Continuous anatomical radiation modulation for the acquisition of coronary computed tomography angiography in difficult to scan patients: a prospective cohort study Authors: Yosef A. Cohen, M. Robert Peters, Mrinali Shetty, Luca Bremner, Michelle Castillo, David Bae, Cyrus Attia, Seth Uretsky, Andrew J. Einstein Published: 2026 ABSTRACT BackgroundA novel continuous anatomically modulated (CAM) retrospectively-gated scan mode for coronary computed tomography angiography (CCTA) modulates tube current as a function of patient anatomy, offering potential for good image quality (IQ) and radiation dose reduction in patients undergoing CCTA on a traditional scanner with significant ectopy (SE) or high heart rate (HHR) who are not candidates for prospectively-triggered scanning. We evaluated the effect of CAM on IQ and radiation dose in a large group of patients.MethodsThree hundred and twenty consecutive patients at a single center (Advanced Cardiovascular Imaging, New York, NY) undergoing CCTA were prospectively protocoled by clinicians to CAM retrospectively-gated scanning if they had SE/HHR or prospectively-triggered scanning if they had regular heart rates. IQ was graded on a 5-point scale by two readers and radiation dose-length product (DLP) evaluated. Radiation exposure was assessed using DLP, with effective dose estimated using a cardiac-specific conversion factor. Associations between patient weight and radiation dose were evaluated using log-linear regression models.ResultsOf 320 patients, 166 (52%) were assigned to prospective triggering (PT) and 154 (48%) to continuous anatomically modulated retrospectively gated (CAM RG). Patients in the CAM RG group had higher heart rates throughout the scanning process. Mean DLP was higher in the CAM RG group compared with the PT group (411.6±308.9 vs. 324.0±208.4 mGy·cm; P=0.03), corresponding to effective doses of 10.7±8.0 vs. 8.4±5.4 mSv, respectively. Despite these differences, IQ remained high across all groups, with mean composite scores exceeding 4.5. When diagnostic quality was defined as an overall IQ score ≥4, the proportion of diagnostic-quality studies did not differ significantly between groups (92.8% vs. 89.6%, P=0.42). In adjusted logistic regression models, scan protocol was not independently associated with diagnostic-quality imaging. Patient weight demonstrated a strong association with radiation dose, with DLP increasing by approximately 13–16% per 10-lb increase in weight, independent of scan protocol.ConclusionsOn a standard single-source CT scanner, CAM-based retrospectively gated CCTA provides consistently diagnostic IQ in patients with more challenging heart rate profiles, at the cost of modestly higher radiation exposure that is largely driven by patient size. CAM RG represents a practical, software-based acquisition strategy for centers without advanced CT hardware, enabling effective imaging of complex patients while maintaining acceptable radiation dose. Highlight box Key findings • Continuous anatomically modulated (CAM) retrospectively gated coronary computed tomography angiography (CCTA) provides consistently diagnostic image quality (IQ) in patients with significant ectopy or high heart rates on standard single-source scanners. While radiation exposure was modestly higher compared with prospectively triggered scanning, no differences were observed in rates of diagnostic-quality imaging. Patient weight was the strongest independent determinant of radiation dose, with dose increasing by approximately 13–16% per 10-lb increase. What is known and what is new? • Prospectively triggered CCTA is the preferred low-radiation approach but is often not feasible in patients with irregular rhythms or elevated heart rates. Retrospective gating enables imaging in these patients but is associated with higher radiation exposure, and prior solutions have largely relied on advanced or costly computed tomography (CT) hardware. • This study demonstrates that CAM scanning can achieve diagnostic IQ in difficult-to-scan patients using standard CT hardware. Importantly, CAM preserves diagnostic performance despite more challenging heart rate profiles, though it does not fully eliminate the inherent radiation penalty of retrospective gating. What are the implications, and what should change now? • CAM retrospectively gated scanning represents a practical and scalable solution for