Influence of Scan Field of View on Operator Radiation Exposure in Computed Tomography Fluoroscopy

Abstract

Purpose:

This study aimed to evaluate the effect of the scan field of view on operator radiation exposure during computed tomography fluoroscopy in computed tomography-guided interventional radiology procedures.

Material and Methods:

Two cylindrical phantoms (diameters: 220 mm and 330 mm) were scanned, and the tube currents were adjusted to achieve equivalent standard deviation values for scan field of view M (320 mm) and scan field of view L (500 mm), where standard deviation refers to the standard deviation of computed tomography attenuation values measured within a centrally placed square region of interest in the phantom image (expressed in Hounsfield units). Equivalent doses at the operator's position were measured under these standard deviation-equivalent conditions.

Results:

The results showed that the equivalent doses at the operator's position were significantly lower when using scan field of view M. The difference in radiation exposure was more pronounced with the larger phantom and increased by up to 15% when using the larger scan field of view (L), with scan field of view M as the reference.

Conclusions:

These findings suggest that using a smaller scan field of view can effectively reduce radiation exposure to the operator when a wide field of view is unnecessary for computed tomography fluoroscopy.

Introduction

CT fluoroscopy (CTF) is widely used in interventional radiology (IR) and has been shown to improve procedural safety and efficiency compared with conventional intermittent CT-guided IR [1-3]. However, these benefits are accompanied by increased radiation exposure to both the patient and the operator, which is a particular concern in cases with prolonged CTF duration and in settings with a limited number of trained personnel [1-3].

The International Commission on Radiological Protection (ICRP) Publication 118 lowered the threshold absorbed dose to the lens from 150 mSv/year to a five-year average of 20 mSv/year (with a maximum of 50 mSv/year), and optimization of protection against lens radiation exposure is required [4]. In addition to personal protective equipment, system-based dose-reduction techniques such as half-scan reconstruction [5], angular beam modulation [6], and iterative reconstruction [7] have been proposed, although their availability varies among CT systems.

The scan field of view (SFOV) is typically selected according to patient size in routine CT imaging. A larger SFOV broadens the X-Y beam width, and the choice of SFOV may therefore influence the amount of scattered radiation during CTF. Nevertheless, no previous studies have evaluated how SFOV selection affects operator radiation exposure.

This study aimed to determine the influence of the SFOV on operator radiation exposure during CTF examinations.

Material and Methods

Experimental procedures

This study consisted of two sequential experimental components. First, we quantified image noise at each SFOV to determine tube current settings that produced equivalent image quality, ensuring that differences in radiation exposure were not confounded by image noise. Second, using these noise-equivalent conditions, we measured the equivalent dose at the operator's position to directly assess the impact of SFOV on operator radiation exposure.

Evaluation of image noise at each SFOV

During standard CT imaging, automatic exposure control (AEC) adjusts the tube current according to patient size to maintain a target image standard deviation (SD), thereby optimizing radiation dose [8]. Given that AEC was not available for CTF on the CT systems at our facility, we directly measured SD at scan field of view M (SFOV M, 320 mm) and scan field of view L (SFOV L, 500 mm) to determine tube current settings that yielded equivalent SDs. Scans were performed using a 320-slice Aquilion ONE GENESIS Edition (Canon Medical Systems, Tochigi, Japan) with the following parameters: 120 kVp, 0.5 s/rotation, detector collimation 3 × 4.0 mm, display field of view 218.75 mm, and filtered back projection (kernel FC11). The CT-guided IR imaging tool on this system allows selection between continuous fluoroscopy and an intermittent (one-shot) mode; in this study, the intermittent mode was used so that SD could be measured at the same mAs per rotation. Two water-filled cylindrical acrylic phantoms were used: a small phantom with a diameter of 220 mm and a large phantom with a diameter of 330 mm (Figure 1). Each phantom was positioned at the gantry isocenter. For the small phantom, the tube current was varied from 10 to 200 mA in 10 mA increments; for the large phantom, it was varied from 60 to 250 mA in 10 mA increments. Each condition was scanned five times, yielding three images per scan (15 images per condition). In each image, a 106.8 × 106.8 mm square ROI was placed at the center of the phantom, and the SD of the CT attenuation values (expressed in Hounsfield units) within this ROI was recorded (Figure 2).

