The US Region Calibration Module has been introduced into the ultrasound IOD to provide access to the full range of data that may be present in a single US image. US images often contain multiple regions that have independent data regions, e.g., quad screen loops that may have different calibration information. The data presented in the various regions of a US image can represent a multiplicity of physical parameters, e.g., spatial distance, blood velocity, time, volume, etc., and these are often contained in the value of the pixel itself. It is therefore imperative that physical information be available for the various regions of a single region independent of each other.
Table C.8-17 contains IOD Attributes that describe an ultrasound region calibration.
Table C.8-17. US Region Calibration Module Attributes
|
The bounds of a rectangle specifying the location of the region, x0,y0,x1,y1. See Section C.8.5.5.1.14 for further explanation. |
|||
|
The bounds of a rectangle specifying the location of the region, x0,y0,x1,y1. See Section C.8.5.5.1.14 for further explanation. |
|||
|
The bounds of a rectangle specifying the location of the region, x0,y0,x1,y1. See Section C.8.5.5.1.14 for further explanation. |
|||
|
The bounds of a rectangle specifying the location of the region, x0,y0,x1,y1. See Section C.8.5.5.1.14 for further explanation. |
|||
|
The physical units of the dimensions of the region. See Section C.8.5.5.1.15 for Enumerated Values. |
|||
|
The physical units of the dimensions of the region. See Section C.8.5.5.1.15 for Enumerated Values. |
|||
|
The physical value increments per positive X pixel increment. The units are as specified in the Physical Units X Direction (0018,6024). See Section C.8.5.5.1.17 for further explanation. |
|||
|
The physical value increments per positive Y pixel increment. The units are as specified in the Physical Units Y Direction (0018,6026). See Section C.8.5.5.1.17 for further explanation. |
|||
|
This coordinate pair, x0,y0, defines the location of a virtual "reference" pixel. See Section C.8.5.5.1.16 for further explanation. |
|||
|
This coordinate pair, x0,y0, defines the location of a virtual "reference" pixel. See Section C.8.5.5.1.16 for further explanation. |
|||
|
The Physical Value at the reference pixel x location. The units are specified in the Physical Units field. |
|||
|
The Physical Value at the reference pixel y location. The units are specified in the Physical Units field. |
|||
|
The spatial organization of the data within the region. See Section C.8.5.5.1.1 for Enumerated Values. |
|||
|
The type of data within the region. See Section C.8.5.5.1.2 for Enumerated Values. |
|||
|
Flags used for special handling of the region. See Section C.8.5.5.1.3 for Enumerated Values and further explanation. |
|||
|
Describes how the components of a pixel can be described. Required if pixel component calibration exists for this region. See Section C.8.5.5.1.4 for Enumerated Values and further explanation. |
|||
|
This value is ANDed with the composite pixel code for each pixel within the region, then shifted right by the number of contiguous least significant zeros in the mask to obtain what will be referred to as the "Shifted Masked Composite Pixel Code" (SMCPC). Required if Pixel Component Organization = Bit aligned. See Section C.8.5.5.1.5 for further explanation. |
|||
|
Defines the start of the numeric range of values within the composite pixel where calibration is to be defined by the "pixel physical calibration table". To be used only when ranges are used to describe the portion of the composite pixel. |
|||
|
Defines the stop of the numeric range of values within the composite pixel where calibration is to be defined by the "pixel physical calibration table". To be used only when ranges are used to describe the portion of the composite pixel. |
|||
|
The physical units to be applied to the pixel component. Required if Pixel Component Organization exists. See Section C.8.5.5.1.6 for further explanation. |
|||
|
The type of data for the pixel component. Required if Pixel Component Organization exists. See Section C.8.5.5.1.7 for further explanation. |
|||
|
The number of break point coordinate pairs used to describe a piece wise linear curve. Required if Pixel Component Organization equals 0 or 1. Otherwise not used. See Section C.8.5.5.1.8 for further explanation. |
