The knee was used as a starting point and reference for the remainder of experimentation, so the importance of getting these techniques correct was heightened. To get the starting values, an earlier experiment was referenced (from a week preceding this experiment), in which Bontrager was consulted to see that the appropriate kVp range for a knee is 65-80. 70 kVp was selected, with a mAs of 10. The phantom was positioned and an image was taken, resulting in a DI of 4.8. The calculations and experimentation in that experiment lead to a technique of 60 kVp and 8 mAs with a resulting DI of 0. This technique was set as the starting point and tested on the phantom. This resulted in a DI of -1.1. This discrepancy could have been due to different phantoms, different x-ray tubes, different consoles, or any other variable between experiments. These variables were controlled in the following experiment by using the same phantom for each image, the same room, and the same equipment (including tube, console, FPD). Another thing that could cause discrepancies was the positioning of the part. After the first image, another step was added to make the positioning, centering, FPD orientation, and collimation as close to perfect as possible so that each subsequent image would result in a DI value that was consistently based only on exposure factors, and not varying anatomy in the field of view. After some readjustment to make the positioning correct, another image was taken without adjusting the technique. The image was taken at 60 kVp and 8 mAs using the small focal spot, and resulted in a DI value of -0.2. This was not ideal for two reasons. First, as the knee was going to be used as a starting point, it needed to have a DI closer to 0. Second, the appropriate kVp for a knee is 65-80 (Bontrager), so it needed to have more penetration. An increase of 1 kVp was sufficient to move the DI to 0.1, then the 15% rule (an increase in kVp of 15% is equal to a 100% increase in exposure, allowing a decrease in mAs of 50%) was applied to bring up the kVp.

15% x 60 = 69, 8 / 2 = 4

 The resulting image was taken at 69 kVp and 4 mAs, which produced a DI of -0.5. The mAs was increased by one station to 4.5 and another image was taken. This resulted in a DI of 0. These techniques were used to take 2 more images (DI’s 0.1 and 0) so that the DI values could be averaged to ensure that the results were consistent. The final technique for the AP knee was set as 69 kVp at 4.5 mAs using the small focal spot, with an average DI of 0.033.

According to proportional anatomy, a knee requires twice the technique of an ankle, and a foot is half of an AP ankle (Quinn). While AP ankle was not in the goal of this experiment, it provides a bridge to the technique for the foot, which is one of the goals and therefore a necessary step. 

To get the starting values for AP Ankle, the mAs from AP knee was halved. This experiment seeks to produce a fixed kVp technique chart, so mAs will be adjusted first and primarily.

4.5 / 2 = 2.25

The actual technique set is subject to the mAs and kVp stations available on the console, so the first technique used to capture an image was 69 kVp and 2.2 mAs. This resulted in a DI of -1.3. This was close to ideal, but still needed to be adjusted to get to +/- 0.5, inaccurate positioning needed to be corrected, and Bontrager consulted to get kVp into the correct range. Positioning was corrected and mAs was increased by 25% (each DI value is roughly equivalent to an increase in exposure by 25% or a decrease by 20%).

2.2 x 125% = 2.75 

The closest available station was 2.8 mAs, so the image was taken at 69 kVp and 2.8 mAs, resulting in a DI of -0.5. At low mAs stations, it is hard to make minor adjustments but even one station change can increase or decrease the exposure by an entire DI value. To make it easier to fine tune techniques the 15% rule was applied, decreasing the kVp to 59 and increasing the mAs to 5.6.

69 x 85% (100% - 15%) = 58.65,    2.8 x 2 = 5.6

This resulted in a DI of -0.4, which was expected because the 15% rule should not change the exposure by much. After some readjustment in positioning to get closer to an ideal image, an x-ray was taken at the same factors (59 kVp and 5.6 mAs), resulting in a DI of 0.3. The appropriate range for AP ankle for kVp is 60-75 (Bontrager), so the 15% rule was applied to increase the kVp back to an appropriate level.

