Computer-aided detection of artificial pulmonary nodules using an ex vivo lung phantom: Influence of exposure parameters and iterative reconstruction

Influence of CT dose reduction on AI-driven malignancy estimation of incidental pulmonary nodules

OBJECTIVES

The purpose of this study was to determine the influence of dose reduction on a commercially available lung cancer prediction convolutional neuronal network (LCP-CNN).

METHODS

CT scans from a cohort provided by the local lung cancer center (n = 218) with confirmed pulmonary malignancies and their corresponding reduced dose simulations (25% and 5% dose) were subjected to the LCP-CNN. The resulting LCP scores (scale 1-10, increasing malignancy risk) and the proportion of correctly classified nodules were compared. The cohort was divided into a low-, medium-, and high-risk group based on the respective LCP scores; shifts between the groups were studied to evaluate the potential impact on nodule management. Two different malignancy risk score thresholds were analyzed: a higher threshold of ≥ 9 ("rule-in" approach) and a lower threshold of > 4 ("rule-out" approach).

RESULTS

In total, 169 patients with 196 nodules could be included (mean age ± SD, 64.5 ± 9.2 year; 49% females). Mean LCP scores for original, 25% and 5% dose levels were 8.5 ± 1.7, 8.4 ± 1.7 (p > 0.05 vs. original dose) and 8.2 ± 1.9 (p < 0.05 vs. original dose), respectively. The proportion of correctly classified nodules with the "rule-in" approach decreased with simulated dose reduction from 58.2 to 56.1% (p = 0.34) and to 52.0% for the respective dose levels (p = 0.01). For the "rule-out" approach the respective values were 95.9%, 96.4%, and 94.4% (p = 0.12). When reducing the original dose to 25%/5%, eight/twenty-two nodules shifted to a lower, five/seven nodules to a higher malignancy risk group.

CONCLUSION

CT dose reduction may affect the analyzed LCP-CNN regarding the classification of pulmonary malignancies and potentially alter pulmonary nodule management.

CLINICAL RELEVANCE STATEMENT

Utilization of a "rule-out" approach with a lower malignancy risk threshold prevents underestimation of the nodule malignancy risk for the analyzed software, especially in high-risk cohorts.

KEY POINTS

• LCP-CNN may be affected by CT image parameters such as noise resulting from low-dose CT acquisitions. • CT dose reduction can alter pulmonary nodule management recommendations by affecting the outcome of the LCP-CNN. • Utilization of a lower malignancy risk threshold prevents underestimation of pulmonary malignancies in high-risk cohorts.

"Will I change nodule management recommendations if I change my CAD system?"-impact of volumetric deviation between different CAD systems on lesion management

OBJECTIVES

To evaluate and compare the measurement accuracy of two different computer-aided diagnosis (CAD) systems regarding artificial pulmonary nodules and assess the clinical impact of volumetric inaccuracies in a phantom study.

METHODS

In this phantom study, 59 different phantom arrangements with 326 artificial nodules (178 solid, 148 ground-glass) were scanned at 80 kV, 100 kV, and 120 kV. Four different nodule diameters were used: 5 mm, 8 mm, 10 mm, and 12 mm. Scans were analyzed by a deep-learning (DL)-based CAD and a standard CAD system. Relative volumetric errors (RVE) of each system vs. ground truth and the relative volume difference (RVD) DL-based vs. standard CAD were calculated. The Bland-Altman method was used to define the limits of agreement (LOA). The hypothetical impact on LungRADS classification was assessed for both systems.

RESULTS

There was no difference between the three voltage groups regarding nodule volumetry. Regarding the solid nodules, the RVE of the 5-mm-, 8-mm-, 10-mm-, and 12-mm-size groups for the DL CAD/standard CAD were 12.2/2.8%, 1.3/ - 2.8%, - 3.6/1.5%, and - 12.2/ - 0.3%, respectively. The corresponding values for the ground-glass nodules (GGN) were 25.6%/81.0%, 9.0%/28.0%, 7.6/20.6%, and 6.8/21.2%. The mean RVD for solid nodules/GGN was 1.3/ - 15.2%. Regarding the LungRADS classification, 88.5% and 79.8% of all solid nodules were correctly assigned by the DL CAD and the standard CAD, respectively. 14.9% of the nodules were assigned differently between the systems.

