Imaging Techniques in Medicine

Medical imaging is a critical field in biomedical engineering, visualizing internal body structures and functions for diagnosis, monitoring, and treatment planning.
Techniques involve various methods with strengths, limitations, and applications.
The evolution of medical imaging has shifted from qualitative to quantitative analysis enabling evidence-based medical decisions.

Medical diagnosis is determining a disease or disorder.
The process involves consultation (patient information), physical examination (inspection, auscultation, measurements), medical tests, and laboratory analysis (biosignal analysis, image analysis).

A monochrome image is a 2D function f(x, y), where x and y represent spatial coordinates, and f represents the pixel intensity.
Pixel values range from 0 to 255 (8-bit grayscale) where 0 is black, 255 is white, and intermediate values represent shades of grey.

This profile represents pixel intensity variations along a line or region in the image.
Essential for analyzing spatial intensity changes useful in edge detection, and texture/uniformity assessment, and object boundary analysis.

RGB images comprise three components (red, green, and blue (R,G,B)).
Each pixel is defined by intensity values corresponding to the three components (red, green, and blue), producing a wide array of colors through additive mixing. (red + green = yellow, red + blue = magenta, green + blue = cyan, red + green + blue = white).

Three-dimensional imaging combines multiple 2D images using techniques like CT and MRI to produce complex volume data.
Useful for comprehensive visualization of anatomical structures.

A digital image is represented as a 2D matrix of pixels with numerical data (intensity for grayscale or color for RGB images).
Key processes are discretization (dividing image into pixels) and quantization (assigning intensity level to pixel based on detected signal).

A computer vision system replicates human visual perception utilizing computational methods for analyzing and interpreting visual data.
The core components are image acquisition, preprocessing, segmentation, feature extraction, analysis, and interpretation.
The different filters used in computer vision systems are linear and nonlinear filters.
- Linear Filters are designed to modify pixel values by calculating a weighted average of neighbouring pixels.
- Nonlinear Filters are more robust to outliers and focus on preserving edges.
  - Median Filters: replaces each pixel value with the median of its neighbourhood. Ideal for salt-and-pepper noise reduction.

Filtering techniques: both linear and nonlinear techniques are used to improve image quality (e.g., smooth images, reduce noise, or enhance features).
Segmentation: this process involves dividing an image into meaningful regions for further analysis. Segmentation methods include k-means clustering and multilayer perceptrons.

Radiography (X-ray Imaging): Uses X-rays to produce images on film or digital receptors.
Computed Tomography (CT): Combines multiple X-ray images to generate 3D cross-sectional views.
Magnetic Resonance Imaging (MRI): Uses magnetic fields and radio waves to detect hydrogen atoms in tissues, producing high-quality images of soft tissues.
Ultrasonography (USG): Uses high-frequency sound waves to create real-time images.
Nuclear Medicine (PET, SPECT): Involves radiotracers that emit gamma rays to visualize metabolic processes.
Endoscopy: Uses cameras and light sources for direct internal visualization of internal organs.
Thermography: Uses infrared radiation to detect temperature variations.

The electromagnetic spectrum encompasses a range of EM waves that vary in frequency and wavelength. The range in this spectrum includes gamma rays, x-rays, ultraviolet, visible light, infrared, microwaves, and radio waves.

Image clarity is the measure of how well information is displayed, and is affected by unsharpness, contrast, noise, and distortion/artifacts.
Image contrast differentiation of subtle features, depends in part on the imaging technique, tissue properties, and contrast agents.
Image noise refers to extraneous information that interferes with the visualization of critical features reducing diagnostic accuracy.

The Larmor frequency is related to the precession frequency of a proton in a magnetic field and it is measured with the following formula: ƒ= γB
Where Y (gamma) is a gyromagnetic ratio and B is field induction in Tesla. The Larmor equation is measured in MHz/Tesla; for a 2 Tesla magnetic field the Larmor equation ≈ 85.2 MHz.
The Larmor frequency is important in MRI as it allows for tuning RF waves resulting in resonance enabling signal detection related to the protons spin.

Computer-aided diagnosis (CAD): computer algorithms to enhance the analysis of IR images, increasing detection accuracy.
Active Dynamic Thermography (ADT): Analyzing thermal transients in IR imaging, enhances the detail quality.
Image artifacts: undesired distortions or features present in the image.
Image resolution and noise: these parameters are crucial for proper image interpretation.
Image degradation models: models describing how images are degraded.

Different medical imaging modalities provide unique advantages and disadvantages. Each type of modality excels in imaging particular structures or types of diseases, offering various tradeoffs in image quality resolution, cost, invasiveness, and radiation exposure.

Various medical applications for infrared thermography are possible, including detecting temperature variations and tissue abnormalities, and detecting patterns in different types of tissue.

Techniques to improve image clarity, such as enhancing or removing noise, minimizing distortions, and enhancing contrast.

Different detector materials for infrared imaging and their respective spectra range. Examples include Indium gallium arsenide, Germanium, Lead Sulfide, Lead Selenide, Indium Antimonide, Indium Arsenide, Platinum Silicide, Mercury Cadmium Telluride, Mercury Zinc Telluride, Lithium tantalate, Triglycine sulfate

QWIPs (Quantum Well Infrared Photodetectors): utilize a sandwiching technique to enhance the sensitivity of the photodetectors.
Microbolometer fabrication: Describes the different steps in constructing the microbolometer FPA.