Weeks 1 & 2

ENIAC

The Electronic Numerical Integrator and Calculator (1946), recognized as the first general-purpose electronic computer. It contained over 18,000 vacuum tubes.

Transistor

An electronic switch, developed in 1948, that alternately allows or prevents electronic signals from passing. Its invention enabled the development of the "stored program" computer.

Microprocessor

Developed in 1971 by Ted Hoff of Intel Corporation, this component is the central processing unit (CPU) of a microcomputer, placing millions of circuits on a single chip.

Four Generations of Computers

The developmental stages of computers, distinguished by their electronic technology:

• 1st (1939-1958): Vacuum tubes

• 2nd (1958-1964): Individually packaged transistors

• 3rd (1964-1971): Integrated circuits (ICs)

• 4th (1975-present): Large-Scale Integration (LSI) and Very Large-Scale Integration (VLSI), placing millions of circuits on a single chip.

Computer (General-purpose, stored-program)

An electronic digital device. General purpose means the computer can be used to solve various problems. Stored program means the computer holds instructions and data in its memory.

Analog vs. Digital

Analog refers to a continuously varying quantity (e.g., the hands on a mechanical watch). Digital refers to a system using only two discrete values (e.g., 0 and 1) that vary through coding.

Hardware

The visible, physical components of a computer system, including input, output, memory, and processing devices.

Software

The set of computer programs that instruct the hardware on what to do and how to store and manipulate data.

Binary Number System

The number system used by computers, which has only two digits: 0 and 1. All computer operations convert characters and values into binary.

Bit

A single binary digit (a 0 or a 1), which is the smallest unit of measure in computer language.

Byte

A group of 8 bits. Computer capacity is often expressed by the number of bytes it can accommodate.

Word (Computer Word)

A group of bytes (typically 2, 4, or 8) that are interpreted and processed by the computer as a single unit. Standard word lengths are 32-bit or 64-bit.

Kilobyte (kB) (in computer use)

Equal to 1024 bytes (2¹⁰), not the metric 1000. It is used to describe image size, dynamic range, and storage capacity.

Gigabyte (GB)

A unit of computer capacity equal to 1024 megabytes (MB), or approximately 1 billion bytes.

Terabyte (TB)

A unit of computer capacity equal to 1024 gigabytes (GB), or approximately 1 trillion bytes.

Hexadecimal Number System

A number system with 16 symbols (0-9 and A-F) used in assembly-level applications. Each symbol represents a set of four bits.

Algorithm

A step-by-step process used to solve a problem, much like a recipe is used to bake a cake.

FORTRAN

(FORmula TRANslation) The oldest programming language for scientific, engineering, and mathematical problems, developed in 1956.

BASIC

(Beginners All-purpose Symbolic Instruction Code) An easy-to-learn, algebraic programming language developed in 1964 for students.

C++

A programming language developed in 1980 that incorporates object-oriented programming (OOP) methods to manage complex programs.

Central Processing Unit (CPU)

The primary element of a computer that manipulates data and executes software instructions. It consists of a control unit and an arithmetic/logic unit (ALU).

Control Unit

The part of the CPU that directs the flow of data, tells the computer how to carry out software instructions, and controls data transfer between memory and I/O hardware.

Arithmetic/Logic Unit (ALU)

The part of the CPU that performs all arithmetic and logic calculations and temporarily holds the results.

Processor

The combination of electronic circuitry that performs the actual computations and the memory that supports this function.

Memory vs. Storage

Memory is active, primary, and internal (e.g., RAM), holding data for immediate use. Storage is archival, secondary memory (e.g., hard drive) for long-term filing.

Random Access Memory (RAM)

Active, primary memory where data can be accessed randomly from any location in roughly the same amount of time. Its contents are temporary.

Read-Only Memory (ROM)

Memory containing firmware—information supplied by the manufacturer that cannot be written on or erased. It includes instructions for starting the computer (bootstrap).