imaging patients with high heart rates or ectopy in centers without access to advanced CT technology. • These findings support broader adoption of CAM as a protocol-level strategy to maintain diagnostic IQ in complex patients, with an emphasis on patient size-based dose optimization. Introduction Proper control of heart rate (HR) in patients undergoing coronary computed tomography angiography (CCTA), with target HR typically <65 beats per minute (bpm), is important in obtaining images of diagnostic quality in CCTA using most traditional scanners. This also allows for scanning by prospective triggering (PT) thereby reducing radiation (1). Patients with significant ectopy (SE) or high heart rates (HHRs) often demonstrate significant coronary motion on CCTA that poses challenges for accurate evaluation. A number of studies have analyzed techniques to scan patients with difficult to scan heart rates but they typically require advanced, costly computed tomography (CT) hardware that is not widely available. In this study, we utilize a novel continuous anatomically modulated retrospectively gated (CAM RG) scan acquisition mode to evaluate its capabilities of obtaining diagnostic quality images in difficult to scan patients on a scanner with 4 cm z-axis coverage and modest gantry rotation time. Continuous anatomically modulated (CAM) scanning is a form of retrospectively gated acquisition in which tube current is continuously adjusted throughout the cardiac cycle based on both temporal and spatial considerations. In contrast to conventional retrospective gating, which applies relatively uniform radiation exposure across all cardiac phases, CAM dynamically modulates tube current according to the expected diagnostic contribution of each phase and the underlying attenuation characteristics of the scanned anatomy. Specifically, higher tube current is delivered during phases of the cardiac cycle most likely to yield diagnostic images, while reduced current is applied during less critical phases, thereby avoiding unnecessary radiation exposure. In addition, anatomical modulation allows adjustment of tube current in response to patient-specific attenuation, further optimizing dose distribution. As a result, CAM functions as a dose-optimization strategy within a retrospectively gated framework, aiming to preserve image quality (IQ) in challenging patients while mitigating, but not eliminating, the inherently higher radiation exposure associated with retrospective gating. We propose that CAM mode would maintain IQ in patients with SE or HHRs while not increasing radiation dose as compared to PT scanning. We present this retrospective analysis in accordance with the STROBE reporting checklist (available at https://cdt.amegroups.com/article/view/10.21037/cdt-2026-1-0095/rc). Methods Scanning protocol Consecutive patients at an outpatient cardiac radiology practice (Advanced Cardiovascular Imaging, New York, NY) who underwent CCTA on an Aquilion Prime 160 (Canon, Tustin, CA) were enrolled in the study. This is a single-source CT scanner with 80 detector rows, 4 cm z-axis coverage and 0.35 s gantry-rotation time, equipped with SureExposure CAM tube current modulation. Patients were prescribed by the supervising imaging clinician to either PT scanning (Group 1), or retrospectively gated scanning with CAM (Group 2) at the time of protocol selection. Assignments were based on anticipated heart rate and rhythm stability, with the goal of maintaining diagnostic IQ while minimizing radiation exposure. Patients without contraindications were premedicated with 50 mg of metoprolol tartrate approximately 12 hours before scanning and with an additional 50–100 mg one hour prior to scanning. Supplemental oral or intravenous (IV) metoprolol was given for patients with persistently elevated heart rate. All patients were administered 0.8 mg of nitroglycerin five minutes prior to image acquisition. Patient categorization Patients assigned to PT scanning (Group 1) were subclassified into: ❖ Group 1a: patients who successfully underwent PT scanning as planned; and ❖ Group 1a: patients who successfully underwent PT scanning as planned; and ❖ Group 1b: patients initially protocolled to PT scanning who were automatically mode-switched by the scanner to CAM RG due to SE or heart rate variability at the time of acquisition. ❖ Group 1b: patients initially protocolled to PT