CONCLUSIONS

Patient management may be affected by the volumetric inaccuracy of the CAD systems and hence demands supervision and/or manual correction by a radiologist.

KEY POINTS

• The DL-based CAD system was more accurate in the volumetry of GGN and less accurate regarding solid nodules than the standard CAD system. • Nodule size and attenuation have an effect on the measurement accuracy of both systems; tube voltage has no effect on measurement accuracy. • Measurement inaccuracies of CAD systems can have an impact on patient management, which demands supervision by radiologists.

How scan parameter choice affects deep learning-based coronary artery disease assessment from computed tomography

Recently, algorithms capable of assessing the severity of Coronary Artery Disease (CAD) in form of the Coronary Artery Disease-Reporting and Data System (CAD-RADS) grade from Coronary Computed Tomography Angiography (CCTA) scans using Deep Learning (DL) were proposed. Before considering to apply these algorithms in clinical practice, their robustness regarding different commonly used Computed Tomography (CT)-specific image formation parameters-including denoising strength, slab combination, and reconstruction kernel-needs to be evaluated. For this study, we reconstructed a data set of 500 patient CCTA scans under seven image formation parameter configurations. We select one default configuration and evaluate how varying individual parameters impacts the performance and stability of a typical algorithm for automated CAD assessment from CCTA. This algorithm consists of multiple preprocessing and a DL prediction step. We evaluate the influence of the parameter changes on the entire pipeline and additionally on only the DL step by propagating the centerline extraction results of the default configuration to all others. We consider the standard deviation of the CAD severity prediction grade difference between the default and variation configurations to assess the stability w.r.t. parameter changes. For the full pipeline we observe slight instability (± 0.226 CAD-RADS) for all variations. Predictions are more stable with centerlines propagated from the default to the variation configurations (± 0.122 CAD-RADS), especially for differing denoising strengths (± 0.046 CAD-RADS). However, stacking slabs with sharp boundaries instead of mixing slabs in overlapping regions (called true stack ± 0.313 CAD-RADS) and increasing the sharpness of the reconstruction kernel (± 0.150 CAD-RADS) leads to unstable predictions. Regarding the clinically relevant tasks of excluding CAD (called rule-out; AUC default 0.957, min 0.937) and excluding obstructive CAD (called hold-out; AUC default 0.971, min 0.964) the performance remains on a high level for all variations. Concluding, an influence of reconstruction parameters on the predictions is observed. Especially, scans reconstructed with the true stack parameter need to be treated with caution when using a DL-based method. Also, reconstruction kernels which are underrepresented in the training data increase the prediction uncertainty.

Diagnostic validation of a deep learning nodule detection algorithm in low-dose chest CT: determination of optimized dose thresholds in a virtual screening scenario

OBJECTIVES

This study was conducted to evaluate the effect of dose reduction on the performance of a deep learning (DL)-based computer-aided diagnosis (CAD) system regarding pulmonary nodule detection in a virtual screening scenario.

METHODS

Sixty-eight anthropomorphic chest phantoms were equipped with 329 nodules (150 ground glass, 179 solid) with four sizes (5 mm, 8 mm, 10 mm, 12 mm) and scanned with nine tube voltage/current combinations. The examinations were analyzed by a commercially available DL-based CAD system. The results were compared by a comparison of proportions. Logistic regression was performed to evaluate the impact of tube voltage, tube current, nodule size, nodule density, and nodule location.