59 x 115% = 67.85,    5.6 / 2 = 2.8

An image was taken at the closest stations to these values, at 68 kVp and 2.8 mAs, resulting in a DI of -0.1. Two more images were taken for consistency, both with a DI of -0.1. The final technique for AP ankle was 68 kVp and 2.8 mAs using the small focal spot with an average DI of -0.1.

According to proportional anatomy, the shoulder is equal to a knee plus 8 kVp (Quinn). The knee technique used as a reference throughout this experiment is 69 kvp at 4.5 mAs.

69 + 8 = 77 kVp

The starting values therefore, are 77 kVp and 4.5 mAs. The image was taken and resulted in a DI of 0. After adjusting for proper centering and positioning, an image was taken again and resulted in a DI of 0.2. The range for kVp for AP shoulder is 70-85 (Bontrager), which shows that the experimental values are appropriate. The image was taken two more times for consistency, yielding DI values of 0.2 both times. The final technique for AP shoulder was 77 kVp and 4.5 mAs using the small focal spot, with an average DI of 0.2.

To get a PA chest using a phantom, a decubitus position was chosen to avoid having to move and position a heavy phantom on and off the table. The phantom was positioned on its left side and supported by positioning sponges. A 6:1 gridcap was placed over the FPD with a 103 lpi frequency and a 40-72 inch focal distance. The IR was placed in a holder and positioned against the phantom’s chest. A PA chest is proportionally roughly equivalent to an AP shoulder (Quinn), so the shoulder techniques were used for the starting image. 77 kVp and 4.5 mAs resulted in a chest image with a DI of -2. After recentering and adjusting positioning and collimation to get closer to a consistent and ideal image, the kVp was increased 3.75% (an increase in kVp by 3.75% or a decrease by 3% is roughly equal to 1 DI value). For the chest, kVp was increased first because of the much longer scale contrast necessary for chest imaging.

77 x 103.75% = 79.89 kVp

The next image was taken at 80 kVp and 4.5 mAs, resulting in a DI of -1.6. With one final field of view adjustment (centering and collimation), an image was taken with the same technique and produced a DI of -0.9. From here, mAs was increased 25% (approximately 1 DI value).

4.5 x 125% = 5.625 mAs

The closest station available on the console was 5.6, so the image taken at 80 kVp and 5.6 mAs resulted in a DI of 0.3. This was an acceptable deviation index value, but the kVp was still much too low for chest imaging (110-125 kVp (Bontrager)). To account for this, the 15% rule (described in AP Knee) was applied.

80 x 115% = 92,     5.6 / 2 = 2.8

The resulting image taken at 92 kVp and 2.8 mAs had a DI of -0.3, which is still acceptable. The same process was followed again.

92 x 115% = 105.8,     2.8 / 2 = 1.4

The resulting image taken at 106 kVp and 1.6 mAs (closest available station to 1.4) had a DI of -0.7, which is not acceptable. The same process would result in a mAs value of less than 1, which can be inconsistent and is outside the scope of this experiment. Therefore, as the DI was almost 1 unit underexposed, the kVp was increased by 3.75% once again without altering the mAs. This had the dual purpose of continuing to increase kVp and therefore contrast scale, while also hopefully increasing exposure to an appropriate level.

106 x 103.75% = 109.975

The next image at 110 kVp and 1.6 mAs resulted in a DI of -0.2. This image fulfilled both the correct kVp range for chest imaging, and appropriate exposure. This technique was used to capture two more images for consistency, producing DI values of -0.1 and -0.2. The final technique for a decubitus position PA chest was 110 kVp and 1.6 mAs using the large focal spot, with an average DI of -0.167.

