Annex RR Ophthalmic Refractive Reports Use Cases (Informative)

RR.1 Introduction

Refractive instruments are the most commonly used instruments in eye care. At present many of them have the capability for digital output, but their data is most often addressed by manual input into a paper or electronic record.

Refractive instruments address the power of a lens or of a patient's eye to bend light. In order for a patient to see well light must be focused on the retina in the back of the eye. If the natural optics of a patient's eye do not accomplish this, corrective lenses can bend incident light so that it will be focused on the retina after passing through the optics of the eye. The power of an optical system such as a spectacle lens or the eye is measured by its ability to bend light, and is measured in diopters (D). In practical clinical applications, this is measured to 3 decimal points, in increments of 0.125 D. The power of a lens is measured in at least two major meridians. A spherical lens power occurs when the power is the same in all meridians (0-180 degrees). A cylindrical lens power occurs when there is a difference in lens power across the various meridians. The shape of the anterior surface of the eye largely determines what type of correcting lens is needed. An eye that requires only spherical lens power is usually shaped spherically, more like a ball, while an eye that requires cylindrical lens power is ellipsoid and shaped more like a football.

Lenses can also bend light without changing its focal distance. This type of refraction simply displaces the position of the image laterally. The power of a prism to bend light is measured in prism diopters. In practical clinical applications this is measured to 1 decimal point, in increments of 0.5 prism diopters. Prism power is required in a pair of spectacles most commonly when both eyes are not properly aligned with the object of regard. Clinical prisms are considered to bend all light coming in from the lens either up, down, in toward the nose, or out away from the nose, in order to compensate for ocular misalignment.

Visual acuity is measured in various scales , all of which indicate a patient's vision as a fraction of what a reference standard patient would see at any given distance. For example, if a patient has 20/30 vision it means that he sees from a distance of 20 feet what a reference standard patient would see from a distance of 30 feet. These measurements are determined by presentation of standardized objects or symbols (optotypes) of varying sizes calibrated to reference standard vision (20/20). The smallest discernable optotype defines the patient’s visual acuity expressed in a variety of formats (letters, numbers, pictures, tumbling E, Landolt C, etc).

Visual acuity is measured in two categories of viewing distances : distance, and near. Distance visual acuity is measured at 20’ or six meters. This distance is roughly equivalent to optical infinity for clinical purposes. The near viewing distance can vary from 30cm to 75 cm depending on a variety of other conditions, but most commonly is measured at 40 cm.

Visual acuity is measured under several common viewing conditions : 1) Uncorrected vision is measured using the autoprojector to project the above mentioned optotypes for viewing, with no lenses in front of the patient's eyes. The line of smallest optotypes of which the patient can see more than half is determined, and that information is uploaded to a computer system. 2) The patient's vision using habitual correction is measured in a similar fashion using whichever vision correction the patient customarily wears. 3) Pinhole vision is measured in a similar fashion, with the patient viewing the optotypes through a pinhole occluder held in front of the eye. Pinhole visual acuity testing reduces retinal blur, providing an approximation of what the patient's vision should be with the best possible refractive correction (spectacles) in place. 4) Best corrected visual acuity is the visual acuity with the best refractive correction in place. 5) Crowding visual acuity measures the presence and amount of disparity in acuity between single optotype and multiple optotype presentations.

A patient’s spectacle prescription may or may not represent the same lenses that provided best corrected visual acuity in his refraction. Subjective comfort plays a role in determining the final spectacle prescription.

1.) Autolensometer: an autolensometer is used to measure the refractive power of a patient's spectacles. This is done by the automatic analysis of the effect of the measured lens upon a beam of light passing through it. Output from an autolensometer can be uploaded to a phoropter to provide a baseline for subjective refraction (discussed below), and it can be uploaded to a computerized medical record. Lenses may also be measured to confirm manufacturing accuracy.

2.) Autorefractor: an autorefractor is used to automatically determine, without patient input, what refractive correction should provide best corrected visual acuity. Output from an autorefractor can be uploaded to a phoropter to provide a baseline for subjective refraction (discussed below), and it can be uploaded to a computerized medical record.

3.) Phoropter (or phoroptor): an instrument containing multiple lenses, that is used in the course of an eye exam to determine the individual’s subjective response to various lenses (subjective refraction) and the need for glasses or contact lenses.. The patient looks through the phoropter lenses at an eye chart that may be at 20 ft or 6m or at a reading chart that may be at 40 cm. Information from the subjective refraction can be uploaded from an autophoropter to a computer. The best corrected vision that was obtained is displayed in an autoprojector, and that information can also be uploaded to a computer.