|||
|
An array of X values used to create the piece wise linear curve. Required if Pixel Component Organization equals 0 or 1. Otherwise not used. See Section C.8.5.5.1.9 for further explanation. |
|||
|
An array of Y values used to create the piece wise linear curve. Required if Pixel Component Organization equals 0 or 1. Otherwise not used. See Section C.8.5.5.1.9 for further explanation. |
|||
|
The number of entries in the Table of Pixel Values. Required if the value of Pixel Component Organization (0018,6044) is 2 or 3. Otherwise not used. See Section C.8.5.5.1.11 for further explanation. |
|||
|
A table of Pixel Values used in conjunction with the Table of Parameter Values (0018,605A) or Pixel Value Mapping Code Sequence (0040,9098) to provide a mapping from Pixel Value to a real world value. Required if the Pixel Component Organization equals 2. Otherwise not used. See Section C.8.5.5.1.12 for further explanation. |
|||
|
A table of Parameter Values used in conjunction with the Table of Pixel Values (0018,6058) to provide a mapping from Pixel Value to Parameter Value. Required if the value of Pixel Component Organization (0018,6044) is 2. Otherwise not used. See Section C.8.5.5.1.13 for further explanation |
|||
|
Sequence that, in conjunction with the Table of Pixel Values (0018,6058), provides a mapping from a Pixel Value to an associated Coded Concept. One or more Items shall included in this Sequence. The number of Items shall be equal to the value of Number of Table Entries (0018,6056). Required if the value of Pixel Component Organization (0018,6044) is 3 (Code Sequence look up). See Section C.8.5.5.1.18 for further explanation. |
|||
|
For IVUS Baseline CID 3497 “IVUS Arterial Morphology”; no Baseline CID is defined otherwise. |
|||
|
The manufacturer defined description of center frequency of the interrogating ultrasound energy. The units are kilohertz. |
|||
|
The ultrasound pulse repetition frequency, as defined by the manufacturer, used to collect data in the region. The units are in hertz. |
|||
|
The steering angle, as defined by the manufacturer, used for a steered 2D image. The units are degrees. |
|||
|
The x displacement, in pixels, from the Reference pixel to the center of the Doppler sample volume. |
|||
|
The y displacement, in pixels, from the Reference pixel to the center of the Doppler sample volume. |
|||
|
The starting and ending coordinates pairs of the m-line. Where the X0,Y0are the starting point and X1,Y1are the end point of the tm-line. See Section C.8.5.5.1.10 for further explanation. |
|||
|
The starting and ending coordinates pairs of the m-line. Where the X0,Y0are the starting point and X1,Y1are the end point of the tm-line. See Section C.8.5.5.1.10 for further explanation. |
|||
|
The starting and ending coordinates pairs of the m-line. Where the X0,Y0are the starting point and X1,Y1are the end point of the tm-line. See Section C.8.5.5.1.10 for further explanation. |
|||
|
The starting and ending coordinates pairs of the m-line. Where the X0,Y0are the starting point and X1,Y1are the end point of the tm-line. See Section C.8.5.5.1.10 for further explanation. |
|||
Values for Region Spatial Format (0018,6012) indicate the spatial organization of the data within the region.
Values for Region Data Type (0018,6014) indicate the type of data within the region.
Enumerated Values:
- 0000H
- 0001H
- 0002H
- 0003H
- 0004H
- 0005H
- 0006H
- 0007H
- 0008H
- 000AH
- 000BH
- 000CH
- 000DH
- 000EH
- 000FH
- 0010H
- 0011H
- 0012H
Region Flags (0018,6016) specify characteristics of US Regions.
Bit 0 of the Region Flags specifies the relative priority of the pixel component calibration specified by an US Region in the case where the US Region intersects with other US Regions. The calibration supplied by one or more of the regions may not be valid in the area that they intersect.
Enumerated Values for Bit 0 (lsb):
A high priority region overwrites data of a low priority region when they overlap, thus invalidating any pixel component calibration specified for a low priority region. pixel component calibration of overlapping regions of the same priority is indeterminate where they overlap. Figure C.8-6 shows an example of intersecting regions.