Ribs are proportionally exactly the same as chest, as they both image the same area. The positioning, however, is different and must be accounted for. The technique for PA chest must be converted from the standard 72 inches, to the required 40 inches for rib imaging. The kVp must then also be brought down significantly to be in the appropriate range to produce the correct contrast scale for the ribs. First, because the intensity of the beam is inversely proportional to the square of the distance from the source to the image, the following equation was used to calculate the new mAs needed at a closer distance.

mAs2 = mAs1 (SID2 / SID1)2

mAs2 = 1.6 (40 / 72)2

mAs2 = 0.4938

The 15% rule was then applied twice (discussed in AP knee) to shift the mAs up and the kVp down. For the experiment one goal was to keep the mAs from dropping below 1 where possible, and the kVp for ribs should be between 75 and 85 (Bontrager).

110 x 85% = 93.5,     0.4938 x 2 = 0.9876

93.5 x 85% = 79.475,    0.9876 x 2 = 1.9752

79 kVp at 2 mAs were used as the starting technique and applied to the phantom in the AP oblique, LPO position, for the ribs above the diaphragm. The resulting DI value was -7.8. This extreme underexposure was not unexpected as the technique was obtained through many approximations (decubitus PA chest may not have been at exactly 72 inches, a 6:1 grid was used for PA chest, while the table bucky with a 10:1 grid was used for ribs, the 15% rule is an approximation as kVp and mAs affect exposure differently, and the 15% rule was applied multiple times before the base exposure was established because this was a preliminary estimation). 1 DI value is equal to approximately a 25% increase or a 20% decrease in exposure, so the mAs was increased by 25% eight times to approximate an 8 point shift upwards in DI value.

2 x 125% = 2.5 x 125% = 3.125 x 125% = 3.90625 x 125% = 4.88 x 125% = 6.1 x 125% = 7.63 x 125% = 9.54 x 125% = 11.92

As the 20-25% exposure is an estimate, an image was taken at 6.3 mAs (referencing the 6.1 after 5 increases and using the nearest station), which produced a DI of -1.3. This only required 1 DI increase, so 6.3 was increased by 25%

6.3 x 125% = 7.875

The nearest station available was 8, so an image was taken with 79 kVp and 8 mAs, resulting in a DI of -0.1. This was a near perfect exposure and an acceptable kVp for rib imaging, so two more images were taken using this technique for consistency, resulting in DI values of 0.3 and 0.3. The final technique for oblique ribs was 79 kVp and 8 mAs using the large focal spot, with an average DI of 0.167.

A lateral chest is proportionally 3-4 times an AP/PA chest (Quinn). To continue obtaining images using a heavy phantom on the x-ray table, cross-table lateral was used with the same 6:1 grid cap as used for the PA chest. The technique used for the PA chest was 110 kVp and 1.6 mAs. The mAs was multiplied by 4 to get our starting technique for the cross-table lateral.

1.6 x 4 = 6.4

The first image was taken at 110 kVp and 6.3 mAs (closest available station to 6.4), resulting in a DI of 3.9. To drop this value by four standard deviations (DI values), the mAs was decreased by 20% four times.

6.3 x 80% = 5.04 x 80% = 4.03 x 80% = 3.23 x 80% = 2.58

An image was taken at 110 kVp and 2.8 mAs (nearest station to 2.6), resulting in a DI value of 0.7. A full standard deviation decrease was not necessary, so the mAs was decreased by 15% rather than the full 20%.

2.8 x 80% = 2.38

An image was taken at 110 kVp and 2.5 mAs (nearest station to 2.38), resulting in a DI of 0.1. Two more images were taken for consistency, both with DI values of 0.1. The final technique for lateral chest was 110 kVp and 2.5 mAs using the large focal spot, with an average DI of 0.1. The difference in starting values from PA chest and the final values for lateral chest were much smaller than the 3-4 times suggested in proportional anatomy, but this could be due to the specific habitus of the phantom used.