4.) Autokeratometer: an autokeratometer is used to measure the curvature, and thus the refractive power, of a patient's cornea. Two measurements are generally taken, one at the steepest and one at the flattest meridian of the cornea. The meridian measured is expressed in degrees, whole integers, in increments of 1 degree. If the measurement is expressed as power, the unit of measurement is diopters, to 3 decimal points, in increments of 0.125D. If the measurement is expressed as radius of curvature, the unit of measurement is millimeters, to 2 decimal points, in increments of 0.01 mm.

RR.2 REFERENCE TABLES for EQUIVALENT VISUAL ACUITY NOTATIONS

RR.2.1 Background

Visual acuity is defined as the reciprocal of the ratio between the letter size that can just be recognized by a patient, relative to the size just recognized by a standard eye . If the patient requires letters that are twice as large (or twice as close), the visual acuity is said to be 1/2 ; if the letters need to be 5x larger, visual acuity is 1/5, and so on.

Note that the scales in the tables extend well above the reference standard (1.0, 20/20, the ability to recognize a letter subtending a visual angle 5 min. of arc), since normal acuity is often 1.25 (20/16), 1.6 (20/12.5) or even 2.0 (20/10).

Today, the ETDRS chart and ETDRS protocol, established by the National Eye Institute in the US, are considered to represent the de-facto gold standard for visual acuity measurements The International Council Of Ophthalmology, Visual Standard, Aspects and Ranges of Vision Loss (April, 2002) is a good reference document.

The full ETDRS protocol requires a wide chart, in the shape of an inverted triangle, on a light box, and cannot be implemented on the limited screen of a projector (or similar) chart.

For most routine clinical measurements projector charts or traditional charts with a rectangular shape are used; these non-standardized tools are less accurate than ETDRS measurements.

This appendix contains two lookup tables, one for traditional charts and one for ETDRS measurements.

RR.2.2 Notations

Various notations may be used to express visual acuity. Snellen (in 1862) used a fractional notation in which the numerator indicated the actual viewing distance; this notation has long been abandoned for the use of equivalent notations, where the numerator is standardized to a fixed value, regardless of the true viewing distance. In Europe the use of decimal fractions is common (1/2 = 0.5, 1/5 = 0.2); in the US the numerator is standardized at 20 (1/2 = 20/40, 1/5 = 20/100), while in Britain the numerator 6 is common (1/2 = 6/12, 1/5 = 6/30).

The linear scales on the right side of the tables are not meant for clinical records. They are required for statistical manipulations, such as calculation of differences, trends and averages and preferred for graphical presentations. They convert the logarithmic progression of visual acuity values to a linear one, based on Weber-Fechner’s law, which states that proportional stimulus increases lead to linear increases in perception.

The logMAR scale is calculated as log (MAR) = log (1/V) = – log (V). LogMAR notation is widely used in scientific publications. Note that it is a scale of vision loss, since higher values indicate poorer vision. The value “0” indicates “no loss”, that is visual acuity equal to the reference standard (1.0, 20/20). Normal visual acuity (which is better than 1.0 (20/20)) is represented by negative logMAR values.

The VAS scale (VAS = Visual Acuity Score) serves the same purpose. Its formula is: 100 – 50 x logMAR or 100 + 50 x log (V). It is more user friendly, since it avoids decimal values and is more intuitive, since higher values indicate better vision. The score is easily calculated on ETDRS charts, where 1 point is credited for each letter read correctly. The VAS scale also forms the basis for the calculation of visual impairment ratings in the AMA Guides to the Evaluation of Permanent Impairment.

RR.2.3 Use of the lookup table

Data input: Determine the notation used in the device and the values of the lines presented. No device will display all the values listed in each of the traditional columns. Convert these values to the decimal DICOM storage values shown on the left of the same row. DICOM values are not meant for data display. In the table, they are listed in scientific notation to avoid confusion with display notations.

In the unlikely event that a value must be stored that does not appear in the lookup table, calculate the decimal equivalent and round to the nearest listed storage value.

Data display: If the display notation is the same as the input notation, convert the DICOM storage values back to the original values. If the notation chosen for the display is different from the input notation, choose the value on the same row from a different column. In certain cases this may result in an unfamiliar notation; unfortunately, this is unavoidable, given the differences in size progressions between different charts. If a suffix (see attribute “Visual Acuity Modifiers” (0046,0135)) is present, that suffix will be displayed as it was recorded.