In this example, Region B is Color Flow while Region A is Tissue Echo. If Region B Color Flow values share the same bit planes as Region A Tissue Echo values, then it is indeterminate whether a pixel in this region is a Color Flow pixel or a Tissue Echo pixel. Since the pixels of the Color Flow region overwrite those of the Tissue Echo region, the Region Flag of the Tissue Echo region is assigned low priority and the Region Flag of the color region is assigned high priority. This means that if both the Tissue Echo and Color Flow regions define pixel component calibration that only the calibration specified by the Color Flow region can be applied to the pixel data value at Point X.
The measurement in Figure C.8-6 is a line between Point Y and Point Z. Both points are in Region A so the distance between them can be calculated using the Region A scaling (assuming that Region A defines both the Physical Units X Direction and Y Direction as being cm). If the points are in Region B, and hence also in Region A, it is still possible to calculate the distance because the region scaling is identical in both regions. The lower priority of Region B only applies to its pixel component calibration, not its X and Y direction scaling.
Ultrasound systems should set this to 1 if the image is scaled automatically by the ultrasound system. If the image is frame-grabbed and scaling is not available then it should be set to 0. If the region is protected, the region can not be manually rescaled. That is the data defined by the region calibration Module can not be overridden by a reader of that image.
Valid for PW and CW regions only. Indicates which type of Doppler scale is used.
Enumerated Values for Bit 3-4 Scrolling Region:
Pixel Component Organization (0018,6044) provides an Enumerated Value describing how the components of a pixel can be described. The absence of this data element means that pixel component calibration does not exist for this region.
Other values reserved for future use.
Pixel Component Organization defines the way in which the composite pixel values are mapped into real world values with physical units, as illustrated in Figure C.8-7.
An example of Component Calibration for an ultrasound image is shown in Figure C.8-8.
In this example, some pixels lie within two Regions. One Region specifies pixel component calibration for Doppler velocity values. The second Region specifies pixel component calibration for Doppler magnitude. A particular Pixel Data (7FE0,0010) value will thus map to a displayed value, a Doppler velocity and magnitude value.
The example has a Palette Color Photometric Interpretation with 16 Bits Allocated and Bits Stored per sample. The Palette Color Lookup Tables also have 16 bits for each entry. The fact that the example has just one sample per pixel means that each composite pixel value is identical to the single Pixel Data value. An example Pixel Data value is shown in brackets along with the output values resulting from each step where it is processed.
The Pixel Data value is mapped to red, green, and blue values from the supplied Palette Color Lookup Tables before being displayed. The display device supports 8 bits per sample and thus requires the scaling of the output values from the 16 bit per entry LUTs.
The Doppler Velocity Region maps each pixel value in the Region to the Doppler velocity. The Pixel Component Organization (0018,6044) has a value of zero, indicating bit aligned positions with a bit mask. The Pixel Component Mask (0018,6046) specifies that the least significant 4 bits of the most significant byte convey the Doppler velocity of each pixel. The Pixel Component Physical Units (0018,604C) are cm/sec, and the Pixel Component Data Type (0018,604E) indicates color flow velocity. The Table of X Break Points (0018,6052) and Table of Y Break Points (0018,6054) map each masked composite pixel value to a Doppler velocity value in cm/sec.
The Doppler Magnitude Region maps each pixel value in the Region to the Doppler magnitude. The Pixel Component Organization (0018,6044) has a value of zero, indicating bit aligned positions with a bit mask. The Pixel Component Mask (0018,6046) specifies that the most significant 4 bits of the most significant byte convey the Doppler magnitude of each pixel. The Pixel Component Physical Units (0018,604C) is set to dB, and the Pixel Component Data Type (0018,604E) indicates color flow magnitude. The Table of X Break Points (0018,6052) and Table of Y Break Points (0018,6054) map each masked composite pixel value to a Doppler magnitude value in dB.
Pixel Component Mask (0018,6046) is ANDed with the Composite Pixel Code (see Section C.7.6.3.1.1) for each pixel within the region, then shifted right by the number of contiguous least significant zeros in the mask to obtain what will be referred to as the "Shifted Masked Composite Pixel Code".
The mask will most likely (but not necessarily) contain a block of contiguous ones, surrounded by leading and trailing zeros. The purpose of this mask is to keep only those bits within the composite pixel code that pertain to the region. It is to be used only when Pixel Organization is bit aligned positions.