Motherboard

The main circuit board in a computer, containing the microprocessor, RAM and ROM chips, and expansion slots for adding other components.

Secondary Memory

Archival storage for data not currently in use, such as CDs, DVDs, hard drives, and flash drives.

Solid-State Drive (SSD)

A form of internal data storage that uses solid-state principles for faster data access and greater durability compared to traditional hard drives, but is also more expensive.

Redundant Array of Independent Discs (RAID)

A system of two or more hard drives in a single cabinet that act as one storage system, providing greater reliability in case one disk fails.

Jukebox (Optical Disc)

A multi-disc drive that can handle up to 2,000 optical discs. Modeled after the classic music jukebox, it is often used in medical imaging departments to replace film file rooms for archiving.

Operating System

A series of instructions that organizes the flow of data through the computer and is the program most closely related to the hardware (e.g., MAC-OS, Windows).

Application Programs (Apps)

Software written to guide the computer to perform a specific user function, such as word processing, image reconstruction, or financial analysis.

Compiler

A program that translates an application program written in a high-level language into a machine language program that the computer can execute.

Input Hardware

Devices that convert data into a form the computer can use, such as keyboards, mice, scanners, and voice-recognition systems.

Output Hardware

Devices that translate computer information into a form humans can understand, such as digital display screens and printers.

Soft Copy

The visual output seen on a digital display device.

Telecommunications

The transfer of data from a sender to a receiver across a distance.

Teleradiology

The practice of transferring medical images and patient reports to remote sites for interpretation by radiologists.

Modem

A device (modulator/demodulator) that converts a computer's digital signals to analog for transmission over analog lines and converts incoming analog signals back to digital.

Broadband

High-speed internet access that is always on and faster than traditional dial-up, including technologies like DSL, cable modem, and fiber optics.

Pixel

A "picture element," one of the small squares that make up a rectangular digital image.

Image Matrix

The arrangement of pixels in columns and rows that form a digital image. Common sizes in medical imaging include 256x256 to 1024x1024 for CT and MRI, 1024x1024 for digital fluoroscopy, and up to 4096x4096 for digital radiography and mammography.

Dynamic Range (Bit Depth)

The number of shades of gray that can be represented by a pixel, determined by its bit depth. For example, a digital mammogram has a 16-bit dynamic range, meaning it can display 2¹⁶ or 65,536 shades of gray.

DICOM

(Digital Imaging and Communications in Medicine) A standard method developed by ACR and NEMA for storing and transmitting diagnostic medical images.

PACS

(Picture Archiving and Communication System) A system that manages the storage, transfer, and archiving of digital medical images, enabling teleradiology.

Artificial Intelligence (AI)

The adaptation of computer technology to mimic human cognitive activities like learning and problem-solving. In medical imaging, it aids in tasks like pattern recognition.

Machine Learning

A subset of AI where a computer is "trained" by being provided with a large collection of data (e.g., medical images) from which it learns to perform a task.

Supervised Learning

A machine learning training method where the computer is given a labeled dataset (e.g., images with abnormalities identified by a programmer) to learn from a known "ground truth."

Unsupervised Learning

A machine learning training method where the computer is given an unlabeled dataset and must identify patterns, clusters, and abnormalities on its own.

Deep Learning

A subset of machine learning that uses multi-layered artificial neural networks. The "deep" refers to the large number of steps (layers) used to analyze input data.

Artificial Neural Networks (ANNs)

Computer models, inspired by the human brain, containing connected nodes (neurons) that transmit information through layers to process data and perform analysis.

Quantum Computing

A computing paradigm based on the principles of quantum mechanics, using qubits instead of classical bits to achieve a leap forward in processing speed and capacity.

Qubit

A "quantum bit." Like a bit, it can be in a 0 or 1 state, but it can also exist in a superposition of both states simultaneously, allowing for many more values.

