Impact of patient habitus and acquisition protocol on iodine quantification in dual-source photon-counting computed tomography

Abstract.

Purpose

Evaluation of iodine quantification accuracy with varying iterative reconstruction level, patient habitus, and acquisition mode on a first-generation dual-source photon-counting computed tomography (PCCT) system.

Approach

A multi-energy CT phantom with and without its extension ring equipped with various iodine inserts (0.2 to 15.0 mg/ml) was scanned over a range of radiation dose levels (${\mathrm{CTDI}}_{\mathrm{vol}}$ 0.5 to 15.0 mGy) using two tube voltages (120, 140 kVp) and two different source modes (single-, dual-source). To assess the agreement between nominal and measured iodine concentrations, iodine density maps at different iterative reconstruction levels were utilized to calculate root mean square error (RMSE) and generate Bland–Altman plots by grouping radiation dose levels (ultra-low: $<1.5$; low: 1.5 to 5; medium: 5 to 15 mGy) and iodine concentrations (low: $<5$; high: 5 to 15 mg/mL).

Results

Overall, quantification of iodine concentrations was accurate and reliable even at ultra-low radiation dose levels. RMSE ranged from 0.25 to 0.37, 0.20 to 0.38, and 0.25 to 0.37 mg/ml for ultra-low, low, and medium radiation dose levels, respectively. Similarly, RMSE was stable at 0.31, 0.28, 0.33, and 0.30 mg/ml for tube voltage and source mode combinations. Ultimately, the accuracy of iodine quantification was higher for the phantom without an extension ring (RMSE 0.21 mg/mL) and did not vary across different levels of iterative reconstruction.

Conclusions

The first-generation PCCT allows for accurate iodine quantification over a wide range of iodine concentrations and radiation dose levels. Stable accuracy across iterative reconstruction levels may allow further radiation exposure reductions without affecting quantitative results.

1. Introduction

Intravenous administration of iodinated contrast during computed tomography (CT) acquisitions can improve diagnostic utility for many clinical indications by improving contrast between structures with similar attenuation on non-contrast scans. In addition to visual assessment of contrast enhancement, iodine measurements provide the ability to assess contrast uptake of various tissues, organs, and lesions quantitatively.^1–4 However, conventional, or non-spectral, CT systems are only capable of assessing the amount of iodine at best in a semi-quantitative way because highly attenuating materials, such as calcifications or bone, can have HU values in the same range as iodine.⁵ With beam hardening artifacts present in conventional CT, these materials can make it difficult for contrast-enhancing features to be differentiated from non-enhancing features.⁶

Spectral CT can overcome challenges of conventional CT, such as beam hardening or non-quantitative measurements, to accurately determine iodine levels using iodine density maps. The latest addition to the family of spectral CT instruments is photon-counting CT (PCCT).⁷ Many years of preclinical research have demonstrated the superiority of photon-counting detectors (PCDs) over energy-integrating detectors (EIDs) in image quality (high contrast-to-noise ratio, high spatial resolution, low electronic noise, and reduced artifacts), quantitative imaging, and dose efficiency.^8–14 Since PCCT has become clinically available,¹⁵ studies quantifying iodine with patient data have reinforced its superiority over EID-CT in applications like abdominal^16–18 and cardiac imaging.^19–21

Accurate iodine quantification is needed to ensure reliable clinical results. Consequently, the purpose of this study was to characterize the iodine quantification performance of a first-generation clinical dual-source PCCT system (NAEOTOM Alpha, Siemens Healthineers). A phantom containing iodine inserts of various concentrations was utilized to investigate and evaluate the effects of patient habitus, acquisition mode, radiation dose, and iterative reconstruction on iodine quantification.