scanning who were automatically mode-switched by the scanner to CAM RG due to SE or heart rate variability at the time of acquisition. Patients assigned to CAM RG scanning (Group 2) were subclassified into: ❖ Group 2a: patients with possible concern for ectopy or elevated heart rate, for whom CAM RG scanning was selected based on clinician judgment; and ❖ Group 2a: patients with possible concern for ectopy or elevated heart rate, for whom CAM RG scanning was selected based on clinician judgment; and ❖ Group 2b: patients with unequivocal SE or HHR, for whom CAM RG scanning was deemed mandatory. ❖ Group 2b: patients with unequivocal SE or HHR, for whom CAM RG scanning was deemed mandatory. These groupings were defined prospectively at the time of protocol selection and are illustrated in Figure 1. Patient scan protocol classification. Patients undergoing CCTA were evaluated for the presence of ectopy or elevated heart rates and subsequently protocolled to either a prospectively triggered acquisition or a retrospective acquisition using continuous anatomically modulated software. Patients protocolled to prospective triggering were either scanned as planned or automatically mode-switched to retrospective acquisition at the time of scanning if significant ectopy or high heart rate occurred. Those protocolled to retrospective scanning were classified based on the degree of clinical concern for ectopy or high heart rate, ranging from possible to significant concern. CAM, continuous anatomically modulated; CCTA, coronary computed tomography angiography; HHR, high heart rate. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by Columbia University’s institutional review board (No. AAAQ2654) and individual consent for this retrospective analysis was waived. Statistical analysis Patient demographics, clinical characteristics, HR parameters and scan acquisition data were collected. IQ parameters—including overall IQ, motion artifact, and image noise—were determined using an established (2) 5-point scales (5 best, 1 worst) on the basis of the consensus of a cardiac radiologist and an advanced-imaging cardiologist with over 10 years of experience. A composite IQ score was calculated for each patient as the mean of the four component scores. Dose-length product (DLP) was compared as a measure of radiation dose. Effective dose (mSv) was estimated using a cardiac specific conversion factor (k=0.026 mSv·mGy–1·cm–1) (3). Continuous variables are reported as mean ± standard deviation, and categorical variables as counts with percentages. Group comparisons were performed using Wilcoxon rank-sum tests for continuous variables and χ2 or Fisher’s exact tests for categorical variables, as appropriate. To evaluate the association between patient size and radiation dose, log-linear regression models were constructed with log(DLP) as the dependent variable. Weight was modeled as a continuous predictor and effects were reported as percent change in DLP per 10-lb increase in weight with 95% confidence intervals. Models were fit overall and separately within each scan group. Diagnostic-quality imaging was defined as an overall IQ score ≥4. Multivariable logistic regression models were constructed to evaluate the independent association between scan protocol and diagnostic-quality imaging after adjustment for patient weight, heart rate at scan, and age. All analyses were performed using R (version 4.4.2). A two-sided P value <0.05 was considered statistically significant. Results Three hundred and twenty consecutive patients were enrolled in the study. Of 320 patients, 166 (52%) were protocoled to Group 1 and 154 (48%) patients to Group 2 (Table 1). Within group 1, 151 patients completed PT scanning as planned (Group 1a), while 15 patients were scanner switched to CAM RG due to HR irregularity (Group 1b). Of those protocoled to Group 2, 89 patients were classified as having SE or HHRs requiring CAM RG scanning (Group 2b). Therefore, in Group 2, 65 patients ultimately underwent PT scanning. Table 1Baseline characteristics, heart rate, radiation dose, and image qualityVariableGroup 1 (PT) (n=166)Group 2 (CAM RG) (n=154)P (1 vs. 2)Group 1a (n=151)Group 2b (n=89)P (1a vs. 2b)Age, years60.3 (12.2)62.7 (12.1)0.04660.0 (12.4)62.0 (12.4)0.18Weight, lb182.3 (42.3)177.5 (37.4)0.22183.1 (42.9)180.1 (39.1)0.40Diabetes16 (9.6%)18 (11.7%)0.6816 (10.6%)10 (11.2%)>0.99Asthma8 (4.8%)10 (6.5%)0.677 (4.6%)7 (8.0%)0.44HR at admission, bpm65.2 (9.2)72.7 (12.6)<0.00165.0 (9.4)75.6 (13.3)<0.001HR after beta-blocker, bpm59.1 (4.1)63.3 (6.7)<0.00158.9 (4.1)64.1 (7.5)<0.001Max HR at scan, bpm59.0 (7.9)69.6 (13.9)<0.00158.5 (7.6)72.4 (14.9)<0.001DLP, mGy·cm324.0 (208.4)411.6 (308.9)0.03316.0 (198.1)417.8 (308.9)0.02Effective dose, mSv 8.42 (5.42) 10.70 (8.03)<0.0018.22 (5.15)10.86 (8.03)<0.001Image quality composite4.75 (0.40)4.66 (0.39)0.0084.77 (0.40)4.65 (0.38)0.005 Diagnostic quality (overall ≥4) 154 (92.8%)138 (89.6%)0.42141 (93.4%)78 (87.6%)0.20 Overall4.52 (0.74)4.34 (0.76)0.0084.55 (0.75)4.31 (0.79)0.005 Motion artifact4.57 (0.65)4.40 (0.67)0.014.60 (0.63)4.39 (0.67)0.009 Step artifact4.92 (0.40)4.93 (0.35)0.894.93 (0.40)4.93 (0.25)0.37 Noise4.99 (0.08)4.98 (0.14)0.284.99 (0.08)4.98 (0.15)0.29Values are shown as mean (standard deviation) unless otherwise indicated. Group 1 includes patients prospectively assigned to PT. Group 2 includes patients prospectively assigned to retrospectively gated scanning with tube current modulation (CAM RG). Group 1a includes patients assigned to PT who successfully underwent PT scanning as planned. Group 2b includes patients with unequivocal significant ectopy or elevated heart rate for whom CAM RG scanning was deemed mandatory. P values compare Group 1 versus Group 2 and Group 1a versus Group 2b, respectively. Image quality is shown both as a composite score and as its four component scores. Effective dose (mSv) estimated from DLP using k=0.026 mSv·mGy−1·cm−1 (mSv = k × DLP). Diagnostic quality was defined as overall image quality score ≥4. CAM RG, continuous anatomically modulated retrospectively gated; DLP, dose-length product; HR, heart rate; PT, prospective triggering. Baseline characteristics, heart rate, radiation dose, and image quality Values are shown as mean (standard deviation) unless otherwise indicated. Group 1 includes patients prospectively assigned to PT. Group 2 includes patients prospectively assigned to retrospectively gated scanning with tube current modulation (CAM RG). Group 1a includes patients assigned to PT who successfully underwent PT scanning as planned. Group 2b includes patients with unequivocal significant ectopy or elevated heart rate for whom CAM RG scanning was deemed mandatory. P values compare Group 1 versus Group 2 and Group 1a versus Group 2b, respectively. Image quality is shown both as a composite score and as its four component scores. Effective dose (mSv) estimated from DLP using k=0.026 mSv·mGy−1·cm−1 (mSv = k × DLP). Diagnostic quality was defined as overall image quality score ≥4. CAM RG, continuous anatomically modulated retrospectively gated; DLP, dose-length product; HR, heart rate; PT, prospective triggering. Baseline demographic characteristics were generally similar across groups (Table 1). Mean age was slightly higher in the CAM RG group compared with the PT group. Body weight, prevalence of diabetes mellitus and asthma did not differ significantly. Patients undergoing CAM RG scanning had higher heart rates throughout the scanning process. Compared with the PT group, the CAM RG group demonstrated higher heart rate at admission, after beta-blocker administration, and at the time of scan acquisition (all P<0.001). These differences were preserved in the pre-specified comparison between Group 1a and Group 2b, confirming that CAM RG was preferentially used in patients with more challenging heart rate profiles. Mean radiation dose was higher in the CAM RG group compared with the PT group (411.6±308.9 vs. 324.0±208.4 mGy·cm; P=0.03) (Table 1). A similar difference was observed in the comparison between Group 1a and Group 2b. When converted to effective dose, mean exposure was 8.4±5.4 mSv in the PT group and 10.7±8.0 mSv in the CAM RG group. Patient weight demonstrated a strong and consistent association with radiation dose. In adjusted log-linear models, DLP increased by approximately 13–16% per 10-lb increase in body weight, depending on scan group (all P<0.001). This relationship was observed both overall and within each protocol subgroup (Table 2, Figures 2,3). Table 2Adjusted predictors of radiation doseVariablePercent change in DLP (95% CI)P valueCAM RG vs. PT25.7% (13.3% to 39.4%)<0.001Weight (per 10 lb)13.6% (12.2% to 14.9%)<0.001Max HR at scan (per 10 bpm)−2.4% (−6.4% to 1.7%)0.25Age (per 10 years)7.8% (3.4% to 