A clavicle is listed in “Group 2” of proportional anatomy in Quinns work on the subject, and is compared to an AP shoulder and a scapula. Using this relationship, the techniques for AP shoulder were used as a starting point for a PA axial clavicle. The first image was taken at 77 kVp and 4.5 mAs, resulting in a DI of -5.3. The mAs was increased by 25% five times to shift the deviation index up by five standard deviations.

4.5 x 125% = 5.625 x 125% = 7.03 x 125% = 8.79 x 125% = 10.99 x 125% = 13.73

The next image was taken at 77kVp and 14 mAs (nearest available station to 13.73), resulting in a DI of 0. While this registered as a perfect exposure, 14 mAs is much too high for a clavicle and is needlessly adding dose to the patient. The suggested range for a AP/PA axial clavicle is 70-85 (Bontrager), so while there was not a lot of increasing or decreasing possible, there were several little adjustments that could be made to maintain an appropriate contrast scale while also minimizing patient dose. The comparison of shoulder to clavicle did not take into consideration the change in SID that would be caused by the angle on the x-ray tube. To ameliorate this problem, the tube was adjusted down by 5 inches, to compensate for the 25% caudal angle on the central ray. The image was taken again using the first set of techniques (77 kVp and 4.5 mAs), resulting in a DI of -3.9. This was still significantly underexposed, but would not need as drastic of an increase in mAs as did the -5.3 from before the SID was adjusted. The mAs was increased four times (see math above), to 10.99 and another image was taken at 77 kVp and 11 mAs, resulting in a DI of 0.2. This would be an ideal time to apply the 15% rule ton increase kVp and lower mAs, but an increase of the full 15% would drive the kVp above the acceptable range for clavicle. So, instead of 15%, a 7.5% increase in kVp was applied, with a joint decrease in mAs of 25% rather than the full 50%.

77 x 107.5% = 82.78,     11 x 75% = 8.25

The next image was taken with 83 kVp and 8 mAs, resulting in a DI of 0.3. The image was taken two more times for consistency, resulting in DI values of 0.3 and 0.4. The final technique for PA axial Clavicle was 83 kVp and 8 mAs using the small focal spot, with an average DI of 0.33.

A foot is roughly equal to one half of an AP ankle. Generally this is achieved by using 8 less kVp rather than adjusting the mAs (Quinn). The ankle technique achieved in this experiment was 68 kVp and 2.8 mAs. For this reason, the starting technique selected for AP foot was 60 kVp and 2.8 mAs. The resulting image had a Deviation Index of 1.8. The mAs was decreased by 20% (approximately 1 DI value) twice to lower the exposure to an appropriate value.

2.8 x 80% = 2.24 x 80% = 1.79

The next image was taken at 60 kVp and 1.8 mAs. The resulting DI was 2.3. This value did not make sense as the exposure factors were lower, but the console was registering an increase in exposure on the image. The factors were reset to the original values and another exposure was taken, yielding the same DI of 1.8. The mAs was again lowered to 1.8 and again produced a DI of 2.3. The system was restarted and the foot was repositioned to achieve as close to a perfect image as possible. The malfunction in the system continued to occur, producing increasingly high DI values, the last of which was 3 when an image was taken with 1.8 mAs. Possible solutions to this error were considered, but limited by the facts that this experiment would only be consistent if done and repeatable on the same console with the same phantom and same computational systems. Finally, it was decided to adjust the kVp as the system reacts differently to different energy levels applied to the IR. The mAs was lowered again by 20%, this time three times to account for the DI of 3, then the 15% rule (discussed in AP Knee) was applied twice to get to a higher mAs (further away from 1 as x-ray systems can be unreliable at very small mAs values).

1.8 x 80% = 1.44 x 80% = 1.52 x 80% = 0.92

Note: While the appropriate estimation for moving exposure down by one standard deviation is a 20% decrease in exposure, an error was made at this point and the actual decrease applied was 25% as shown below.

1.8 x 75% = 1.35 x 75% = 1.0125 x 75% = 0.759

The resulting technique before application of the 15% rule was 60 kVp and 0.76 mAs.