Suffixes: Suffixes may be used to indicate steps that are smaller than a 1 line difference. On traditional charts, such suffixes have no defined numerical value. Suffixes +1, +2, +3 and -1, -2, -3 may be encountered. These suffixes do not correspond to a defined number of rows in the table.

RR.2.4 Traditional Charts

The Traditional charts used in clinical practice are not standardized; they have an irregular progression of letter sizes and a variable number of characters per line. Measurement accuracy may further suffer from hidden errors that cannot be captured by any recording device, such as an inconsistent, non-standardized protocol, inaccurate viewing distance, inaccurate projector adjustment and contrast loss from room illumination. Therefore, the difference between two routine clinical measurements should not be considered significant, unless it exceeds 5 rows in the table (1 line on an ETDRS chart).

Table RR-1 contains many blank lines to make the vertical scale consistent with that used in Table RR-2. Notations within the same gray band are interchangeable for routine clinical use, since their differences are small compared to the clinical variability, which is typically in the order of 5 rows (1 ETDRS line).

TABLE RR-1Reference table for use with TRADITIONAL CHARTS

DICOM Notations for Clinical Use with Traditional Charts Scales for statistics and graphical displays

Decimal Visual Acuity Traditional scales Linear Scales
Decimal US 6 m LogMAR VAS
2.00 E+00 2.0 20/10 6/3 -0.30 115
1.74 E+00 -0.24 112
1.66 E+00 -0.22 111
1.38 E+00 -0.14 107
1.30 E+00 1.3 20/15 6/4.5 -0.12 106
1.10 E+00 1.1 20/18 6/5.5 -0.04 102
1.05 E+00 -0.02 101
8.70 E-01 0.06 97
8.30 E-01 0.08 96
7.00 E-01 0.7 20/28 6/8.7 0.16 92
6.60 E-01 0.66 20/30 6/9 0.18 91
5.50 E-01 0.26 87
5.25 E-01 0.28 86
4.37 E-01 0.36 82
4.17 E-01 0.38 81
3.50 E-01 0.46 77
3.33 E-01 0.33 20/60 6/18 0.48 76
2.75 E-01 0.56 72
2.63 E-01 0.58 71
2.20 E-01 0.66 67
2.10 E-01 0.68 66
1.74 E-01 0.76 62
1.66 E-01 0.17 20/120 6/36 0.78 61
1.38 E-01 0.86 57
1.30 E-01 0.13 20/150 6/45 0.88 56
1.10 E-01 0.96 52
1.05 E-01 0.98 51
8.70 E-02 1.06 47
8.30 E-02 0.083 20/240 6/72 1.08 46
7.00 E-02 1.16 42
6.60 E-02 0.065 20/300 6/90 1.18 41
5.50 E-02 1.26 37
5.25 E-02 1.28 36
4.40 E-02 1.36 32
4.20 E-02 1.38 31
3.50 E-02 1.46 27
3.33 E-02 1.48 26
2.75 E-02 1.56 22
2.63 E-02 1.58 21
2.20 E-02 1.66 17
2.10 E-02 1.68 16
1.74 E-02 1.76 12
1.66 E-02 1.78 11
1.38 E-02 1.86 7
1.30 E-02 1.88 6
1.10 E-02 1.96 2
1.05 E-02

Decimal Visual Acuity Use with suffixes Calculated values Linear Scales

Figure SS.3-1: Colon radiograph as Described in Example 1

The content tree structure would resemble:

Node Code Meaning of Concept Name Code Meaning or Example Value TID
1 Colon CAD Report 4120
1.1 Language of Content Item and Descendants English 1204
1.2 Image Set Properties 4122
1.2.1 Frame of Reference UID 1.2.840.114191.123 4122
1.2.2 Study Instance UID 1.2.840.114191.456 4122
1.2.3 Study Date 20060924 4122
1.2.4 Study Time 090807 4122
1.2.5 Modality CT 4122
1.2.6 Horizontal Pixel Spacing 0.80 mm 4122
1.2.7 Vertical Pixel Spacing 0.80 mm 4122
1.2.8 Slice Thickness 2.5 mm 4122
1.2.9 Spacing between slices 1.5 mm 4122
1.2.10 Recumbent Patient Position with respect to gravity Prone 4122
1.3 CAD Processing and Findings Summary All algorithms succeeded; without findings 4121
1.4 Summary of Detections Succeeded 4120
1.4.1 Successful Detections 4015
1.4.1.1 Detection Performed Nodule 4017
1.4.1.1.1 Algorithm Name “Colon Polyp Detector” 4019
1.4.1.1.2 Algorithm Version “V1.3 4019
1.4.1.1.3 Series Instance UID 1.2.840.114191.789 4017
1.5 Summary of Analyses Not Attempted 4120

SS.3.2 Example 2: Colon Polyp Detection with Findings

A colon CAD device processes a screening colon case with several hundred images, and a colon polyp detected. The colon radiograph resembles:

Colon CT Slice [pic] Slice 103 [pic] Slice 104 [pic] Slice 105

Figure SS.3-2: Colon radiograph as Described in Example 2

The content tree structure in this example is complex. Structural illustrations of portions of the content tree are placed within the content tree table to show the relationships of data within the tree. Some content items are duplicated (and shown in boldface) to facilitate use of the diagrams.

