Low-Dose Chest CT Protocols for Imaging COVID-19 Pneumonia: Technique Parameters and Radiation Dose

Chest computed tomography (CT) plays a vital role in the early diagnosis, treatment, and follow-up of COVID-19 pneumonia during the pandemic. However, this raises concerns about excessive exposure to ionizing radiation. This study aimed to survey radiation doses in low-dose chest CT (LDCT) and ultra-low-dose chest CT (ULD) protocols used for imaging COVID-19 pneumonia relative to standard CT (STD) protocols so that the best possible practice and dose reduction techniques could be recommended. A total of 564 articles were identified by searching major scientific databases, including ISI Web of Science, Scopus, and PubMed. After evaluating the content and applying the inclusion criteria to technical factors and radiation dose metrics relevant to the LDCT protocols used for imaging COVID-19 patients, data from ten articles were extracted and analyzed. Technique factors that affect the application of LDCT and ULD are discussed, including tube current (mA), peak tube voltage (kVp), pitch factor, and iterative reconstruction (IR) algorithms. The CTDIvol values for the STD, LDCT, and ULD chest CT protocols ranged from 2.79–13.2 mGy, 0.90–4.40 mGy, and 0.20–0.28 mGy, respectively. The effective dose (ED) values for STD, LDCT, and ULD chest CT protocols ranged from 1.66–6.60 mSv, 0.50–0.80 mGy, and 0.39–0.64 mSv, respectively. Compared with the standard (STD), LDCT reduced the dose reduction by a factor of 2–4, whereas ULD reduced the dose reduction by a factor of 8–13. These dose reductions were achieved by applying scan parameters and techniques such as iterative reconstructions, ultra-long pitches, and fast spectral shaping with a tin filter. Using LDCT, the cumulative radiation dose of serial CT examinations during the acute period of COVID-19 may have been inferior or equivalent to that of conventional CT.


Introduction
Chest X-ray radiography (CXR) and computed tomography (CT) are essential tools for the diagnosis and follow-up of COVID-19 pneumonia [1]. COVID-19 is a type of pneumonia that was first identified in Wuhan, Hubei Province, China, in 2019 [2,3]. Affected patients present with fever and cough in addition to non-specific symptoms, including fatigue, dyspnea, muscle soreness, and headache [4]. Timely detection, quick intervention, and appropriate health interventions can prevent the rapid spread of COVID-19 and reduce morbidity and mortality [5].

Radiation Dose Quantities in CT
Three types of CT dose quantities were used to report chest CT COVID-19 radiation dose results: CTDI vol , DLP, and ED [34][35][36]. The CTDI vol conveys information regarding the average dose absorbed in the scanned region, whereas the DLP reflects the total integrated absorbed dose for a complete CT examination. Finally, ED is a quantity that provides the idea of weighted radiation risk.
Primarily, the CT dose is modeled in terms of the CT dose index (CTDI 100 ), which is measured at the center (CTDI 100,c ) and periphery (CTDI 100,p ) of a standard head or body CT dosimetry phantom using a pencil ionization chamber with an active length of 100 mm. The weighted CT dose index was computed as shown in Equation (1): CTDI vol considers the helical pitch or axial scan spacing and, hence, is related to CTDI w via: where p is the CT pitch factor, which was calculated using the formula: p = l/NT. where N is the number of simultaneously acquired tomographic slices, T is the slice thickness, and l is the couch movement per helical rotation. The overall energy delivered by a given scan protocol is better represented using the CT air kerma-length product, as shown in Equation (2) [34][35][36].
where L is the scan length. In the reviewed literature, the CTDI vol and DLP values in most centers were extracted from the DICOM header. The effective dose is used to express the radiation risk for partial-body irradiation. It is defined as the sum of the dose absorbed by each of the specified body organs and tissues multiplied by the tissue-weighting factor for the same organ or tissue [8]: ED refers to the average effective dose over age and sex, W T is the tissue weighting factor, and H T is the dose for organ T. In this literature survey, ED values were estimated by multiplying the DLP value by a conversion factor for chest CT equal to 100 kV: 0.014-mSv/mGy·cm [37]. CT dosimetry software such as CT Expo and automatic dose registry software have also been used for effective dose calculations [38,39]. For multiple CT examinations, the cumulative ED can be obtained by summing the observed EDs of radiation received during hospitalization per patient undergoing the CT protocols. Figure 1 is a PRISMA flowchart showing the article search, inclusion, and exclusion processes.