60 x 85% = 51,       0.76 x 2 = 1.52

51 x 85% = 43.35,     1.52 x 2 = 3.04

The resulting technique for the next exposure was 43 kVp and 3.2 mAs (nearest available station to 3). The resulting DI was -1. At this point the kVp was outside the appropriate range for a foot (55-65 (Bontrager)), but the system algorithms needed to be tested and attuned to, then the kVp could be brought back up to an acceptable range while maintaining correct exposure. The mAs was increased by 25% (approximately 1 DI value), and another exposure was taken.

3.2 x 125% = 4

The next image was taken at 43 kVp and 4 mAs and resulted in a DI of 0.1. Now that the system was responding appropriately to changes in mAs, the kVp could be increased while maintaining approximately the same exposure using the 15% rule.

43 x 115% = 49.45,      4 / 2 = 2

50 kVp approached the correct range, but a full 15% increase in kVp and halving of mAs would bring the mAs to 1, which was hypothesized to be one of the causal variables that created errors earlier in the experiment. Because of this, the kVp was only increased by 7.5%, and the mAs decreased by 25%.

49.45 x 107.5% = 53.16,      2 x 75% = 1.5

The resulting technique used for the next image was 53 kVp and 1.4 mAs (nearest available station to 1.5). This produced a DI of -0.5. The system continued to respond as expected, so the same 7.5% increase was applied again.

53 x 107.5% = 56.975,   1.4 x 75% = 1.05

The next image was taken at 57 kVp and 1 mAs, yielding a DI value of -0.7. The mAs was increased by 25% to increase exposure.

1 x 125% = 1.25

The final image was taken at 57 kVp and .25 mAs, yielding a DI of 0.4. This technique was inside acceptable range and also produced correct exposure. The same technique was used to take two more images for consistency, producing DI values of 0.4 and 0.3. The final technique for AP foot was 57 kVp and 1.25 mAs using the small focal spot, with an average DI of 0.37.

Getting from foot to hand is slightly more complicated than some other parts, but Quinn states that a wrist is equal to half of an elbow, which is equal to half of a foot, which is equal to 1.5 times a hand. If only part of this association is used, it can be stated that half of a foot is equal to 1.5 times a hand. This can be stated as follows.

(1 / 2) foot = 1.5 x hand

Substituting the mAs from hand, it is now stated:

(1 / 2) (1.25) = 1.5X

X = 0.42

This would make the technique 57 kVp and 0.42 mAs. Ideally, the mAs should be closer to or above 1, so the 15% rule was applied (discussed in AP Knee).

57 x 85% = 48.45,    0.42 x 2 = 0.84

Note: At this point, an error was made in which the kVp was increased rather than decreased by the appropriate 15% as shown below.

57 x 115% = 65.55

The first image was taken at 66 kVp and 0.8 mAs. The resulting DI was 3.6. This highly overexposed image could be expected because of the error causing a 15% increase in kVp, which doubles exposure, while also doubling mAs. To account for the overexposure, the mAs was decreased by 20% (roughly equivalent to 1 DI value) four times.

0.8 x 80% = 0.64 x 80% = 0.512 x 80% = 0.41 x 80% = 0.328

The 15% rule was applied once again to avoid using a mAs value as low as 0.33

66 x 85% = 56.1,    0.33 x 2 = 0.66

The next image was taken with a kVp of 56 and a mAs of 0.63 (nearest available station to 0.66). The resulting DI was -0.5. This value indicates an acceptable exposure, and the kVp is within the appropriate range of 55-65 (Bontrager). The image was taken two more times for consistency yielding DI values of -0.4 both times. The final technique for PA hand was 56 kVp and 0.63 mAs using the small focal spot, with an average DI of -0.43.