Figure 1.

Photographs of the small and large phantoms.

(A) Small phantom. (B) Large phantom.

Figure 2.

Overview of the phantom and examples of standard deviation measurements.

(A) Large phantom with a diameter of 330 mm. (B) A magnified reconstruction was performed with a display field of view of 218.75 mm. A square region of interest was placed at the center of the phantom, and the standard deviation was measured. A scale bar is provided on the right side of the figure, with 10 mm per division (tick mark).

Measurement of equivalent dose at the operator's position

An ionization chamber-type survey meter (ICS-323C; ALOKA, Tokyo, Japan) was used to measure the equivalent dose at the operator's position. The operator's position was defined as 60 cm caudal to the center of the gantry and 60 cm to the left of the bed toward the gantry. In this study, measurements were performed on the left side only, assuming bilateral symmetry in scatter distribution, and we did not measure the equivalent dose when the operator was positioned on the right side of the gantry. This assumption was based on the rotational symmetry of the gantry and the central positioning of the phantom, which are expected to generate a broadly symmetric scatter field around the isocenter under the experimental conditions. The ionization chamber-type survey meter was oriented horizontally and set at a 45° angle to the CT table and gantry. The two phantoms (from the ‘Evaluation of image noise at each SFOV’ section) were placed at the gantry isocenter, and the equivalent dose was measured 10 times at heights of 90 cm (abdomen) and 160 cm (eye) (Figure 3). Tube currents were 160/190 mA (SFOV M/L) for the small phantom and 180/210 mA (SFOV M/L) for the large phantom; the corresponding console-reported volume CT dose index (CTDIvol) values were 5.8/6.2 mGy and 6.5/6.9 mGy, respectively.

Figure 3.

Overview of the equivalent dose measurement location. (A) View from the ceiling. (B) View from the caudal side.

Statistical analysis

Statistical analyses were performed using EZR software (Saitama Jichi Medical University, Saitama, Japan) [9]. Continuous data were tested for normality using the Shapiro-Wilk test. Student's t-test and Mann-Whitney U test were performed depending on the variance assumptions. Statistical significance was set at p < 0.05.

Results

Image noise at each SFOV

The relationship between the tube current and SD at each SFOV is presented in Tables 1 and 2. For the same tube current, the SD was lower for SFOV M, regardless of the size of the phantom. The tube current-SD pairs showing nearly identical SDs were 160/190 mA (SFOV M/L; SD 8.82/8.82) for the small phantom and 180/210 mA (SFOV M/L; SD 25.04/25.07) for the large phantom; these SDs did not differ significantly (Student's t-test, p > 0.05).

Table 1.

Image Noise at Each Scan Field of View (Small Phantom).

Tube current(mA)	Standard deviation (HU)
	SFOV: M	SFOV: L
10	41.54 ± 0.26	45.87 ± 0.35
20	26.27 ± 0.17	28.90 ± 0.14
30	20.76 ± 0.13	22.76 ± 0.17
40	17.71 ± 0.09	19.49 ± 0.16
50	15.74 ± 0.09	17.28 ± 0.11
60	14.23 ± 0.12	15.64 ± 0.12
70	13.16 ± 0.09	14.45 ± 0.11
80	12.25 ± 0.08	13.42 ± 0.10
90	11.91 ± 0.12	13.02 ± 0.09
100	11.27 ± 0.07	12.37 ± 0.09
110	10.70 ± 0.07	11.74 ± 0.09
120	10.2 ± 0.09	11.20 ± 0.07
130	9.80 ± 0.06	10.72 ± 0.07
140	9.43 ± 0.08	10.31 ± 0.09
150	9.11 ± 0.06	9.97 ± 0.06
160	8.82 ± 0.05	9.65 ± 0.04
170	8.58 ± 0.07	9.37 ± 0.04
180	8.32 ± 0.06	9.09 ± 0.04
190	8.09 ± 0.05	8.82 ± 0.04
200	7.87 ± 0.06	8.61 ± 0.07

Data are mean ± standard deviation.