The effective dose is used to express the radiation risk for partial-body irradiation. It is defined as the sum of the dose absorbed by each of the specified body organs and tissues multiplied by the tissue-weighting factor for the same organ or tissue [8]: ED refers to the average effective dose over age and sex, WT is the tissue weighting factor, and HT is the dose for organ T. In this literature survey, ED values were estimated by multiplying the DLP value by a conversion factor for chest CT equal to 100 kV:0.014-mSv/mGy.cm [37]. CT dosimetry software such as CT Expo and automatic dose registry software have also been used for effective dose calculations [38,39]. For multiple CT examinations, the cumulative ED can be obtained by summing the observed EDs of radiation received during hospitalization per patient undergoing the CT protocols. Figure 1 is a PRISMA flowchart showing the article search, inclusion, and exclusion processes.   Table 1 summarizes the surveyed studies, detailing the objectives of each study, the subjects and equipment used, and the major findings. The survey results are analyzed in the following subsections: Table 1. Summary of the surveyed studies, detailing the objectives of each study, the subject and equipment used, and the major findings.

Study Objectives Study Population Major Findings
Italy: Agostini et al. [40] To study the feasibility of a ULD with fast, long-pitch, dual-source scanning 10 adult COVID-positive patients aged >18 years (average age: 53 years) Reduced radiation dose with acceptable image quality was obtained using ULD with 100 Sn kV spectral shaping, dual-source, and ultra-long pitch.
France: Greffier et al. [41] To compare the diagnostic performance of LDCT and STD using objective and subjective image quality measures 380 patients (M:F = 195:185) with a mean age of 66.3 ± 18.7 years; 97 CTs showed viral pneumonia ULD is less effective than STD CT for examining the interstitial space and is not recommended for follow-up of these diseases.
China: Kang and Zhou [42] To minimize radiation dose during chest CT imaging of COVID-19 using an LDCT protocol An LDCT protocol was implemented using energy spectrum tin-filtering technology (Sn100 low-dose protocol) The LDCT protocol resulted in a radiation dose reduction from 1/8 to 1/9 of the STD without significant loss of image quality. Hamper et al. [48] To determine LDCT parameters sufficient for acceptable image quality 36 patients positive for SARS-CoV-2 (M/F = 27/9) were retrospectively included for chest CT scans.
LDCT gives radiation doses in the submillisievert range without compromising image quality.
Steuwe et al. [49] To evaluate the image quality and radiation dose of an LDCT protocol and its diagnostic accuracy.
The authors developed an LDCT protocol with high sensitivity to diagnose COVID-19 with sufficient image quality.

Scan Parameters and Techniques
Different strategies have been implemented for dose reduction in chest CT, including the lowering of kVp and mAs as well as the use of modulation along with the use of an ultra-long pitch factor, without adversely affecting the quality of diagnostic information. Table 2 presents the features of the LDCT, ULD, and STD chest CT protocols for imaging COVID-19 pneumonia. Based on the literature surveyed, these factors are discussed below. Tube current (mA): Reducing the tube current is the first and most efficient way to reduce the radiation dose [40][41][42][43][44][45][46][47][48][49]. This is primarily achieved using tube current modulation (TCM), which adjusts the tube current according to a patient's body characteristics. Survey studies have used either TCM or vendor-specific mA modulation software to achieve a dose reduction in up to 50% in standard-sized patients [50], including CareDose 4D in the mA modulation software from Siemens. The use of special mA modulation programs has helped professionals achieve true LDCT for imaging COVID-19.