The Number of Table Break Points (0018,6050) gives the number of entries in each of two tables: the Table of X Break Points (0018,6052) and Table of Y Break Points (0018,6054). These tables are used to designate a curve mapping the value of a pixel component to its actual physical value, as described in Section C.8.5.5.1.9.
Table of X Break Points (0018,6052) and Table of Y Break Points (0018,6054) are individual arrays of coordinates that interpreted together are used to create a piecewise linear curve. Each X value from the Table of X Break Points is matched with the corresponding Y value from the Table of Y Break Points yielding an (X,Y) coordinate. The set of (X,Y) coordinates describes a piecewise linear curve mapping the value of a pixel component to its actual physical value (in units defined in Pixel Component Physical Units data element (0018,604C)).
The X direction on the curve has no units, and represents actual pixel component values. If the Pixel Component Organization (0018,6044) is "Bit aligned positions", and the width of the Pixel Component Mask is n bits then the X coordinates are in the range 0 through 2n-1. If the Pixel Component Organization is Ranges, then the X coordinates are in the range 0 through 2 number of bits in the composite pixel- 1.
Note
The X value is NOT relative to the Pixel Component Range Start (0018,6048). Not all possible X values in the range need be covered by the curve.
For any pixel component value in the range of the curve described by this table, the corresponding Y value is the actual physical value for that pixel, in units specified in Pixel Component Physical Units (0018,604C). If the pixel component value is NOT within the range of specified X values for the curve, then no pixel calibration is defined by this region. It may be possible for pixel calibration to be defined by other spatial regions intersecting this one.
The TM-Line Position X0 (0018,603D) and TM-Line Position Y0 (0018,603F) are the coordinates of the starting point and TM-Line Position X1 (0018,6041), TM-Line Position Y1 (0018,6043) are the coordinates of the end point of the TM-line. The coordinate is defined as the displacement, in pixels, from the Reference pixel. Typically used for M-mode line and CW Doppler.
The Number of Table Entries (0018,6056) gives the number of entries in the Table of Pixel Values, the number of entries in the Table of Parameter Values (0018,605A), if present, and the number of items in the Pixel Value Mapping Code Sequence (0040,9098), if present.
The Table of Pixel Values (0018,6058) specifies the pixel values that are mapped to real world parameter values or coded concepts (tissue characterizations). The number of entries in the table is given by Number of Table Entries (0018,6056).
A pixel is calibrated (mapped to a real-world value) by finding an entry in the Table of Pixel Values that matches its Composite Pixel Code (see Section C.7.6.3.1.1). The offset index of this entry is used as an index into the Parameter Value Table (0018,605A) or as a sequence item number in the Pixel Value Mapping Code Sequence (0040,9098) to select the real world value. The first Table of Pixel Values entry corresponds to sequence item 1.
Note
If a Composite Pixel Code has no matching value in the Pixel Value Table then there is no unambiguous way to determine the corresponding Parameter Value. A method may exist to determine a valid Parameter Value but the specification of such a method is outside the scope of the DICOM standard. No assumption should be made that linear interpolation will produce a valid result.
The Table of Parameter Values (0018,605A) provides the real world values for pixel values identified in the Table of Pixel Values (0018,6058). The number of table entries is given by Number of Table Entries (0018,6056) and the physical units are given by Pixel Component Physical Units (0018,604C). Values may repeat when a parameter value is associated with more than one Composite Pixel Code value.
These attributes specify the location of the region, Region Location Min X0 (0018,6018), Region Location Min Y0 (0018,601A), Region Location Max X1 (0018,601C), Region Location Max Y1 (0018,601E) expressed as offsets to the pixel coordinates. The upper left corner of the entire image is x=0,y=0 and the lower right corner is x=image width - 1, and y=image length - 1. Thus, a region will be specified as within these bounds. Where x0,y0 is the coordinate of the upper left corner of the region and x1,y1 is the coordinate of the lower right corner of the region.
Physical Units X Direction (0018,6024) and Physical Units Y Direction (0018,6026) indicate the physical units of the dimensions of the region.