RESULTS

The combination with the lowest effective dose (E) and unimpaired detection rate was 80 kV/50 mAs (sensitivity: 97.9%, mean false-positive rate (FPR): 1.9, mean CTDIvol: 1.2 ± 0.4 mGy, mean E: 0.66 mSv). Logistic regression revealed that tube voltage and current had the greatest impact on the detection rate, while nodule size and density had no significant influence.

CONCLUSIONS

The optimal tube voltage/current combination proposed in this study (80 kV/50 mAs) is comparable to the proposed combinations in similar studies, which mostly dealt with conventional CAD software. Modification of tube voltage and tube current has a significant impact on the performance of DL-based CAD software in pulmonary nodule detection regardless of their size and composition.

KEY POINTS

• Modification of tube voltage and tube current has a significant impact on the performance of deep learning-based CAD software. • Nodule size and composition have no significant impact on the software's performance. • The optimal tube voltage/current combination for the examined software is 80 kV/50 mAs.

Effect of New Model-Based Iterative Reconstruction on Quantitative Analysis of Airway Tree by Computer-Aided Detection Software in Chest Computed Tomography

OBJECTIVE

Compared the performance of computer-aided detection (CAD) software for quantitative analysis of airway using computed tomography (CT) images reconstructed with versions of model-based iterative reconstruction (MBIR) that either balances spatial and density resolution (MBIRSTND) or prefers spatial resolution (MBIRRP20), and adaptive statistical iterative reconstruction (ASIR) with lung kernel.

METHODS

Thirty patients were included who were scanned for pulmonary disease using a routine dose multidetector CT system. Data were reconstructed with ASIR, MBIRSTND, and MBIRRP20. Airway dimensions from the 3 reconstructions were measured using an automated, quantitative CAD software designed to segment and quantify the bronchial tree automatically using a skeletonization algorithm. For each patient and reconstruction algorithm, the right middle lobe bronchus was selected as a representative for measuring the bronchial length of the matched airways. Two radiologists used a semiquantitative 5-point scale to rate the subjective image quality of MBIRSTND and MBIRRP20 reconstructions on airway trees analysis.

RESULTS

Algorithm impacts the measurement variability of bronchus length in chest CT, MBIRRP20 were the best, whereas ASIR were the worst (P < 0.05). In addition, the optimal reconstruction algorithm was found to be MBIRSTND for the airway trees being assessed about subjective noise and MBIRRP20 about bronchial end shows, and there were no significant differences in the continuity and completeness of bronchial wall, whereas ASIR performed inferiorly compared with them (P < 0.05).

CONCLUSIONS

Compared with ASIR, MBIRSTND, and MBIRRP20 from MBIRn algorithm potentially allow the desired airway quantification accuracy to be achieved on the performance of CAD, especially for MBIRRP20.

Effect of CT reconstruction settings on the performance of a deep learning based lung nodule CAD system

PURPOSE

To study the effect of different reconstruction parameter settings on the performance of a commercially available deep learning based pulmonary nodule CAD system.

MATERIALS AND METHODS

We performed a retrospective analysis of 24 chest CT scans, reconstructed at 16 different reconstruction settings for two different iterative reconstruction algorithms (SAFIRE and ADMIRE) varying in slice thickness, kernel size and iterative reconstruction level strength using a commercially available deep learning pulmonary nodule CAD system. The DL-CAD software was evaluated at 25 different sensitivity threshold settings and nodules detected by the DL-CAD software were matched against a reference standard based on the consensus reading of three radiologists.

RESULTS

A total of 384 CT reconstructions was analysed from 24 patients, resulting in a total of 5786 found nodules. We matched the detected nodules against the reference standard, defined by a team of thoracic radiologists, and showed a gradual drop in recall, and an improvement in precision when the iterative strength levels were increased for a constant kernel size. The optimal DL-CAD threshold setting for use in our clinical workflow was found to be 0.88 with an F₂ of 0.73 ± 0.053.

CONCLUSIONS

The DL-CAD system behaves differently on IR data than on FBP data, there is a gradual drop in recall, and growth in precision when the iterative strength levels are increased. As a result, caution should be taken when implementing deep learning software in a hospital with multiple CT scanners and different reconstruction protocols. To the best of our knowledge, this is the first study that demonstrates this result from a DL-CAD system on clinical data.

