).
MRI (formerly NMR): Developed in the 1970s by Dr. Paul Lauterbur and Sir Peter Mansfield (Nobel Prize 2003).
SPECT: Developed late 1970s/early 1980s.
PET: Developed mid-1970s.
SPECT/CT: Developed late 1980s/early 1990s; first clinical use in 1999.
PET/CT: Named TIME Magazine Medical Invention of the Year in 2000.
PET/MRI: Developed in 2010.

Clinical Advantages and Limitations of Hybrid Systems

Advantages: - Anatomical localization of uptake. - Metabolic functional imaging. - Attenuation correction. - Increased diagnostic specificity and reporter confidence. - Detection of incidental findings.
Disadvantages: - Increased equipment, shielding, and training costs. - Increased radiation exposure: Low-dose CT adds $1-5\text{\,mSv}$ ; diagnostic CT adds $2-20\text{\,mSv}$ or more.

Principles of X-ray Computed Tomography (CT)

Utilizes a rotating x-ray tube with a rotation speed of approximately $0.4\text{\,sec}$ per $360^{\circ}$ .
Detectors are arranged in a ring formation around the patient.
Helical CT: The x-ray tube rotates continuously while the bed moves, allowing larger areas to be scanned without stopping (e.g., within a single breath hold), reducing motion artifacts.
CT Density/Hounsfield Units: - High Density (Bone/Metal): Displays as light grey or white. - Low Density (Air/Fat): Displays as dark grey or black.

Radiation Safety and Dose Optimization in SPECT/CT

CT significantly increases patient dose. Mispositioning of the patient can increase dose by $20\%$ .
Optimization Strategies: - Minimum scan length (only scan what is necessary). - Raising the patient’s arms improves image quality and decreases dose.
Low-dose CT is usually sufficient for NM attenuation correction and localization.

CT Attenuation Correction (CTAC) and Hybrid Image Registration

Attenuation Problem: Photons from deeper tissues are more likely to be absorbed (attenuated), leading to a loss of counts and artifacts, particularly in cardiac imaging.
CTAC Process: A low-dose CT is acquired. The reconstructed CT produces attenuation coefficients used to create an attenuation map. This map is applied to SPECT images to compensate for photon loss.
Registration: Overlaying images from two modalities. - Limitations include differences in slice thickness, organ status (e.g., bladder filling), and respiratory motion. - Best utilized for imaging of the brain and skeleton.
Hardware Configuration: SPECT and CT may be on the same gantry or separate gantries. SPECT is usually acquired first. Most cameras have a "no CT zone" and require the patient to remain in the middle of the field of view (FOV).