HU: Hounsfield unit; SFOV: M: scan field of view M (320 mm); SFOV: L: scan field of view L (500 mm)

Table 2.

Image Noise at Each Scan Field of View (Large Phantom).

Tube current (mA)	Standard deviation (HU)
	SFOV: M	SFOV: L
60	44.93 ± 0.21	49.28 ± 0.30
70	41.05 ± 0.35	44.68 ± 0.41
80	37.86 ± 0.28	41.18 ± 0.30
90	36.64 ± 0.26	39.80 ± 0.37
100	34.44 ± 0.30	37.41 ± 0.37
110	32.39 ± 0.28	35.30 ± 0.32
120	30.87 ± 0.23	33.50 ± 0.17
130	29.56 ± 0.24	32.00 ± 0.22
140	28.23 ± 0.26	30.72 ± 0.25
150	27.21 ± 0.22	29.54 ± 0.19
160	26.96 ± 0.31	29.26 ± 0.26
170	26.08 ± 0.18	28.21 ± 0.23
180	25.04 ± 0.20	27.36 ± 0.22
190	24.30 ± 0.23	26.43 ± 0.18
200	23.67 ± 0.16	25.65 ± 0.17
210	23.01 ± 0.12	25.07 ± 0.18
220	22.44 ± 0.19	24.46 ± 0.13
230	21.85 ± 0.18	23.78 ± 0.16
240	21.42 ± 0.15	23.29 ± 0.17
250	20.90 ± 0.12	22.72 ± 0.20

Data are mean ± standard deviation.

HU: Hounsfield unit; SFOV: M: scan field of view M (320 mm); SFOV: L: scan field of view L (500 mm)

Comparison of equivalent doses at the operator's position

The equivalent doses at the operator's position for each SFOV are listed in Table 3. The equivalent doses were lower with SFOV M than with SFOV L, with a more pronounced relative increase for the large phantom than for the small phantom.

Table 3.

Comparison of Equivalent Dose at the Operator’s Position.

Phantom size(height)	Dose equivalent (μSv)		Ratio (L/M)	p-Value
	SFOV: M(95% CI)	SFOV: L(95% CI)
Small (90 cm)	3.64 ± 0.05 (3.60-3.68)	3.96 ± 0.05 (3.92-4.00)	1.09	p < 0.001
Small (160 cm)	3.43 ± 0.05 (3.40-3.47)	3.74 ± 0.05 (3.70-3.78)	1.09	p < 0.001
Large (90 cm)	4.80 ± 0.07 (4.75-4.85)	5.53 ± 0.08 (5.47-5.59)	1.15	p < 0.001
Large (160 cm)	3.54 ± 0.05 (3.50-3.58)	4.07 ± 0.05 (4.04-4.11)	1.15	p < 0.001

Data are mean ± standard deviation.

CI: confidence interval; SFOV M: scan field of view M (320 mm); SFOV L: scan field of view L (500 mm)

Discussion

In this study, we investigated how SFOV selection affects operator radiation exposure during CTF. Increasing tube current reduces image noise but increases radiation exposure to both the patient and the operator. Therefore, to evaluate the effect of SFOV on operator radiation exposure, we adjusted the tube current to match SD between SFOV settings, ensuring that differences in the measured dose were not confounded by image noise. The tube currents were 160/190 mA (SFOV M/L) for the small phantom and 180/210 mA (SFOV M/L) for the large phantom. Under these SD-equivalent conditions, the SD difference was <0.2% and not statistically significant.

