Session 5 Instrumentation

Image Storage Devices

• The ultrasound machine's image processing portion consists of scan converter, preprocessing, image memory, post-processing, digital storage, and display.
• Echo signal information converts into images by the image processor after being digitized, filtered, detected, and compressed.

Image Processing

• Until this point, the echo data travels in scan-line form serially through the system, with one scan line at a time.
• No image has been formed until after detection of the echo voltage amplitudes.
• The scan-line data enters the Scan Converter, the first step in Image Processing.
• The data is preprocessed in the scan converter and stored in image form.
• Imaging processing uses a computer and binary memory for image information storage.
• Conversion becomes necessary because the transducer sends out an analog signal that needs to be converted back to analog for display.
• Signals subsequently travel from the receiver to the scan converter.
• The scan converter consists of pre-processing (A to D converter), digital memory, and post-processing (D to A converter).
• Scan conversion is done for each scan line to create an image in a fraction of a second, then repeated.
• Frame rate is related to the speed of scan conversion, which allows for real-time imaging.

Scan Converter

• Format of echo data changes into image form for image processing, storage, and display in the scan converter.
• Pulses from a transducer penetrating the body provide ultrasound information.
• Multiple penetrations, or spokes, become necessary for each two-dimensional image.
• The monitor displays video information as horizontal lines.
• The scan converter translates the information from the spoke format into the video format.
• Video is presented as horizontal lines converting the penetration pattern into the horizontal line pattern.
• A pulse straight down into the body results in a series of echoes in a column in memory at various row locations.
• A pulse emitted in an off-vertical direction results in a series of echoes located in various rows and columns.

Analog vs Digital

• Analog converters could store many signal amplitudes allowing for many shades of gray.
• Analog means infinite.
• A dimmer switch exemplifies an analog device.
• Everyday numbers are "real world" numbers known as Analog numbers.
• Analog numbers can have an unlimited and continuous range of values; a person's weight exemplifies this.
• Compared to analog, digital means finite, with far less choices available.
• The on-off switch represents a "digital" device with discrete, finite choices.
• Digital numbers have only discrete values; a person's weight on a digital scale represents this.
• Signals travel from the receiver to the scan converter.
• During reception, low voltage electrical signals are produced by a transducer to create an image.
• Weak signals become susceptible to contamination by electrical noise.
• Converting information from analog to digital form becomes advantageous because digital information faces a far reduced risk of contamination.
• Translating image information from the real world(analog) to the computer world (digital) and back functions as a 5 step process.
• Electrical signals created by the transducer during reception convert from analog to digital form by an analog to digital (A-to-D) converter.
• A string containing only 0s and 1s contains the digital information.
• The digital information becomes stored in the scan converter's computer memory.
• Any processing of the reflected signal before storage is called preprocessing.
• Limitations in the analog scan converter have made it obsolete.
• These limitations include image fade because stored charges on the silicon wafer dissipate over time.
• Image flicker is caused by switching between read and write mood.
• Instability degrades picture quality depending on usage length, room temperature, and humidity.
• Deterioration gradually degrades image as the instrument ages.

Digital Scan Converter Advantages

• Computer technology allows digital scan converters to convert images into numbers via digitizing.
• Computer memory stores each image in the digital converter as a series of zeroes and ones, storing echo reflection amplitudes.
• The image's numerical representation processes and then re-translates into an image on a monitor.
• Digital scan converters provide uniformity by creating consistent grayscale quality throughout the image.
• The image is stable because it does not fade or drift and shows durability against aging or heavy use.
• Processing yields speed creating a nearly instantaneous action, and accuracy creates error-free scans.

Image Processor - Preprocessing

• Occurs as part of the scan conversion process.
• It is performed before the image is stored in memory.
• Manipulation of image data before storage in the scan converter takes place here.
• The analog signal leaving the receiver prepares for the digital computer memory in the scan converter.
• The sonographer controls preprocessing, which alters the image data forever and cannot be reversed.
• Process of scan-line information going into image memory allows processing functions on the image.
• Preprocessing occurs before the echo data are stored in the image memory.
• In the A-to-D converter, incoming signals are assigned shades of gray based on their amplitudes.
• The image is still “live” when preprocessing occurs.
• Any function such as gain changes on a live image occurs during preprocessing, after which information goes to the digital storage, and then off to Postprocessing.

Fill-In (Pixel) Interpolation

• This method constructs new simulated data points to fill in missing data that is undetectable by the observer.
• Image improvement can be accomplished via this tool.
• Two-dimensional ultrasound images result from multiple ultrasound pulses directed into the body.
• A sector scan in the process leaves out intervening pixels when scan lines have increasing separation.
• Gaps, or missing data, exists between the scan lines, especially as the lines spread apart.
• A brightness value is assigned to a missed pixel during interpolation based on an average of brightnesses of adjacent pixels.
• Interpolation falls under a form of preprocessing that increases line density, improving special resolution.
• This process enhances the ability to precisely visualize round structures' borders.