12.4%)<0.001Percent changes are derived from a multivariable log-linear regression model with log-transformed DLP as the dependent variable. Estimates represent the adjusted percent change in DLP associated with each predictor, with 95% CIs. Scan protocol was modeled as CAM RG scanning versus PT. Weight, maximum heart rate at scan, and age were modeled continuously and scaled as shown in the table. Multivariable log-linear regression with DLP as the dependent variable and protocol type, weight (per 10 lb), maximum heart rate at scan (per 10 bpm), and age (per 10 years) as independent variables. Sex was not included because it was not available in the dataset. CAM RG, continuous anatomically modulated retrospectively gated; CI, confidence interval; DLP, dose-length product; HR, heart rate; PT, prospective triggering. Adjusted predictors of radiation dose Percent changes are derived from a multivariable log-linear regression model with log-transformed DLP as the dependent variable. Estimates represent the adjusted percent change in DLP associated with each predictor, with 95% CIs. Scan protocol was modeled as CAM RG scanning versus PT. Weight, maximum heart rate at scan, and age were modeled continuously and scaled as shown in the table. Multivariable log-linear regression with DLP as the dependent variable and protocol type, weight (per 10 lb), maximum heart rate at scan (per 10 bpm), and age (per 10 years) as independent variables. Sex was not included because it was not available in the dataset. CAM RG, continuous anatomically modulated retrospectively gated; CI, confidence interval; DLP, dose-length product; HR, heart rate; PT, prospective triggering. Adjusted predictors of radiation dose. Forest plot showing adjusted percent change in DLP associated with scan protocol, weight, maximum heart rate at scan, and age. Estimates were derived from a multivariable log-linear regression model with log-transformed DLP as the dependent variable. Points represent adjusted percent change in DLP, and horizontal bars represent 95% CIs. The vertical reference line indicates no percent change. CAM RG scanning was associated with higher DLP compared with PT, while greater weight and older age were also independently associated with higher DLP. Maximum heart rate at scan was not significantly associated with DLP after adjustment. CAM RG, continuous anatomically modulated retrospectively gated; CI, confidence interval; DLP, dose-length product; HR, heart rate; PT, prospective triggering. Radiation dose versus weight by protocol. Scatterplot showing the association between patient weight and DLP according to scan protocol. The y-axis is shown on a logarithmic scale. Points represent individual patients, and fitted lines represent protocol-specific log-linear associations between weight and DLP. PT is shown with the solid line, and CAM RG scanning is shown with the dashed line. The interaction P value represents whether the association between weight and DLP differed by scan protocol. The adjusted CAM RG versus PT estimate reflects the percent difference in DLP associated with CAM RG after adjustment for weight, maximum heart rate at scan, and age. CAM RG, continuous anatomically modulated retrospectively gated; DLP, dose-length product; PT, prospective triggering. Composite IQ scores were slightly higher in the PT group compared with the CAM RG group, but mean scores in all groups exceeded 4.5 (Table 1). In adjusted logistic regression models, scan protocol was not independently associated with diagnostic-quality imaging (Table 3, Figure 2). When diagnostic quality was defined as an overall IQ score ≥4, the proportion of diagnostic-quality studies remained high and did not differ significantly between groups (92.8% vs. 89.6%, P=0.42), including in the comparison between Group 1a and Group 2b (Figure 4). Table 3Adjusted predictors of diagnostic-quality imagingVariableAdjusted OR for diagnostic quality (95% CI)P valueCAM RG vs. PT0.80 (0.32 to 2.03)0.64Weight (per 10 lb)0.99 (0.89 to 1.09)0.82Max HR at scan (per 10 bpm)0.85 (0.61 to 1.18)0.33Age (per 10 years)1.01 (0.70 to 1.47)0.96Adjusted odds ratios are derived from a multivariable logistic regression model with diagnostic-quality imaging as the dependent variable. Diagnostic-quality imaging was defined as an overall image quality score ≥4. Estimates represent the adjusted OR for diagnostic-quality