2. Methods

2.1. CT Phantom

A multi-energy CT phantom (Multi-energy CT, Sun Nuclear) (Fig. 1) was used with and without an extension ring to simulate two patient sizes: 20 cm diameter (“small”) and $30\times 40\mathrm{cm}$ (“large”). The phantom contained varying concentrations of iodine (0.2, 0.5, 1, 2, 5, 10, and 15 mg/ml), blood with iodine (blood + 2 mg/ml iodine and blood + 4 mg/ml iodine), blood, and brain tissue-equivalent inserts. In this study, analysis of iodine measurements was limited to 0.2 to 15 mg/ml iodine inserts. Iodine inserts were grouped as all (0.2 to 15 mg/ml), low (0.2 to 2 mg/ml), and high (5 to 15 mg/ml) for selected analysis. Ground-truth iodine densities were based on information provided by the manufacturer. The arrangement of iodine inserts in the phantom is displayed in Fig. 1.

Fig. 1.

Experimental setup. (a) and (b) Photographs of a first-generation dual-source PCCT with multi-energy CT phantom in large and small sizes, respectively. (c) and (d) Reconstructed iodine map slice of large and small phantom sizes with numbered iodine inserts of varying concentrations: 1, iodine 2 mg/ml; 2, iodine 0.2 mg/ml; 3, iodine 0.5 mg/ml; 4, iodine 5 mg/ml; 5, iodine 1 mg/ml; 6, iodine 10 mg/ml; and 7, iodine 15 mg/ml. (WW,WL): (8, 3) mg/ml.

2.2. Image Acquisition and Reconstruction

All image acquisitions were performed on a first-generation clinical DS-PCCT in two acquisition modes: single-source (SS) mode and dual-source (DS) mode each at tube voltages of 120 and 140 kVp. For both phantom sizes, the phantom was placed in the iso-center of the scanner (Fig. 1) and a standard abdominal protocol was implemented for data acquisition and image reconstruction (Table 1). Without using any dose modulation, scans were performed from ultra-low to medium radiation dose levels: CT dose index (${\mathrm{CTDI}}_{\text{vol}}$) of 0.6, 0.8, 1.2, 1.6, 2, 4, 6, 10, and 15 mGy. The ${\mathrm{CTDI}}_{\text{vol}}$ range is equivalent to an effective dose range between 0.54 mSv and 13.50 mSv ($k=0.015\mathrm{mSv}$ ${\mathrm{mGy}}^{-1}$ ${\mathrm{cm}}^{-1}$) for an abdomen scan with a length of 60 cm. Based on the size of the patient, the selected dose range represents the clinical range from ultra-low doses to standard doses. Each experiment was repeated three times without moving the phantom between scans. Individual scans were reconstructed with three levels of quantum iterative reconstruction (QIR 0, QIR 2, and QIR 4) in spectral mode to generate iodine density maps. Additional virtual monoenergetic images (VMIs) at 70 keV were reconstructed for the 6.0 mGy scan for each phantom size to automate region of interest (ROI) placement.

Table 1.

Acquisition and reconstruction parameters.

Scanner model	NAEOTOM Alpha
Acquisition mode	SS, DS
Tube voltage	120, 140 kVp
Rotation time	0.25
Spiral pitch factor	1
Collimation	$144\times 0.4\mathrm{mm}$
Slice thickness	3 mm
Iterative reconstruction	QIR0, QIR2, QIR4
Reconstruction filter	Qr48
Reconstructed field of view	450 mm
Matrix size	$512\times 512$
Pixel spacing (in $x$ and $y$)	0.88 mm

2.3. Image Analysis

ROIs were prescribed on selected iodine-containing inserts on VMI 70 keV images at 6.0 mGy for each phantom size and combination of acquisition parameters. These ROIs were copied to the iodine density images at other radiation dose levels from the corresponding phantom size and combination of acquisition parameters. Mean and standard deviation were calculated across a total of 30 slices (3 repeated scans × 10 consecutive central slices) for each insert at each dose level. Measured iodine density corresponding to 0.2, 0.5, 1, 2, 5, 10, and 15 mg/ml iodine concentrations was plotted against expected values, and error bars depict the standard deviation ($\sigma$) of the mean. Within each iodine insert, measured iodine density from different phantom sizes (small/large) and tube combinations (SS120, SS140, DS120, DS140) were arranged by increasing radiation dose groups (ultra-low: $<1.5$, low: 1.5 to 4, and medium: 5 to 15 mGy). The corresponding root-mean-square error (RMSE-iodine) for each unique combination of iodine insert, tube voltage, source mode, and phantom size was calculated and plotted separately.