[pic]

Figure SS.3-3: Content Tree Root of Example 2 Content Tree

The content tree structure would resemble:

Node Code Meaning of Concept Name Code Meaning or Example Value TID
1 Colon CAD Report 4120
1.1 Language of Content Item and Descendants English 1204
1.2 Image Set Properties 4122
1.3 CAD Processing and Findings Summary All algorithms succeeded; with findings 4121
1.4 Summary of Detections Succeeded 4120
1.5 Summary of Analyses Not Attempted 4120

Node Code Meaning of Concept Name Code Meaning or Example Value TID
1.2 Image Set Properties 4122
1.2.1 Frame of Reference UID 1.2.840.114191.1122 4122
1.2.2 Study Instance UID 1.2.840.114191.3344 4122
1.2.3 Study Date 20070924 4122
1.2.4 Study Time 090807 4122
1.2.5 Modality CT 4122
1.2.6 Horizontal Pixel Spacing 0.80 mm 4122
1.2.7 Vertical Pixel Spacing 0.80 mm 4122
1.2.8 Slice Thickness 2.5 mm 4122
1.2.9 Spacing between slices 1.5 mm 4122
1.2.10 Recumbent Patient Position with respect to gravity Prone 4122

[pic]

Figure SS.3-4: CAD Processing and Findings Summary Portion of Example 2 Content Tree

Node Code Meaning of Concept Name Code Meaning or Example Value TID
1.3 CAD Processing and Findings Summary All algorithms succeeded; with findings 4121
1.3.1 Composite Feature Polyp 4125
1.3.1.1 Rendering Intent Presentation Required:… 4125
1.3.1.2 Algorithm Name “Colon Polyp Detector” 4019
1.3.1.3 Algorithm Version “V1.3 4019
1.3.1.4 Composite Type Target content items are related spatially 4126
1.3.1.5 Scope of Feature Feature detected on multiple images 4126
1.3.1.6 Center SCOORD3D POINT 4129
1.3.1.7 Outline SCOORD3D ELLIPSOID 4129
1.3.1.8 Associated Morphology Pedunculated 4128
1.3.1.9 Diameter 20 mm 1406
1.3.1.9.1 Path SCOORD3D POLYLINE 1406

[pic]

Figure SS.3-5: Summary of Detections Portion of Example 2 Content Tree

Node Code Meaning of Concept Name Code Meaning or Example Value TID
1.4 Summary of Detections Succeeded 4120
1.4.1 Successful Detections 4015
1.4.1.1 Detection Performed Polyp 4017
1.4.1.1.1 Algorithm Name “Colon Polyp Detector” 4019
1.4.1.1.2 Algorithm Version “V1.3 4019
1.4.1.1.3 Series Instance UID 1.2.840.114191.111222 4017

Node Code Meaning of Concept Name Code Meaning or Example Value TID
1.5 Summary of Analyses Not Attempted 4120

SS.3.3 Example 3: Colon Polyp Detection, Temporal Differencing with Findings

The patient in Example 2 returns for another colon radiograph. A more comprehensive colon CAD device processes the current colon radiograph, and analyses are performed that determine some temporally related content items for Composite Features. Portions of the prior colon CAD report (Example 2) are incorporated into this report. In the current colon radiograph the colon polyp has increased in size.