imaging associated with each predictor, with 95% CIs. Scan protocol was modeled as CAM RG scanning versus PT. Weight, maximum heart rate at scan, and age were modeled continuously and scaled as shown in the table. Multivariable logistic regression with diagnostic quality (overall image quality score ≥4) as the dependent variable and protocol type, weight (per 10 lb), maximum heart rate at scan (per 10 bpm), and age (per 10 years) as independent variables. CAM RG, continuous anatomically modulated retrospectively gated; CI, confidence interval; HR, heart rate; OR, odds ratio; PT, prospective triggering. Adjusted predictors of diagnostic-quality imaging Adjusted odds ratios are derived from a multivariable logistic regression model with diagnostic-quality imaging as the dependent variable. Diagnostic-quality imaging was defined as an overall image quality score ≥4. Estimates represent the adjusted OR for diagnostic-quality imaging associated with each predictor, with 95% CIs. Scan protocol was modeled as CAM RG scanning versus PT. Weight, maximum heart rate at scan, and age were modeled continuously and scaled as shown in the table. Multivariable logistic regression with diagnostic quality (overall image quality score ≥4) as the dependent variable and protocol type, weight (per 10 lb), maximum heart rate at scan (per 10 bpm), and age (per 10 years) as independent variables. CAM RG, continuous anatomically modulated retrospectively gated; CI, confidence interval; HR, heart rate; OR, odds ratio; PT, prospective triggering. Diagnostic-quality imaging in difficult-to-scan patients. Bar plot comparing the proportion of scans meeting diagnostic-quality criteria between patients who successfully underwent prospective triggering as planned (Group 1a) and patients for whom continuous anatomically modulated retrospectively gated scanning was deemed mandatory because of significant ectopy or elevated heart rate (Group 2b). Diagnostic-quality imaging was defined as an overall image quality score ≥4. Values above each bar indicate the number and percentage of patients with diagnostic-quality scans in each group. CAM RG, continuous anatomically modulated retrospectively gated; PT, prospective triggering. Discussion In this study of consecutive patients undergoing CCTA on a standard single-source scanner, we evaluated the use of CAM RG acquisition in patients with anticipated heart rate or rhythm instability. Our findings demonstrate that CAM RG enables the acquisition of consistently diagnostic-quality images in patients with more challenging heart rate profiles, with modestly higher radiation exposure that is strongly influenced by patient size. Importantly, these results were obtained on a scanner with modest gantry rotation time and without advanced hardware features such as dual-source imaging or wide-z-axis volume coverage, underscoring the potential utility of CAM RG in centers with older-generation CT technology. Our initial hypothesis that CAM mode would maintain IQ without increasing radiation dose was not supported. Although CAM RG maintained high rates of diagnostic-quality imaging, radiation dose was significantly higher than with PT scanning, including after multivariable adjustment. This likely reflects the fact that CAM was applied within a retrospectively gated acquisition, which inherently carries a higher radiation burden than prospective triggering, even when tube current is anatomically modulated. Accordingly, the present findings suggest that CAM may help preserve diagnostic imaging performance in difficult-to-scan patients, but it does not eliminate the dose penalty associated with retrospective gating. Prior studies have previously attempted to address the difficulty of imaging patients with ectopy or HHR. Andreini et al. (4) demonstrated in a study of 380 patients undergoing CCTA the utility of using a motion correction algorithm to obtain images of diagnostic quality. However, their study focused on post-processing solutions rather than protocol-level strategies to optimize radiation exposure during image acquisition. Similarly, Miller et al. found that the higher radiation doses in patients with HHRs can be mitigated with use of third generation dual-source scanners (5), however, in centers with older scanners the need for alternative techniques focused on software-based acquisition strategy such as those studied