Computed Radiography (CR)

The first form of digital radiography, introduced in 1981, which uses a photostimulable phosphor plate to acquire the image data.

Photostimulable Luminescence (PSL)

The process where a material (like barium fluorohalide) emits light first upon x-ray exposure, and later when stimulated by a different light source (a laser).

Photostimulable Phosphor (PSP)

The material used in CR plates (e.g., barium fluorohalide) that exhibits photostimulable luminescence to store a latent image.

Europium (Activator)

A small amount of an element added to the PSP that is responsible for the storage property of the phosphor. It is analogous to the sensitivity center of a film emulsion, as without it, no latent image would be possible.

Metastable State

A higher-energy, temporary state into which electrons are excited by x-ray interaction with the PSP. The trapped electrons in this state form the latent image.

Storage Phosphor Screen (SPS)

A screen containing PSPs where the latent image is stored in the form of metastable electrons.

Imaging Plate (IP)

The rugged cassette that houses the Storage Phosphor Screen (SPS). The IP has a lead backing to reduce backscatter x-rays, which improves contrast resolution.

Latent Image (in CR)

The invisible image formed by electrons trapped in a high-energy, metastable state within the photostimulable phosphor after x-ray exposure.

CR Step 1: Expose

An x-ray beam exposes the Imaging Plate (IP), transferring energy that excites electrons in the phosphor into a metastable state, forming the latent image.

CR Step 2: Stimulate

A finely focused red laser beam scans the IP, causing the trapped metastable electrons to return to their ground state.

CR Step 3: Read

As electrons return to their ground state, they emit shorter-wavelength blue light. This light is detected by a photodetector, making the latent image visible as an analog signal.

CR Step 4: Erase

The IP is flooded with intense white light to remove any residual latent image by returning all remaining metastable electrons to their ground state, preventing ghosting on subsequent images.

Ghosting Artifact

An artifact that can appear on a CR image if the previous image's residual latent image is not completely erased from the IP.

CR Reader

A device that combines mechanical, optical, and computer modules to process the IP. It scans the plate with a laser, reads the emitted light, and digitizes the signal.

Slow Scan

The slow, constant movement of the imaging plate (IP) along its long axis within the CR reader.

Fast Scan

The rapid back-and-forth scanning of the laser beam across the IP, perpendicular to the slow scan direction.

Laser (in CR Reader)

The source of stimulating light that causes the release of energy from the trapped electrons in the PSP.

Photodetector (e.g., PD, PMT)

A device (such as a Photodiode or Photomultiplier Tube) that detects the emitted blue light from the PSP and converts it into an analog electrical signal.

Analog-to-Digital Conversion (ADC)

The process of converting the analog signal from the photodetector into a digital signal that the computer can process. It consists of sampling and quantization.

Sampling and Quantization

The two processes of ADC. Sampling is the measurement of the analog signal at discrete points in time. Quantization is the assignment of a discrete numerical value to each sample.

Image Receptor Response Function (CR)

A linear response to x-ray exposure, unlike the S-shaped characteristic curve of screen-film. CR image contrast remains constant over a very wide exposure range.

Exposure Latitude (CR vs. Screen-Film)

CR is characterized by extremely wide exposure latitude, capable of responding to five decades of radiation exposure and producing nearly 100,000 gray levels. This is far greater than the latitude of screen-film.

Image Noise (CR Sources)

Besides scatter radiation, CR noise can arise from: • Mechanical defects: Drive motors • Optical defects: Laser fluctuations, light scatter • Computer defects: Electronic noise, inadequate sampling or quantization.

Patient Dose in CR

Because CR image receptors are "faster" than a 400-speed screen-film system at low exposures, lower patient radiation doses are possible.

Relationship of kVp and mAs in CR

The traditional roles ("kVp for contrast, mAs for density") do not apply. Since CR image contrast is constant regardless of exposure, images can be made with higher kVp and lower mAs, further reducing patient dose.