Persistence

• Aka temporal compounding or temporal averaging.
• Averages sequential frames to smooth the image with reduced noise, higher signal to noise ratio, and improved image quality.
• Used for gray scale and color Doppler, it uses image processing techniques to continue to display information from older images.
• A number of previous frames are superimposed on the most current frame when incorporating persistence, so the current frame on previous frames gives a smoother image.
• Noise is reduced and the image smoothed when several frames containing noise are averaged.
• Dynamic range and contrast resolution improve due to speckle reduction.
• The primary limitation lies in the reduction in the displayed frame rate, reducing temporal resolution because averaged frames no longer show as independent.
• Operators control averaging of sequential frames allowing a selectable number from zero up to a maximum.
• Persistence proves most effective with slowly moving structures (higher levels are appropriate for slower-moving structures)
• Less useful when imaging rapidly moving structures like heart valves, persistence can prevent the accurate observation of rapid motion.
• Lower levels of persistence, or none at all, are appropriate for following rapidly moving structures.

Edge Enhancement

• This image processing method makes pictures look sharper sharpening boundaries to make them more detectable and measurements more precise
• The computer identifies and emphasizes sharp edge boundaries in the image.
• The method works by increasing the image contrast in the area immediately around the edge.

Spatial Compounding

• Method of using sonographic information from several different imaging angles to produce a single image.
• Phasing techniques direct scan lines in multiple directions so ultrasound beam interrogates structures more than once.
• Overlapped or compounded frames combine to form a real time image.
• Averaging frames that view the anatomy from different angles constitutes spatial compounding.
• The technique gets limited to reduced frame rate and temporal resolution..
• The technqiue employs electronic steering of the sound beam used in compound imaging and availability requires phased array transducers.
• The technique averages multiple frames obtained by steering the sound beams in different directions acquiring multiple images from different imaging angles.
• Speckle and clutter reduce caused by artifacts.
• Smooth surfaces are presented more completely at more than one angle, increasing the probability of achieving close to 90-degree incidence.
• Structures previously hidden beneath highly attenuating objects can be visualized by minimizing shadowing artifact.
• Persistence and spatial compounding differ in that all of the images used in persistence has the same view.
• Spatial compounding's images results from different views or angles.
• Some preprocessing, such as persistence and spatial compounding, are operator controllable

Panoramic Imaging

• This gives panoramic images or a "bigger picture”.
• Slide transducer along visualization area.
• New info is added to one end of the image.
• The machine compares echoes so that new info is placed in proper location.

Volume Imaging

• Three-dimensional imaging (3D, volume imaging) is a recent innovation in ultrasound
• A 2-D array with thousands of elements arranged in a checkerboard pattern creates 3-D images.
• Many parallel two-dimensional (2D-slice imaging) scans accomplishes volume imaging.
• Processing this 3D volume of echo information then occurs for presentation on 2D displays.
• Acquired at rates of up to 30 volumes per second.

Freehand Method or Manual Technique

• The sonographer moves the transducer through a path to gather the 2D slices
• As it relies upon the steady hand of the sonographer to move the transducer at the same speed over the tissue, this method remains the most operator dependent.
• Variability in movement across the plane makes 3D image measurements are impossible with this technique.
• The 2D slices, once converted to 3D format, undergo slicing to view coronal, sagittal, and axial planes

Automated Mechanical Technique

• Specialized transducers have been developed that are curved sequenced array transducers mounted onto a motor.
• Screen measurements occur for the 3D image as well as the use of real-time 3D, also known as four-dimensional (4D) ultrasound.
• The motor speed to which the transducer attaches limits the 4D image frame rate.

Electronic Array Scanning

• The latest technology for acquiring a 3D image is the new electronic array, called a2D transducer
• These transducers with greater than 2500 elements acquire real-time volumes.
• Similar to those in other imaging modes (magnetic resonance imaging [MRI) and computed tomography [CT]), serial slice presentations exist.
• 3D images need assembled by assembling several 2D scans into a 3D volume of echo information in the image memory.
• Presentation forms are post-processing choices that present the stored 3D volume of echo information.
• 3-D acquisition has preprocessing attributes whereas 3-D presentation has postprocessing ones.
• Real time 3-D imaging involves 4-D imaging, utilizing a 2-D array.
• Rendering creates an image from 3-dimensional data acquired during the ultrasound examination
• It provides an element of realism to the image, creating a photo like image.
• 3-D rendering on an offline, stand-alone computer system occurs after acquiring and storing ultrasound data.
• Surface renderings are popular specifically for obstetric imaging and echocardiography.

Image Processor - Storing Image Frames

• Image memory stores images after scan conversion and preprocessing.
• One frame freezing from those shown per second creates a freeze frame.
• Allows us to review the last several frames that were acquired before the image was frozen in a Cine loop.