At our facility, when performing CT-guided IR, we review preoperative CT or MRI scans to determine patient positioning and to ensure a safe and accessible route to the lesion. During the initial planning scan, a large SFOV may be required for whole-body visualization; after the puncture site is confirmed, CTF is typically performed with a magnified view of the ROI. Given that whole-body coverage is usually unnecessary during CTF, a smaller SFOV is often sufficient for needle guidance and lesion visualization. However, SFOV selection should also preserve situational awareness: in CT-guided lung biopsy, an overly narrow SFOV may exclude the heart or central pulmonary arteries and delay recognition of complications such as air embolism. Therefore, SFOV-based dose optimization should be balanced against comprehensive intraprocedural monitoring. For the same tube current, the SD was lower for SFOV M than for SFOV L for both phantom sizes, which may reflect differences in bow-tie filter behavior. When the SFOV is increased, a larger fan angle is required, and the X-ray beam passes through more peripheral regions of the filter, where the aluminum-equivalent thickness is greater. Yang et al. [10] reported that this thickness increases toward the filter edges; therefore, wider fan angles would be expected to produce stronger attenuation and more pronounced spectral changes. In some CT systems, changing the SFOV also switches the bow-tie filter itself. Although scatter is influenced by multiple factors, these SFOV-related differences are likely to modify both its magnitude and distribution.

The equivalent dose at the operator's position was lower with SFOV M than with SFOV L, and the dose ratio differed by phantom size. In the small phantom, the subject was fully encompassed by the beam with both SFOV settings, and the primary scattered radiation from the phantom is therefore expected to be similar. In the large phantom, the wider beam width with SFOV L likely increased primary scattered radiation from the subject. Consistent with prior findings that scatter dose increases with subject size [11], this is likely the main reason for the larger equivalent-dose ratio observed with the large phantom. In addition, the relatively short phantom-to-bed distance may have increased secondary scatter from the bed and nearby gantry structures, particularly with the wider SFOV. Therefore, although SFOV-dependent differences in bow-tie filter behavior and beam width may partly account for the results, geometric coupling effects cannot be excluded.

Previous studies have demonstrated procedural advantages of CT-guided IR using CTF, including shorter procedure time and fewer complications compared with conventional CT-guided IR [1]. However, CTF is associated with increased radiation exposure to both the operator and the patient [1-3]; therefore, careful dose optimization remains essential. Based on the results of this study, radiation exposure to the operator's lens and abdomen can be reduced by changing the SFOV to a small value. In principle, this approach can be implemented by appropriately selecting a smaller SFOV on existing CT systems and does not require additional hardware installation or major software upgrades. However, the degree of flexibility in SFOV selection and fluoroscopy configuration varies among manufacturers and models; therefore, the feasibility and workflow for implementing this strategy need to be confirmed for each system.

Previous studies have also reported several CT system-based approaches to reduce radiation dose during CTF. Half-scan reconstruction has been reported to reduce surface dose by 16%-42% [5], angular beam modulation decreases operator hand scatter by approximately 27% [6], and iterative reconstruction reduces median effective dose in CTF-guided abdominal biopsy by about 50% [7]. In comparison, the SFOV-related reduction observed in our study (9%-15%) is smaller, but it represents an additional system-based method that can complement these existing dose-reduction techniques.

Kim et al. [1] reported that the mean whole-body dose to the operator per CTF-guided lung biopsy was 0.054 mSv. Applying the relative reduction observed in our phantom study (approximately 9%-15% lower equivalent dose at the operator's position with SFOV M vs SFOV L) would correspond to an estimated reduction of approximately 0.005-0.008 mSv per procedure under similar clinical conditions. In our phantom experiment, the absolute reduction in equivalent dose at eye level (160 cm) for the large phantom was 0.53 μSv per rotation under SD-equivalent conditions. In clinical CTF, however, tube current and fluoroscopy time vary by case depending on patient habitus, target location, access route, and operator technique, so the absolute benefit will also vary. Although the impact of this small per-procedure reduction is limited relative to the ICRP-recommended annual eye lens limit of 20 mSv/year [4], modest per-case decreases may become meaningful when accumulated over many procedures and years, particularly for high-volume operators. Therefore, SFOV optimization should be considered one component of a cumulative dose-reduction strategy to keep operator exposure as low as reasonably achievable.