A femur is equal to two thirds of an abdomen (Quinn). This can be written as follows:

Femur = (2 / 3) abdomen

Substituting the obtained mAs value for femur, the expression can be simplified as shown below.

7.1 = (2 / 3) X

10.65 = X

The starting image was taken at 77 kVp and 11 mAs (nearest available station to 10.65), resulting in a DI of -1.4. Positioning was corrected and the mAs was increased by 25% (approximately 1 DI value).

11 x 125% = 13.75

The next image was taken at 77 kVp and 14 mAs (nearest available station to 13.75), resulting in a DI of -0.3. 14 mAs is higher than necessary for patient dose, but 77 kVp is firmly between the suggested range of 70-85 (Bontrager), which does not leave much room for drastic adjustments. For this reason, a 7.5% increase in kVp was applied with an associated 25% decrease in mAs.

77 x 107.5% = 82.775,     14 x 75% = 10.5

An exposure was taken at 82 kVp and 11 mAs, resulting in a DI of 0.1, indicating a near perfect exposure. 82 is still within the appropriate range for abdomen imaging, so the image was taken twice more for consistency, producing DI values of 0 and 0.1. The final technique for AP abdomen was 82 kVp and 11 mAs using the small focal spot, with an average DI of 0.07.

While not directly within the scope of this experiment, finding the appropriate technique for the femur can lead to the technique for the abdomen, which will be used in the discovery of other techniques that are within the scope of this experiment. Therefore, finding the correct factors for femur became imperative for the completion of this project. The femur is approximately equal to the shoulder (Quinn), so the starting technique was obtained from earlier data. The first image was taken at 77 kVp and 4.5 mAs. Because this technique would be used to find the technique for AP abdomen, the proximal femur was selected for its similarity to the density of the abdomen. The resulting DI value was -1. The positioning and centering were modified to get closer to an ideal image and eliminate positioning as a variable. The mAs was increased by 25% (approximately 1  DI value).

4.5 x 125% = 5.625

With the adjustment in positioning and mAs, the resulting DI was -1.1. Because of the change in positioning, this minor change was not unexpected. Changing positioning changes what is in the field of view and therefore how much x-ray energy is attenuated and how much passes through to be read by the IR. Now that the positioning was correct however, the factors were the only variable left to be adjusted to achieve an ideal image. The mAs was increased by 25% once again.

5.6 x 125% = 7.03

An image was taken at 77 kVp and 7.1 mAs (nearest available station to 7), and the resulting DI was 0, indicating perfect exposure. The kVp range was inside the suggested 75-85 (Bontrager). The image was taken twice more for consistency, yielding DI values of 0.1 and 0. The final technique for AP proximal femur was 77 kVp and 7.1 mAs using the small focal spot, with an average DI of 0.03.

Quinn places AP hip and AP abdomen together in “Group 1” of proportional anatomy, suggesting similar densities. For this reason, the starting values for AP hip were directly taken from the data on AP abdomen. The first image was taken at 82 kVp and 11 mAs, which produced a DI of -0.7. Usually an increase of 25% in exposure is roughly equivalent to an increase in DI value of one standard deviation. In this case, a full unit increase was not necessary, so an increase of 20% was applied to the mAs.

11 x 120% = 13.2

An image was taken with 82 kVp and 14 mAs (nearest available station to 13.2), yielding a DI of 0.2. The kVp was in the appropriate range of 80-85 (Bontrager), and the DI indicated correct exposure, so two more images were taken with this technique for consistency. The resulting DI values were 0.2 and 0.2. The final technique for AP hip was 82 kVp and 14 mAs using the small focal spot, with an average DI of 0.2.

AP L spine is approximately equivalent to an AP hip (Quinn). The starting values used for this image were taken from the AP hip data, and the first image was taken at 82 kVp and 14 mAs. This produced a DI value of -0.7. The image was taken again after adjusting positioning and it produced a DI of -1.1. The mAs was increased by 25% (approximately 1 DI value).