Peak tube voltage (kVp): LDCT has been performed using tube voltages of less than or equal to 100 kVp [51]. Therefore, 80-100 kVp is commonly used in LDCT. Reducing the voltage from 120 kVp to 80 kVp increases the image noise and would require a fourfold increase in the tube current to maintain the image quality [52][53][54]. For better CT Life 2023, 13, 992 7 of 14 performance, spectral shaping with a tin filter at 100 kVp (100 Sn kVp) was used, which reduced low-energy X-rays, resulting in a significant dose reduction [55,56]. Another technique that can be used is dual-energy CT (DECT) for the LDCT protocol, which is centered on the concurrent acquisition of low (80 kVp) and high (140 kVp) energy X-rays [57]. Similarly, Agostini et al. [40] used high-definition DECT at 90 and 150 Sn kVp and fast, low-dose, long-pitch CT for imaging COVID-19.
Pitch factor: Radiation dose reduction in CT can be achieved by altering scanning parameters, such as mAs, kVp, collimation, and pitch value [35,36]. The use of an ultralong pitch is the main feature of the LDCT protocol, and using a long pitch value prevents motion artifacts, thus improving image quality and reducing patient dose [40][41][42][43][44][45][46][47][48][49]. As demonstrated in this study, ULD, in particular, has been accompanied by a substantial increase in the pitch factor compared to the STD chest protocol [41,42].
Iterative reconstruction (IR) algorithms: Lowering kVp and mAs to lower the dose, as is needed in LDCT for imaging COVID-19 patients, causes more noise, which lowers the quality of the image and makes it necessary to use iterative reconstruction (IR) algorithms. IR algorithms have the potential for radiation dose optimization by lowering image noise [45]. IR algorithms have been used instead of filtered backprojection (FBP) reconstruction algorithms. Equipment vendors have used different models of IR algorithms with particular characteristics: CT scanners built by Siemens use ADMIRE IR algorithms, whereas GE Healthcare uses ASIR [40,41,43]. The use of IR is vital for eliminating increased image noise and artefacts because of the lowered tube current in low-dose CT. According to Gordon et al. [58], IR algorithm use resulted in a 52% reduction in noise compared to FBP. This is obvious in the current survey: relatively high effective doses were required when filtered FBP was used in the study by Tabatabaei et al. [44] and Li et al. [45] compared with other studies in which IR algorithms were used. A significant dose reduction was achieved when TCM was combined with IR algorithms.

Radiation Dose Considerations
This review addresses the differences in cumulative radiation dose exposure due to the successive use of ionizing radiation in different chest CT protocols for imaging COVID-19 pneumonia. Radiation doses are summarized in terms of CTDI vol and ED. CTDI vol quantifies the scanner output and improves equipment performance following the use of vendorrelated optimization measures [59,60]. ED refers to the total radiation risk to the patient and is suitable for comparing different imaging modalities and techniques concerning the radiation burden [61]. Based on our previous dose surveys in Sudan, patients received effective dose values in the range of 4.6-5.2 mSv during routine CT examinations [62,63].
During COVID-19 pneumonia follow-up, a patient may undergo 6-8 chest CT scans, resulting in a significant cumulative dose. Zhou et al. [46] surveyed patient doses in 550 COVID-19 patients who underwent chest CT, noting a cumulative ED of 19.07 mSv for patients undergoing multiple CT examinations during the acute period of the COVID-19 pandemic.