This study primarily evaluated operator radiation exposure. However, SFOV selection also affected patient dose. The CTDIvol was 5.8 vs 6.2 mGy for the small phantom and 6.5 vs 6.9 mGy for the large phantom (SFOV M vs SFOV L), indicating a 6%-7% reduction with the smaller SFOV. With matched SD in the central ROI, lesion and device visibility should be broadly preserved as long as the target remains within the reduced field of view.

This study had several limitations. First, it was performed on a single CT system; different models or SFOV configurations may produce different operator radiation exposure and image quality, limiting generalizability. In addition, AEC for CTF was not available on the system used in this study. In scanners in which AEC is implemented for CTF, different target noise levels or control algorithms may partially compensate for SFOV-related changes by automatically adapting the tube current. Future studies should therefore evaluate how AEC-integrated CTF systems manage SFOV-dependent variations in image noise and scattered radiation. Second, we only examined operator radiation exposure with the phantom positioned centrally. In clinical practice, CT-guided IR is performed in various patient positions depending on lesion location and access route. The bed is often lowered to prevent collision between devices or the operator's hands and the gantry [3, 5], but this would be expected to alter image quality and scatter distribution because X-rays passing through the bow-tie filter would no longer irradiate the patient uniformly. Thus, this study did not account for differences related to phantom height or off-center positioning, as we focused solely on SFOV-related changes in operator radiation exposure. In addition, we assumed that exposure on the right and left sides of the gantry was equivalent and therefore measured only on the left side. Although this assumption is supported by gantry rotational symmetry and central phantom placement, minor asymmetries may still occur owing to room layout, shielding, and operator stance; future studies should include direct bilateral measurements. Third, image quality assessment was limited to SD and did not include subjective or diagnostic evaluation. Subjective evaluation would require testing the visibility of simulated targets in an anthropomorphic phantom, which was not performed. In this study, uniform water phantoms were used to set tube currents, resulting in equivalent SD between SFOVs, but the tube currents required to achieve this differed by phantom size. In anthropomorphic phantoms, matching SD at different SFOVs in CTF would be even more difficult without AEC. Moreover, simple water phantoms do not reproduce heterogeneous tissue composition (e.g., lung, bone, and soft tissue) or realistic body contours and therefore cannot fully replicate clinical scatter conditions. Fourth, although our experimental setup simulated a typical IR workflow, we did not assess clinical outcomes such as intraprocedural detection of complications, preservation of safety margins, or overall situational awareness. Given that reduced SFOV may narrow the visualized region, determining the optimal SFOV in clinical practice will require balancing operator radiation exposure reduction with procedural safety. Prospective clinical studies evaluating SFOV selection during actual biopsy, drainage, or ablation procedures are needed to clarify this balance. Finally, this study did not examine other system-based dose-reduction strategies such as half-scan reconstruction or iterative reconstruction, which may further modify operator radiation exposure; future studies should assess their combined effects with SFOV reduction.

Conclusions

Our results demonstrated that, under SD-equivalent conditions, selecting a smaller SFOV (M, 320 mm) during CTF can reduce operator radiation exposure compared with a larger SFOV (L, 500 mm). Specifically, SFOV M decreased the equivalent dose at the operator's position by approximately 9%-15%, with a more pronounced relative difference for the larger phantom. Therefore, when a wide field of view is not required for intraprocedural monitoring, selecting a smaller SFOV may be considered one practical option for reducing operator radiation exposure during CTF.

Author Contributions

All authors have reviewed the final version to be published and agreed to be accountable for all aspects of the work. Concept and design, acquisition, analysis, or interpretation of data: All authors. Drafting of the manuscript: K. Tsunou. Critical review of the manuscript for important intellectual content: All authors.

Conflicts of Interest

There are no conflicts of interest.

Clinical Registration Number

None

Data Sharing

Data supporting the findings of this study are available from the corresponding author on request.