The mean and measured bias between measured iodine density concentrations and expected values for each phantom size and tube combinations over low radiation dose levels were illustrated using a Bland–Altman (BA) plot. The mean bias and measured bias were separately calculated and plotted for low iodine concentrations (0.2 to 2 mg/ml) and high iodine concentrations (5 to 15 mg/ml). Limits of agreement (LOA) were calculated as mean bias $\pm 1.96$ × ($\sigma$ of bias). Corresponding results for ultra-low and medium radiation dose levels were summarized in separate tables.

To demonstrate the effect of QIR on iodine quantification, RMSE values (RMSE-QIR) were calculated for each QIR level, iodine concentration group (all, low, and high), and radiation dose level across phantom sizes at SS120. RMSE-QIR was shown in three different plots labeled as all, low, and high. In each ${\mathrm{CTDI}}_{\mathrm{vol}}$ column, RMSE-QIR values were arranged by increasing QIR levels. RMSE was similarly calculated to demonstrate the effect of patient habitus on iodine quantification (RMSE-size). RMSE-size values similarly were calculated for each combination of phantom size, radiation dose, and iodine concentration group at SS120. They were visualized in scatter plots. Varying parameters for each evaluation metrics are listed in Table 2.

Table 2.

Varying parameters for each evaluation metric.

	Source and tube voltage	Phantom size	Iodine concentration	Radiation dose	QIR
BA	x	x	Low/high	x
RMSE-iodine	x	x	Each
RMSE-QIR	SS120 only		Low/high	x	x
RMSE-size	SS120 only	x	Low/high	x

3. Results

Measured iodine concentrations and corresponding RMSE-iodine values for each combination of iodine concentrations, phantom size, source mode, tube voltage, and dose range are summarized in Fig. 2. Overall, iodine quantification was accurate across all iodine inserts between measured and reference values, with a larger bias for higher-concentration iodine inserts. For iodine concentrations between 0.2 and 2 mg/ml, the RMSE-iodine values for each combination of acquisition parameters were below 0.3 mg/ml. RMSE-iodine values of less than 0.7 mg/ml were observed for 0.5 and 2 mg/ml iodine inserts for large phantom at DS120 and DS140 with ultra-low radiation dose levels. The majority of the RMSE-iodine values for 5 mg/ml were below 0.4 mg/ml. For 10 mg/ml iodine, the large phantom showed higher RMSE-iodine values ranging from 0.3 to 0.7 mg/ml compared to 0.1 to 0.3 mg/ml for the small phantom. The 15 mg/ml iodine insert demonstrated RMSE-iodine values less than 0.8 mg/ml except for an outlier for the large phantom at DS120 and ultra-low radiation dose levels.

Fig. 2.

Comparison of (a) measured iodine densities and (b) corresponding RMSE-iodine for individual inserts, ordered by increasing iodine concentration (0.2 to 15 mg/ml) for different phantom sizes and acquisition parameters. Within each column, measurements are ordered by increasing ${\mathrm{CTDI}}_{\text{vol}}$ groups (ultra-low, 0.6 to 1.2; low, 1.6 to 4; and medium, 5 to 15 mGy).