This coordinate pair, Reference Pixel X0 (0018,6020), Reference Pixel Y0 (0018,6022) defines the location of a virtual "reference" pixel. This reference pixel location is used to tie the image's pixel coordinate system to the physical coordinate system. For example, the reference pixel could be defined where a depth of zero centimeters occurs in the 2D image, or it could define where the baseline (i.e.,: zero frequency) resides in a spectral display. The reference pixel location is the relative offset from the Region Location Min X0 (0018,6018) and Region Location Min Y0 (0018,601A), not the image origin. The location is not required to be within the region or even within the image boundary. For this reason, the Reference Pixel X0 and Reference Pixel Y0 values can be positive or negative.
The reference pixel location varies depending on the type and spatial organization of the data within the region.
Tissue data is tissue echo intensity displayed as grayscale. The Region Data Type (0018,6014) value is 0001H (Tissue). Color flow is Doppler signal displayed as color and encoded as some function of Doppler magnitude and velocity of blood flow or tissue motion. The Region Data Type value is 0002H (Color flow). For 2D, the Region Spatial Format (0018,6012) is 0001H (2D), meaning that the region is a tomographic image. For such 2D regions the reference pixel location is typically at the center of the transducer face on the tissue-transducer interface (skin line).
Figure C.8-1 shows 2D attribute values of reference pixel location along with Region Location Min and Region Location Max. for 2D-Tissue and 2D-Color Flow Regions:
Both the 2D regions-Tissue and Color Flow-share the same physical location at the skin line but the reference pixel location values (Reference Pixel X0 and Reference Pixel Y0) are relative to their respective region origins at the skin line.
Spectral Doppler is the time varying magnitude of Doppler signal as function of frequency. Region Data Type (0018,6014) value is 0003H (pulsed wave Doppler) or 0004H (continuous wave Doppler). Spectral Doppler regions display the magnitude of Doppler signal with frequency or velocity as the vertical dimension and time as the horizontal dimension. Spectral Doppler regions have a Region Spatial Format (0018,6012) of 0003H (Spectral). The time dimension for the Region Spatial Format displays horizontally with data scrolling toward the left or sweeping toward the right. The reference pixel location is the pixel in the frame where:
Figure C.8-2 shows an example of reference pixel locations in an image with both a Tissue and a scrolling Spectral (CW or PW Doppler) Region. The user adjusts the depth and position of the Doppler sample volume. The system annotates the sample volume position on the 2D region and specifies the location in Doppler Sample Volume X Position (0018,6039) and Doppler Sample Volume Y Position (0018,603B).
The scrolling Spectral Region reference pixel location specifies the horizontal location at the time of the current image frame. Data to the left of this location in the Spectral Region was acquired in the past. Because time increases to the right, the Physical Delta X (0018,602C) for this Region is positive. To specify the location of the most recent data the Reference Pixel x0specifies the time of acquisition, and the Reference Pixel Physical Value X (0018,6028) specifies the reference time to be zero. The Physical Units X Direction (0018,6024) is seconds. For an explanation of how to handle sweeping regions refer to Section C.8.5.5.1.16.7 Treatment of Sweeping Regions.
The Reference Pixel Physical Value Y (0018,602A) value specifies the baseline where velocity or frequency are zero. Typically spectral Doppler regions display positive velocity (cm/Sec) or frequency shift (Hz) above the baseline. This indicates flow toward the transducer face. Negative velocity or frequency information is displayed below the baseline. This indicates flow away from the transducer face. The Physical Delta Y (0018,602E) value is therefore negative because vertical coordinates increment downward.
M-Mode is tissue or color flow with a Region Spatial Format (0018,6012) of 0002H (M-mode). The vertical reference pixel location is the transducer face.
The horizontal reference pixel location is the pixel in the frame where:
Figure C.8-3 shows an example of reference pixel locations for 2D Tissue and M-Mode Regions within the same image frame. The system annotates the sample line position on the 2D tissue region and specifies its position with the TM-Line Position attributes (0018,603D), (0018,603F), (0018,6041) and (0018,6043).
The physical length of the TM-Line corresponds directly to the physical height of the M-Mode Region. The M-Mode region's Reference Pixel y0 can be used to calculate the depth of the M-Mode region and facilitate depth measurements. In this example the M-Mode Region Reference Pixel y0 has a negative value corresponding to the distance between the face of the ultrasound probe and the TM-Line starting point. Note that the negative offset in pixel units is determined using the pixel height-width scaling of the M-Mode - Tissue Region as this could differ from the scaling of the 2D - Tissue Region (as it does in this example).