here are imperative. A key finding of this analysis is the dominant role of patient size in determining radiation dose, irrespective of scan protocol. Across all protocol groups, weight demonstrated a strong, graded association with dose, with DLP increasing by approximately 13–16% per 10-lb increase in body weight. These findings are consistent with prior work demonstrating the influence of body habitus on radiation exposure and highlight the importance of incorporating patient size into dose optimization strategies (6). CAM utilizes these principles to reduce unnecessary radiation exposure within a retrospectively gated acquisition by modulating radiation exposure according to anatomical positioning. However, our findings suggest that any dose savings achieved by the CAM algorithm were outweighed by the higher baseline radiation burden of retrospective gating relative to prospective triggering. Thus, the observed increase in dose likely reflects the combined effects of protocol selection for more challenging patients and the intrinsic radiation cost of retrospective gating, rather than failure of the CAM algorithm itself. Although CAM RG was associated with higher mean radiation dose compared with prospective triggering, IQ remained high across all groups. When applying a clinically relevant threshold for diagnostic quality (overall score ≥4), the proportion of diagnostic-quality studies was similarly high between protocols. These findings suggest that differences in continuous IQ scores are modest and may reflect underlying differences in heart rate and rhythm complexity rather than a true reduction in diagnostic interpretability. Taken together, these findings suggest that CAM RG represents a reasonable and pragmatic tradeoff—accepting a modest increase in dose to preserve diagnostic quality in patients who are suboptimal candidates for prospective triggering. This study has several limitations. Its single-center design introduces the potential for selection bias, as protocol assignment was based on clinician judgment rather than randomized criteria. As a result, patients undergoing CAM RG represented a more challenging cohort, with higher heart rates and greater anticipated difficulty in image acquisition. This fundamental difference between groups limits the ability to make direct causal comparisons of radiation dose and IQ between protocols. Although we performed multivariable adjustment for key confounders including heart rate, weight, and age, residual confounding may persist, and the observed differences should be interpreted in the context of this selection bias. IQ assessment was performed by consensus rather than independent blinded readers, and interobserver variability was not formally assessed. Additionally, the cohort was predominantly older, which may limit generalizability to younger populations with different patterns of heart rate variability. Finally, while this study provides robust descriptive and modeling data, it was not designed to establish equivalence or noninferiority between protocols. Despite these limitations, our findings demonstrate that with appropriate software-based modulation, older-generation CT scanners can be used effectively to image patients with challenging heart rate conditions while maintaining diagnostic IQ and acceptable radiation exposure. CAM RG offers a practical option for centers without access to advanced hardware and may enable more cost-effective utilization of existing imaging infrastructure. Conclusions In conclusion, this study demonstrates that CAM retrospectively-gated scanning enables the acquisition of diagnostic-quality CCTA images in patients with SE or HHRs on a standard single-source CT scanner, at the cost of modestly higher radiation exposure. By leveraging anatomical modulation to optimize tube current, CAM offers a viable alternative for centers lacking advanced hardware, promoting cost-effective utilization of existing equipment. However, the non-randomized design and reliance on subjective clinician judgments limit the findings’ robustness. Prospective, multi-center trials are needed to confirm these results across diverse patient populations and scanner types, potentially expanding CAM’s role in routine cardiovascular imaging. Supplementary The article’s supplementary files as Available at https://cdt.amegroups.com/article/view/10.21037/cdt-2026-1-0095/prf