Figures 3 and 4 illustrate the BA plots for low and high iodine concentrations, respectively, separated for each phantom size and acquisition parameters over low radiation dose levels. For low iodine concentrations, the mean bias measured for large and small phantom sizes ranged from $-0.11$ to $-0.04$ and $-0.22$ to $-0.14\mathrm{mg}/\mathrm{ml}$, respectively. Corresponding mean bias ranges were $-0.11$ to 0.32 and $-0.44$ to $-0.22\mathrm{mg}/\mathrm{ml}$ for large and small phantoms, respectively, with high iodine concentrations. However, only a small difference between individual mean biases was observed across acquisition parameters for both phantom sizes. All bias measurements were within the calculated LOA. For low iodine concentrations (Fig. 3), the widths of the LOA from different acquisition modes ranged from 0.24 to 0.44 and 0.13 to 0.23 mg/ml for the large and small phantoms, respectively. For high iodine concentrations (Fig. 4), the corresponding LOA widths ranged from 0.94 to 1.34 and 0.8 to 0.95 mg/ml. BA results over ultra-low and medium radiation dose levels are summarized in Tables 3 and 4, respectively.

Fig. 3.

BA plots illustrating the measured bias and mean bias between measured and nominal iodine concentrations versus the average of measured and nominal iodine concentrations acquired for low iodine concentrations (0.2 to 2 mg/ml) across low radiation dose levels (${\mathrm{CTDI}}_{\mathrm{vol}}$ 1.6 to 4 mGy), phantom sizes, and QIR. The black dashed line represents the mean bias. The red and blue dashed lines represent the upper and lower limits of the mean bias (from $\pm 1.96\sigma$).

Fig. 4.

BA plots illustrating the measured bias and mean bias between measured and nominal iodine concentrations versus the average of measured and nominal iodine concentrations acquired for high iodine concentrations (5 to 15 mg/ml) across phantom sizes, low radiation dose levels (${\mathrm{CTDI}}_{\text{vol}}$ 1.6 to 4 mGy), and QIR. The black dashed line represents the mean bias while the red and blue dashed lines represent the upper and lower limits of the mean bias (from $\pm 1.96\sigma$).

Table 3.

Bias and limits of agreement for low iodine concentrations (0.2 to 2 mg/ml), phantom sizes, and acquisition mode combinations.

Radiation dose	Phantom size	Acquisition mode	Bias (mg/ml)				Mean bias (mg/ml)	Limits of agreement (mg/ml)
			Iodine 0.2 (mg/ml)	Iodine 0.5 (mg/ml)	Iodine 1 (mg/ml)	Iodine 2 (mg/ml)		Lower	Upper
Ultra-low	Large	SS120	0.11	−0.19	−0.10	−0.08	−0.06	−0.27	0.14
		SS140	0.06	−0.15	−0.11	−0.01	−0.05	−0.21	0.11
		DS120	−0.26	−0.37	−0.21	−0.53	−0.34	−0.58	−0.10
		DS140	−0.27	−0.43	−0.16	−0.49	−0.34	−0.59	−0.08
	Small	SS120	−0.19	−0.19	−0.18	−0.09	−0.16	−0.25	−0.08
		SS140	−0.18	−0.21	−0.13	−0.04	−0.14	−0.27	−0.01
		DS120	−0.14	−0.15	−0.10	−0.08	−0.12	−0.18	−0.06
		DS140	−0.12	−0.09	−0.05	0.02	−0.06	−0.16	0.04
Low	Large	SS120	0.02	−0.13	−0.18	0.02	−0.07	−0.24	0.10
		SS140	0.06	−0.11	−0.07	−0.04	−0.04	−0.16	0.08
		DS120	−0.02	−0.21	−0.23	0.00	−0.11	−0.32	0.09
		DS140	−0.04	−0.21	−0.20	0.05	−0.10	−0.32	0.12
	Small	SS120	−0.22	−0.26	−0.21	−0.13	−0.21	−0.29	−0.12
		SS140	−0.24	−0.28	−0.23	−0.12	−0.22	−0.33	−0.10
		DS120	−0.17	−0.21	−0.17	−0.12	−0.16	−0.23	−0.10
		DS140	−0.16	−0.18	−0.13	−0.07	−0.14	−0.22	−0.05
Medium	Large	SS120	−0.07	−0.20	−0.28	−0.29	−0.21	−0.38	−0.04
		SS140	0.01	−0.17	−0.24	−0.06	−0.12	−0.30	0.07
		DS120	−0.07	−0.24	−0.31	−0.26	−0.22	−0.40	−0.05
		DS140	−0.13	−0.21	−0.23	−0.12	−0.17	−0.27	−0.08
	Small	SS120	−0.12	−0.15	−0.16	−0.16	−0.15	−0.18	−0.12
		SS140	−0.13	−0.11	−0.11	−0.13	−0.12	−0.14	−0.10
		DS120	−0.17	−0.18	−0.19	−0.17	−0.18	−0.20	−0.16
		DS140	−0.17	−0.17	−0.15	−0.13	−0.16	−0.18	−0.13