Waveforms are traces with a Region Spatial Format (0018,6012) value of 0004H (Waveform). The Reference Pixel x0 (0018,6020) specifies the time origin as the time of frame capture. There is typically no baseline position for ECG traces; the Reference Pixel y0 (0018,6022) is arbitrary.
Figure C.8-4 shows an example of reference pixel location for 2D Tissue, M-Mode, and ECG Waveform Regions within the same image frame:
Doppler Traces have a Region Spatial Format (0018,6012) value of 0004H (Waveform) and a Region Data Type value of 0005H (Doppler Mean Trace), 0006H (Doppler Mode Trace) or 0007H (Doppler Max Trace). The Reference Pixel x0 (0018,6020) specifies the time origin as the time of frame capture. The Reference Pixel y0 (0018,6022) is the Doppler Baseline position (zero velocity / frequency position).
For regions with Region Spatial Format (0018,6012) value of 0005H (Graphics) the reference pixel location has no meaning.
Time-based display of data may scroll the acquired data from a fixed horizontal location to the left. Alternatively, sweep-based display increments the horizontal location of the acquired data, overwriting previously acquired data to the right. When the horizontal location corresponding to zero time has completely swept over the older data, writing wraps from the left of the region. Thus, sweep-based displays have a time discontinuity. The measurement of time intervals across the discontinuity requires special treatment. The time interval between two points across the discontinuity is equal to the region's time width minus the point separation. The sweeping area can be treated as a single region. The Reference Pixel x0 should indicate the time origin for the multi-frame image, which will be the location of the sweeping region's discontinuity line for the first frame of the multi-frame image. In order to specify that this is actually the location of the discontinuity line, the Reference Pixel Physical Value X (0018,6028) must be set to 0 seconds. This indicates that this location corresponds to the time at which the first frame was acquired.
It is useful to be able to calculate the location of the discontinuity line for subsequent frames of a multi-frame image. This is necessary if one is to determine whether two points are on opposite sides of the discontinuity line and also to correctly calculate the difference in time between such points. The x-axis location of the discontinuity line, x, for a given frame number, y, can be calculated from the Reference Pixel x0, x0, the Reference Pixel x1, x1, the time offset for frame y, t, (determined from the Frame Time Vector (0018,1065) or Frame Time (0018,1063)) and the Physical Delta X (0018,602C), px, as follows:
x = x0 + modulus((t / px) / (x1 - x0))
Alternatively, two regions can be used, one on each side of the time discontinuity. Figure C.8-5 shows the use of two regions. Note that the two region approach is not valid for multi-frame images, as the same region scaling must apply to all the frames.
The two region approach may also be used in Doppler or physiological sweeping regions.
Time-based display of data may also be a combination of sweeping and scrolling. Sweep-based display is used at the start of acquisition, incrementing the horizontal location of the acquired data from left to right. After the horizontal location corresponding to zero time has completely swept to the right hand limit of the region, writing scrolls to the left from the right hand limit rather than wrapping from the left. A single region should be specified when this combination of behavior is used. The x-axis (zero time) location, x, for a given frame number, y, can be calculated from the Reference Pixel x0, x0, the Reference Pixel x1, x1, the time offset for frame y, t, (determined from the Frame Time Vector (0018,1065) or Frame Time (0018,1063)) and the Physical Delta X, px, as follows:
The Physical Delta X (0018,602C) is the physical value increment per positive X pixel increment, which is left to right. The Physical Delta Y (0018,602E) is the physical value increment per positive Y pixel increment, which is top to bottom.
Note
When displaying Doppler data, ultrasound applications typically display the Doppler strip horizontally, with data sweeping (moving time origin) from left (oldest) to right (newest) or scrolling (static time origin) from right to left. The default display of positive velocity values normally indicates flow toward the transducer; negative velocity values indicate flow away from the transducer. In this case a negative Physical Delta Y is required to specify that the direction of positive velocities or frequencies is upward.