Several studies comparing the STD chest CT protocol with the LDCT or ULD CT protocol used to image COVID-19 patients have been conducted. Table 3 summarizes the patient dose metric values in the STD, LDCT, and ULD chest CT protocols used for imaging COVID-19 patients. Table 4 provides information about the effective dose calculation. Figure 2 presents the aggregate results for both groups' CTDI vol and ED using boxplots. Figure 3 presents a bar chart comparing the LDCT and ULD with STD protocols.   The CTDIvol values ranged from 2.79-13.2 mGy in STD, 0.9-4.4 mGy in LDCT, and 0.20-0.28 mGy in ULD chest CT protocols. ED values ranged from 1.7-6.6 mSv in STD, 0.5-0.8 mSv in LDCT, and 0.39-0.64 mSv in ULD chest CT protocols. In this study, using the ULD chest CT protocol led to a dose reduction factor of 8-13, while using the LDCT chest protocol led to a dose reduction factor of 2-4 [40][41][42][43][44][45][46][47][48]. This survey revealed two ways to achieve dose reduction in CT scans [50]. This includes defining the target image quality that is adequate for a specific diagnostic task and achieving dose reduction by reducing the noise level using IR to improve the image quality. These results were obtained using various dose-reduction techniques, as discussed in the following sections.  The CTDIvol values ranged from 2.79-13.2 mGy in STD, 0.9-4.4 mGy in LDCT, and 0.20-0.28 mGy in ULD chest CT protocols. ED values ranged from 1.7-6.6 mSv in STD, 0.5-0.8 mSv in LDCT, and 0.39-0.64 mSv in ULD chest CT protocols. In this study, using the ULD chest CT protocol led to a dose reduction factor of 8-13, while using the LDCT chest protocol led to a dose reduction factor of 2-4 [40][41][42][43][44][45][46][47][48]. This survey revealed two ways to achieve dose reduction in CT scans [50]. This includes defining the target image quality that is adequate for a specific diagnostic task and achieving dose reduction by reducing the noise level using IR to improve the image quality. These results were obtained using various dose-reduction techniques, as discussed in the following sections.
In summary, Agostini et al. [40] reported an ED of 3.28 mSv in STD compared to 0.28 mSv in ULD protocols, resulting in a dose reduction of 11 times. The IR algorithm ADMIRE was used in conjunction with a 90/150Sn DSCT scanner. In another study, Greffier et al. [41] performed LDCT using a Siemens EDGE/Somatom CT scanner, 100 kVp, 10 mA, 1.7 pitch, and the ADMIRE 4 IR algorithm. Their results revealed a dose reduction by a factor of 8. Karakaş et al. reported ED values of 0.22 mSv in ULD as opposed to ED values of 0.28 mSv, indicating a 13-fold dosage decrease [43]. They used the GE Healthcare Optima 660 SE with 80 kVp and 10 mA ref, as well as the ASIR algorithm. In the ULD chest CT protocol, a significantly low tube current was used in conjunction with IR to preserve image quality. The dose was reduced by approximately 10 times that of the standard CT, comparable to the corresponding radiographic procedures.
Tabatabaei et al. [44] reported an ED of 6 mSv in STD protocols compared with 1.80 mSv in LDCT protocols. This means that the dose was reduced by a factor of 4. Similarly, Li et al. [45] reported ED values of 5.05 mSv in STD compared with 1.22 mSv in LDCT, achieving a dose reduction by a factor of 4. The dose reduction in ULD was almost double that achieved using LDCT. Owing to the degradation of image qualities causing difficult changes in kV and mAs, the ULD includes more technical features to comprehend the expected degradation in image quality.
Owing to the high dose reduction, ULD chest CT is inevitably associated with increased noise and artifacts, which necessitates the use of IR algorithms to compensate for the loss of image quality.

Image Quality Issues
Based on the results of the literature survey, the application of LDCT and ULD chest CT protocols for imaging COVID-19 patients is inevitably accompanied by concerns about image quality. The authors identified two sources that lead to lower image quality: COVID-19 patients are coughing, causing motion artefacts, while a lower radiation dose causes an increase in noise, both of which cause degradation in image quality, which is addressed by using LDCT with a high speed and ultra-long pitch.   In the surveyed literature, the authors evaluated image quality using objective and subjective methods [40,41,48,49], only objective methods [42], only subjective methods [43,45,47], and articles that did not include image quality evaluations [44,46]. Objective image quality evaluations were mainly based on contrast-to-noise (CNR) and signal-to-noise ratio (SNR) measurements [49], whereas subjective image quality evaluations were based on visual image quality ratings based on the radiologist's perceptions.