Table 4.

Bias and limits of agreement for high iodine concentrations (5 to 15 mg/ml), phantom sizes, and acquisition mode combinations.

Radiation dose	Phantom size	Acquisition mode	Bias (mg/ml)			Mean bias (mg/ml)	Limits of agreement (mg/ml)
			Io 5 (mg/ml)	Io 10 (mg/ml)	Io 15 (mg/ml)		Lower	Upper
Ultra-low	Large	SS120	0.09	0.25	−0.54	−0.07	−0.74	0.60
		SS140	0.09	0.51	0.03	0.21	−0.21	0.62
		DS120	−0.47	0.11	−0.97	−0.44	−1.31	0.42
		DS140	−0.17	0.51	−0.12	0.07	−0.54	0.69
	Small	SS120	−0.27	0.00	−0.77	−0.35	−0.97	0.27
		SS140	−0.22	0.12	−0.53	−0.21	−0.73	0.31
		DS120	−0.20	0.06	−0.63	−0.26	−0.82	0.30
		DS140	−0.10	0.28	−0.30	−0.04	−0.51	0.43
Low	Large	SS120	0.10	0.41	−0.34	0.06	−0.55	0.66
		SS140	0.14	0.66	0.16	0.32	−0.15	0.79
		DS120	−0.08	0.30	−0.54	−0.11	−0.78	0.56
		DS140	0.04	0.68	0.25	0.32	−0.20	0.84
	Small	SS120	−0.34	−0.21	−0.78	−0.44	−0.92	0.03
		SS140	−0.34	−0.09	−0.65	−0.36	−0.81	0.09
		DS120	−0.28	−0.11	−0.67	−0.35	−0.81	0.11
		DS140	−0.23	0.04	−0.47	−0.22	−0.62	0.18
Medium	Large	SS120	−0.07	0.35	−0.34	−0.02	−0.58	0.54
		SS140	0.07	0.64	0.14	0.28	−0.22	0.78
		DS120	−0.23	0.16	−0.54	−0.20	−0.76	0.36
		DS140	0.04	0.66	0.22	0.31	−0.20	0.82
	Small	SS120	−0.24	−0.12	−0.53	−0.30	−0.64	0.05
		SS140	−0.21	−0.07	−0.50	−0.26	−0.61	0.09
		DS120	−0.30	−0.23	−0.65	−0.39	−0.75	−0.03
		DS140	−0.26	−0.14	−0.56	−0.32	−0.67	0.03

The highest mean bias ($-0.44\mathrm{mg}/\mathrm{ml}$) for ultra-low radiation dose levels occurred for the large phantom at DS120 with a high iodine concentration. In Table 3, the mean bias ranges for low iodine concentrations were $-0.34$ to $-0.05$ and $-0.16$ to $-0.06\mathrm{mg}/\mathrm{ml}$ for large and small phantoms, respectively. In Table 4, the mean bias ranges for large and small phantoms were $-0.44$ to 0.21 and $-0.35$ to $-0.04\mathrm{mg}/\mathrm{ml}$, respectively. The difference in the mean bias measurements between individual acquisition combinations was minimal and all bias measurements were within the LOA. In Table 3, the widths of the LOA for low iodine concentrations were 0.32 to 0.51 and 0.12 to 0.26 mg/ml for the large and small phantoms, respectively.