14 x 125% = 17.5

An image was taken at 82 kVp and 18 mAs (nearest available station to 17.5), yielding a DI of 0. Whenever possible, it is important to keep the mAs low to keep patient dose as low as possible. The accepted kVp range for AP L spine is 75-90 (Bontrager), so it was possible to apply a 7.5% increase to the kVp with a respective 25% decrease in mAs.

82 x 1.075 = 88.15,      18 x 75% = 13.5

An image was taken at 88 kVp and 14 mAs, resulting in a DI of 0.4. This was very close to an ideal image, but not far enough away to merit a full 20-25% change in mAs. Considering this, the mAs was decreased by one station on the console, resulting in a mAs of 12.5. Another image was taken, yielding a DI of -0.1. Two more images were taken using the same technique for consistency, producing DI values of -0.1 and 0. The final technique for AP lumbar spine was 88 kVp and 12.5 mAs using the small focal spot, with an average DI of -0.07.

A PA Caldwell is roughly equal to two thirds of an abdomen (Quinn). This can be written and simplified as shown below, with substitution of the mAs values obtained earlier in the experiment for an AP abdomen.

Caldwell = (2 / 3) abdomen

X = (2 / 3) (11)

X = 7.33

The first image was taken at 82 kVp and 7.1 mAs (nearest available station to 7.3), resulting in a DI of -4.7. The lack of flexibility in the phantom as well as its weight and instability in movement prompted a supine, prone position for the phantom, with the hips elevated to place the orbitomeatal line as close to perpendicular to the table as possible, with a CR angle of greater than 30 degrees to compensate for the difference. Because of the unorthodox positioning, several adjustments to angulation and position were required before the image was correct enough to eliminate positioning as a variable. Once proper position, CR angulation, and centering were achieved, the DI value was -8.5. The appropriate range for a PA axial skull is 75-85 (Bontrager). The kVp was already very close to the upper limit, so it was only increased by 7.5%. Without an associated decrease in mAs, this should be roughly equivalent to a 2 DI increase in exposure. To increase by the remaining 6.5 DI values, the mAs should be increased by 25% at least 6 times. However, when applying these conversions many times, their accuracy is not always maintained as each conversion is an approximation. For this reason, the mAs was increased only three times to start with, leaving remaining adjustments to be made after an experimental image.

82 kVp x 107.5% = 88.15

7.1 x 125% = 8.875 x 125% = 11.09 x 125% = 13.87

Another image was taken at 88 kVp and 14 mAs, resulting in a DI of -4. The kVp was increased by 7.5% again (roughly equivalent to 2 DI values) and the mAs was increased by 25% (roughly equivalent to 1 DI value).

88 x 107.5% = 94.6,      14 x 125% = 17.5

An image was taken at 95 kVp and 18 mAs, producing a DI of -1.4. The conversions accounted for approximately a 3 DI increase, so this value was not unexpected. Another 3.75% increase was applied to the kVp to account for one more standard deviation in exposure.

95 x 103.75% = 98.56

An image was taken at 99 kVp and 18 mAs, resulting in a DI of -0.6. At this point, the kVp is outside the accepted range for PA axial skull, but decreasing it would increase the dose to the patient by too much. However, further increases to kVp would change the contrast scale and penetration of the beam by too high a factor to continue. Therefore, the kVp was left at 99, and mAs was increased by 15% (A full 25% was not needed as a perfect image was only 0.6 deviations away)

18 x 115% = 20.7

An image was taken at 99 kVp and 20 mAs, resulting in a DI of -0.1. After these many adjustments and changes, all of which took time, the positioning of the phantom had changed slightly. Another image was taken with the same factors after readjusting the phantom positioning, and a DI of -0.4 was attained. The image was taken twice more for consistency, yielding DI values of -0.5 and -0.4. The final technique for PA axial skull was 99 kVp and 20 mAs using the small focal spot, with an average DI of -0.43.