Image memory

• Echo data is preprocessed, so the image frames are stored in the image memory.
• Storing each image in memory while the sound beam is scanned through the anatomy allows.displays individual frames out of the rapid sequence captured in real-time instruments.
• Holding and displaying one frame out of the sequence creates freeze-frame.
• Instruments contain cine loops and review the last several frames acquired before freezing.

Binary System

• The numbers zero and one rather than zero through nine are the only one used within digital devices such as computers.
• Binary numbers are based on only the choices/values of 0 or 1.
• Any signal coming into the computer converts into only zeroes and ones.
• Signals in ultrasound machines get represented by black-and-white dots, or echoes; a black echo represents zero (0) or something "off".
• Pixel and bit both represent two important elements of digital systems.
• A pixel represents the smallest element of a digital picture.
• Pixel density, measured as the number of picture elements per inch, with increased density meaning increased spatial resolution also creates images with greater detail.
• A bit identifies the binary digit and the smallest amount of computer memory with values of either 0 or 1.
• One bit allows for the two shades of black and white.
• Either a value of o or 1 signals that a bit is bistable.
• A series of zeroes and ones, such as 0101010011, represents a binary number.
• The number of bits in memory (per pixel) determines the number of shades of gray.
• Each pixel's shade of gray is determined by the cluster of bits assigned to it.
• The more shades can be displayed and the better an image's contrast resolution with more bits per pixel.
• For example, 4 bits in memory allow 16 shades of gray and 8 bits 256 shades of gray.
• A byte consists of a group of eight bits of computer memory so eight bits translates to 1 byte.
• A word of computer memory has two bytes or 16 bits.
• More bits per pixel create better contrast resolution.
• Four bits exemplify 16 shades of gray.
• Contrast resolution is the ability to distinguish subtle differences in echo strength between adjacent media.
• The amount of columns and rows refer to an image matrix .
• Each cell in the matrix stores a location storing a 1 = on or a 0=off.
• In digital memory containing a digital 10 x 10 matrix with 4 bits of memory each cell stores a one or a zero, so more bits results in more shades of grays can show on the spectrum.
• If a one-bit memory available it would display black and white only, though having the 4 bits provides around 16 shades of gray possible.
• More detail from pixels improves with better spatial resolution.
• Each storage location in memory corresponds to one pixel on the display
• Each pixel on the display has values of being a one for black, it gets translated to a zero for white.
• Calculating the number of gray shades that can be represented by a cluster of bits requires multiplying the number two by itself as many numbers as there exist of bits.
• Example: if 5 bits of memory exist, one multiplies 2 five times to get 32, allowing 32 different gray shades, therefore,ultrasound systems often have 6-bit to 8-bit memories

Postprocessing

• Manipulation of image data after it gets stored in memory gets referred to as this.
• Functions get performed on frozen images.
• Operators control postprocessing so its changes prove reversible.
• Postprocessing functions after data storage in the scan converter including once signals arrive in the D-to-A converter.
• A component of the digital scan converter, so signals convert back into analog form to show, send to PACS, film, etc.
• When an image gets frozen, settings for image alterations display by certain gray shades.
• After A-to-D conversion (digitization) and storage, postprocessing occurs.
• Since all postprocessing changes can be reversed, the initial numerical values can restore.
• Manipulation of a 3D volume image is postprocessing.
• Image memory provides a place for the images gets stored after scan conversion and preprocessing.
• In the postprocessing stage, the process is completed, and any changes or alterations becomes post processing.
• Assigned brightness is the assignment of specific numbers in display through numbers that get retrieved from a type of memory.
• Changes of all types can be added in the "freeze frame” of processing along with black and white inversion, magnification, contrast, 3-D rendering and presentation, measurement, and cine loop
• Contrast determines the range of brilliance in the image and in bistable it indicates the ranges of only shades of white and black.
• Magnification also gets used to help anatomy as it is the sonographer choice to show detail. the selected part of the image gets know as the ROI, this can be both read or written.
• The sonoghaper can improve the visualization of the detail and with the magnification used with the image. This can have 2 forms of read and write.
• With reading: this is to add extra detail to the images, that get scored through data after the "scan reader" and after data gets scored it is a common process.

Read Magnification (post-processing function)

• Zooming becomes its AKA and the ROI may be unreadable.
• Consisting from three steps from anatomy, scanned images show from analog to digital form with ROIs that can identify the system.
• Magnification gets characterized for the same numbers in the images.

Write Magnification (preprocessing function)

• Scaling exists to alter that from analog to digital along with all new acquired data where the images is greater for numbers.
• Magnification occurs through acquired data before storage including scanning from all images that can later get stored in this system.
• Identifying the point it can be shown and it is discarded and re-scanned in the system.
• With color it can become a common process with the option for new colors or bar colors that get applied in these sections.