For medium radiation dose levels, iodine mean bias measurements ranged from $-0.22$ to $-0.12$ and $-0.18$ to $-0.12\mathrm{mg}/\mathrm{ml}$ for large and small phantoms, respectively, considering low iodine concentrations (Table 3). In Table 4, the mean bias measurements ranged from $-0.2$ to 0.31 and $-0.39$ to $-0.26\mathrm{mg}/\mathrm{ml}$ for the large and small phantoms, respectively. Only minor differences were observed between individual acquisition combinations, phantom sizes, and iodine concentration groups. Again, all bias measurements were within the designated LOA. The widths of the LOA for low iodine concentrations (Table 3) ranged from 0.19 to 0.37 and 0.04 to 0.06 mg/ml for the large and small phantom, respectively. The corresponding widths of LOA for high iodine concentrations (Table 4) were noted from 1.0 to 1.12 and 0.69 to 0.72 mg/ml.

Figure 5 presents the effect of QIR across all radiation dose levels acquired at SS120. RMSE-QIR values calculated overall, low, and high iodine concentrations are shown in three plots. Comparable RMSE-QIR values were calculated between individual iterative reconstruction levels across all radiation dose levels in all three cases. On average, overall RMSE-QIR values and those for low iodine concentrations were less than 0.3 and 0.2 mg/ml, respectively. Although higher RMSE-QIR values were calculated for high iodine concentrations, the RMSE-QIR values were less than 0.4 mg/ml.

Fig. 5.

RMSE-QIR for each combination of QIR, iodine concentration groups, and radiation doses at SS120. RMSE-QIR was calculated for all (0.2 to 15 mg/ml, left), low (0.2 to 2 mg/ml, middle), and high (5 to 15 mg/ml, right) iodine concentrations.

Figure 6 represents the influence of phantom size across all radiation dose acquired at SS120. Computed RMSE-size values using all and low iodine concentrations were below 0.4 mg/ml while RMSE-size for high iodine concentrations were less than 0.6 mg/ml. Minimal RMSE-size variations were evident from 1.2 to 6 and at 15 mGy over all iodine concentrations. RMSE-size values calculated over high concentrations showed slight variations across all radiation dose levels except at 0.8 mGy. Finally, RMSE-size values for the small phantom improved with dose level for dose levels higher than 4 mGy.

Fig. 6.

RMSE-size for different phantom sizes, radiation doses, and iodine concentration groups at SS120. RMSE-size was calculated for all (0.2 to 15 mg/ml, left), low (0.2 to 2 mg/ml, middle), and high (5 to 15 mg/ml, right) iodine concentrations.

4. Discussion

In this study, we present an initial systematic evaluation of a clinical DS-PCCT system for the measurement of iodine concentrations. We investigated the influence of patient habitus, acquisition parameters, and iterative reconstruction modes on the accuracy of DS-PCCT iodine maps. Overall, our phantom study demonstrated accurate iodine quantification with minimal effect of patient habitus, tube voltage combinations, or iterative reconstruction on the iodine measurement accuracy of the DS-PCCT system, including at doses considerably lower than doses used in current clinical day-to-day routine. It was noted that despite the minimal increase in iodine bias, PCCT technology, due to its excellent ultra-low dose performance, outperforms other spectral platforms that have demonstrated mean absolute percentage differences of 64% and 5% at 2.5 and 20 mGy, respectively.²²