PRIOR COLON CT SLICE [pic]
CURRENT COLON CT SLICE [pic]

Figure SS.3-7: Colon radiographs as Described in Example 3

Node Code Meaning of Concept Name Code Meaning or Example Value TID
1 Colon CAD Report 4120
1.1 Language of Content Item and Descendants English 1204
1.2 Image Set Properties 4122
1.2.1 Frame of Reference UID 1.2.840.114191.5577 4122
1.2.2 Study Instance UID 1.2.840.114191.7788 4122
1.2.3 Study Date 20080924 4122
1.2.4 Study Time 101827 4122
1.2.5 Modality CT 4122
1.2.6 Horizontal Pixel Spacing 0.80 mm 4122
1.2.7 Vertical Pixel Spacing 0.80 mm 4122
1.2.8 Slice Thickness 2.5 mm 4122
1.2.9 Spacing between slices 1.5 mm 4122
1.2.10 Recumbent Patient Position with respect to gravity Prone 4122
1.3 Image Set Properties 4122
1.3.1 Frame of Reference UID 1.2.840.114191.1122 4122
1.3.2 Study Instance UID 1.2.840.114191.3344 4122
1.3.3 Study Date 20070924 4122
1.3.4 Study Time 090807 4122
1.3.5 Modality CT 4122
1.3.6 Horizontal Pixel Spacing 0.80 mm 4122
1.3.7 Vertical Pixel Spacing 0.80 mm 4122
1.3.8 Slice Thickness 2.5 mm 4122
1.3.9 Spacing between slices 1.5 mm 4122
1.3.10 Recumbent Patient Position with respect to gravity Prone 4122

The CAD processing and findings consist of one composite feature, comprised of single image findings, one from each year. The temporal relationship allows a quantitative temporal difference to be calculated:

Node Code Meaning of Concept Name Code Meaning or Example Value TID
1.4 CAD Processing and Findings Summary All algorithms succeeded; with findings 4121
1.4.1 Composite Feature Polyp 4125
1.4.1.1 Rendering Intent Presentation Required: … 4125
1.4.1.2 Algorithm Name “Polyp Change” 4019
1.4.1.3 Algorithm Version “V2.3 4019
1.4.1.4 Composite Type Target content items are related temporally 4126
1.4.1.5 Scope of Feature Feature detected on multiple images 4126
1.4.1.6 Certainty of Feature 85% 4126
1.4.1.7 Associated Morphology Pedunculated 4128
1.4.1.8 Difference in size 2 mm 4126
1.4.1.8.1 Reference to Node 1.4.1.9.10 4126
1.4.1.8.2 Reference to Node 1.4.1.10.10 4126
1.4.1.9 Composite Feature Polyp 4125
1.4.1.9.1 Rendering Intent Presentation Required: … 4125
1.4.1.9.2 Tracking Identifier “Watchlist #1” 4108
1.4.1.9.3 Algorithm Name “Colon Polyp Detector” 4019
1.4.1.9.4 Algorithm Version “V1.3 4019
1.4.1.9.5 Composite Type Target content items are related spatially 4126
1.4.1.9.6 Scope of Feature Feature detected on multiple images 4126
1.4.1.9.7 Center SCOORD3D POINT 4129
1.4.1.9.8 Outline SCOORD3D ELLIPSE 4129
1.4.1.9.9 Associated Morphology Pedunculated 4128
1.4.1.9.10 Diameter 4 mm 1406
1.4.1.9.10.1 Path SCOORD3D POLYLINE 1406
1.4.1.10 Composite Feature Polyp 4125
1.4.1.10.1 Rendering Intent Presentation Required: … 4125
1.4.1.10.2 [Observation Context content items] 4022
1.4.1.10.3 Algorithm Name “Colon Polyp Detector” 4019
1.4.1.10.4 Algorithm Version “V1.3 4019
1.4.1.10.5 Composite Type Target content items are related spatially 4126
1.4.1.10.6 Scope of Feature Feature detected on multiple images 4126
1.4.1.10.7 Center SCOORD3D POINT 4129
1.4.1.10.8 Outline SCOORD3D ELLIPSE 4129
1.4.1.10.9 Associated Morphology Pedunculated 4128
1.4.1.10.10 Diameter 2 mm 1406
1.4.1.10.10.1 Path SCOORD3D POLYLINE 1406
1.5 Summary of Detections Succeeded 4120
1.5.1 Successful Detections 4015
1.5.1.1 Detection Performed Polyp 4017
1.5.1.1.1 Algorithm Name “Colon Polyp Detector” 4019
1.5.1.1.2 Algorithm Version “V1.3 4019
1.5.1.1.3 Series Instance UID 1.2.840.114191.555666 4017
1.6 Summary of Analyses Succeeded 4120
1.6.1 Successful Analyses 4016
1.6.1.1 Analysis Performed “Temporal correlation” 4018
1.6.1.1.1 Algorithm Name “Polyp Change” 4019
1.6.1.1.2 Algorithm Version “V2.3 4019
1.6.1.1.3 Series Instance UID 1.2.840.114191.111222 4018
1.6.1.1.4 Series Instance UID 1.2.840.114191.555666 4018