Most authors have reported that ULD and ULD protocols are achievable with acceptable image quality. Agostini et al. [40] performed image quality evaluations and reported significant differences in SNR and CNR for several anatomical structures using the STD, LDCT, and ULD protocols. LDCT and ULD based on HD-DECT showed comparable diagnostic performance as well as a substantial reduction in motion artefacts. This is attributed to the use of an energy spectrum tin filter (100Sn kVp), which results in reduced image noise and radiation dose. Similarly, Kang et al. [42] used a tin filter (Sn100 kVp) and reported a significant reduction in the effective dose without a significant reduction in the image quality. Li et al. [45] performed subjective evaluations of image quality. The radiation dose was significantly reduced at acceptable image quality when using CareDose 4D along with Karl 3D IR in LDCT. Similar results were reported by Dangis et al. [47], who conducted subjective image evaluations demonstrating that LDCT at the sub-millisievert dose level permits imaging of suspected COVID-19. Hamper et al. [48] determined the CNR and SNR image quality parameters, demonstrating LDCT imaging with a radiation dose in the submillisievert range without sacrificing the quality of the diagnostic information.
However, few authors have expressed concerns when applying the LDCT and ULD chest CT protocols for imaging COVID-19 patients. According to Agostini et al. [40], the ULD chest protocol cannot be used to image overweight patients because the amount of radiation is insufficient to penetrate the body, resulting in poor image quality. This can be remedied by boosting the tube's voltage and current using a CT scanner with tin filters that alter the spectrum's shape or DECT.
Additionally, Karakaş et al. [43] carried out an objective image quality assessment, which shows a significant difference in the performance of SDT and LDCT. According to this study, LDCT may not be suitable for the initial imaging of suspected COVID-19 patients due to the requirement for higher sensitivity, but it is recommended for the followup of COVID-19 patients. In contrast, Steuwe et al. [49] reported that noise and CNR were significantly superior to ULD than STD and concluded that ULD might not be suitable for the long-term follow-up of viral pneumonia. The results demonstrated that, for lung imaging, there was a decrease in noise and an increased CNR with the ULD compared to the STD. These changes were related to an increase in the iterative level and the use of a softer reconstruction kernel.
The main limitation of this study is that we did not present quantitative image quality parameters. This is because there was no unified set of image quality parameters measured across all centers. However, we have summarized the major findings and recommendations of the reviewed literature, which emphasize the importance of image quality evaluation in conjunction with radiation dose measurement in the application of LDCT and ULD protocols.

Summary
This review summarizes the technical features of current low-dose CT (LDCT) protocols compared with standard chest CT protocols, focusing on radiation dose optimization challenges. During the acute phase of COVID-19 pneumonia, chest CT is used repeatedly to monitor the patients. During this time, a patient can undergo 6-8 CT scans, resulting in a significant cumulative radiation dose. In these cases, the use of standard CT for imaging COVID-19 results in a significantly effective dose, which is known to increase the probability of carcinogenesis. Overall, this has led to concerns regarding the use of ionizing radiation for COVID-19 pneumonia imaging. To address these concerns over radiation risk, CT equipment vendors and professionals have adopted several dose reduction techniques and optimization tools. Most patients with COVID-19 require fast scanning and higher pitch values owing to the shortness of breath or coughing. The major component of the LDCT protocol is the incorporation of dose reduction through both mA modulation and automatic tube voltage selection features, which are available in certain CT scanners. Dose reduction is often achieved by shaping using a tin filter at kVp and dual-energy CT (DECT). However, radiation dose reduction results in image artifacts that degrade image quality and necessitate the use of IR to preserve image quality.