DS-PCCT measurements of various iodine concentrations demonstrated comparable accuracy compared to ground truth over a wide range of values. This included high measurement accuracy at low iodine concentrations resulting in RMSE-QIR of less than 0.3 mg/ml across different phantom sizes, tube settings, and radiation dose groups. Based on data collected from an early prototype PCCT system, Riederer et al. reported similar iodine quantification performance with an RMSE of 0.19 mg/mL.²³ A more recent study with the same PCCT system exhibited an RMSE of 0.5 mg/mL.¹³ Reliable quantification of iodine uptake, particularly of low iodine concentrations, may enable improved diagnosis including early tumor detection, differentiation, and staging.^24,25 On the other hand, high iodine concentrations were associated with a slight increase in RMSE-QIR. Lang et al., on a research PCCT system, demonstrated similar observations with RMSE increasing from 0.4 to 0.9 mg/mL for iodine concentrations of 2 to 15 mg/mL, respectively.²⁶ DECT systems also exhibited this increase in error but more dramatically from $-0.5$ to $-3.2\mathrm{mg}/\mathrm{ml}$ for the same iodine concentrations.²⁷ All biases measured in this study were well within the LOA, though even lower measurement variability was seen with low iodine concentrations, the small phantom, and medium radiation dose levels.

Previous DECT studies reported a significant effect of patient habitus on the accuracy of iodine quantification at low radiation exposures.²² The advent of PCD technology has improved the quality of diagnostic imaging of large patients significantly compared to EIDs.^28–30 In our quantitative evaluation, the calculated RMSE-size for both phantom sizes across all iodine concentrations remained consistently below 0.4 mg/ml, even for very low dose levels. Furthermore, small RMSE-size discrepancies were seen between phantom sizes. These suggest accurate and consistent iodine concentration measurements regardless of the patient size. Sartoretti et al. similarly showed an increase in RMSE from 0.23 to 0.41 mg/mL between their small and large phantoms but noticeably utilized iodine maps from a three-material decomposition designed for liver fat quantification rather than the standard two-material decomposition used to generate standard iodine maps.³¹ Notably, differences in RMSE-size between small and large phantoms increased with higher iodine concentrations, which can be attributed to beam hardening and is consistent with previous evaluations of iodine quantification in DECT.²⁷ Similar iodine quantification performance was observed across the iterative reconstruction levels at all radiation dose levels. These results indicate that the noise reduction, which is associated with QIR strengths, does not significantly affect iodine measurements. Similar observations have been made in phantom studies on available DECT systems^22,32,33 and clinical studies with the DS-PCCT.^18,34

This study has a few limitations. First, a phantom was utilized instead of patient data for analysis, which removes the effect of respiratory motions and other factors that may affect iodine quantification. The phantom, however, allowed for the evaluation of iodine quantification accuracy with a large variation of acquisition and reconstruction parameters. This only included a single reconstruction filter but one specifically designed for quantification. Second, only one diameter of iodine insert was assessed in this study. Future studies will evaluate the effect of iodine insert size on iodine quantification. Third, only two phantom sizes were utilized, which does not sufficiently represent the distribution of patient sizes. Finally, although we did not perform a direct comparison with existing DECT systems, the extensive literature using the same phantom and similar acquisition parameters enables meaningful comparisons.^32,33,35

5. Conclusion

Quantification of iodine uptake can assist in clinical diagnosis, such as distinguishing between a hyperdense cyst and a solid tumor.¹ Accurate and reliable iodine density maps may eliminate the need for the unenhanced phase, which has served as a comparison for determining iodine uptake. It is anticipated that, in the future, accurate and precise iodine quantification will provide opportunities for the development of spectral-based biomarkers, such as those that may be employed to evaluate cancer treatments.

Biography

Biographies of the authors are not available.

Disclosures

The authors declare relationships with the following companies: Siemens Healthineers. Harold I. Litt and Peter B. Noël have a research agreement with Siemens Healthineers. Pooyan Sahbaee is an employee of Siemens Healthineers.

Code and Data Availability

Data presented in this study are available upon reasonable request from the authors. All phantom materials in this study are available from Sun Nuclear.