Influence of acquisition settings and radiation exposure on CT lung densitometry-An anthropomorphic ex vivo phantom study

Objectives

To systematically evaluate the influence of acquisition settings in conjunction with raw-data based iterative image reconstruction (IR) on lung densitometry based on multi-row detector computed tomography (CT) in an anthropomorphic chest phantom.

Materials and methods

Ten porcine heart-lung explants were mounted in an ex vivo chest phantom shell, six with highly and four with low attenuating chest wall. CT (Somatom Definition Flash, Siemens Healthineers) was performed at 120kV_p and 80kV_p, each combined with current-time products of 120, 60, 30, and 12mAs, and was reconstructed with filtered back projection (FBP) and IR (Safire, Siemens Healthineers). Mean lung density (LD), air density (AD) and noise were measured by semi-automated region-of interest (ROI) analysis, with 120kV_p/120 mAs serving as the standard of reference.

Results

Using IR, noise in lung parenchyma was reduced by ~ 31% at high attenuating chest wall and by ~ 22% at low attenuating chest wall compared to FBP, respectively (p<0.05). IR induced changes in the order of ±1 HU to mean absolute LD and AD compared to corresponding FBP reconstructions which were statistically significant (p<0.05).

Conclusions

Densitometry is influenced by acquisition parameters and reconstruction algorithms to a degree that may be clinically negligible. However, in longitudinal studies and clinical research identical protocols and potentially other measures for calibration may be required.

Radiation dose reduction techniques for chest CT: Principles and clinical results

Computer tomography plays a major role in the evaluation of thoracic diseases, especially since the advent of the multidetector-row CT (MDCT) technology. However, the increase use of this technique has raised some concerns about the resulting radiation dose. In this review, we will present the various methods allowing limiting the radiation dose exposure resulting from chest CT acquisitions, including the options of image filtering and iterative reconstruction (IR) algorithms. The clinical applications of reduced dose protocols will be reviewed, especially for lung nodule detection and diagnosis of pulmonary thromboembolism. The performance of reduced dose protocols for infiltrative lung disease assessment will also be discussed. Lastly, the influence of using IR algorithms on computer-aided detection and volumetry of lung nodules, as well as on quantitative and functional assessment of chest diseases will be presented and discussed.

Quantitative Analysis of Airway Tree in Low-dose Chest CT with a New Model-based Iterative Reconstruction Algorithm: Comparison to Adaptive Statistical Iterative Reconstruction in Routine-dose CT

OBJECTIVE

We aimed to evaluate a new model-based iterative reconstruction (MBIRn) algorithm either with spatial resolution and noise reduction balance (MBIR_STND) or spatial resolution preference (MBIR_RP20) for quantitative analysis of airway in low-dose chest computed tomography (CT) with a computer-aided detection (CAD) software, in comparison to adaptive statistical iterative reconstruction (ASIR) in routine-dose CT.

METHODS

Thirty patients who underwent both the routine-dose (noise index [NI] = 14 HU) and low-dose (at 30% level with NI = 28 HU) CT examination for pulmonary disease were included. Image acquisition was performed with 120 kVp tube voltage and automatic tube current modulation. Routine-dose scans were reconstructed with ASIR, whereas low-dose scans were reconstructed with ASIR, MBIR_STND, and MBIR_RP20. Airway dimensions of the right middle lobe bronchus from the four reconstructions were measured using CAD software. Two radiologists used a semiquantitative 5 scoring criteria (-2, inferior to; +2, superior to; -1 slightly inferior to; +1, slightly superior to; and 0, equal to ASIR in routine-dose CT) to rate the subjective image quality of MBIR_STND and MBIR_RP20 of airway trees. The paired t test and Wilcoxon signed-rank test were used for statistical comparison.

RESULTS

The low-dose CT provided 70.76% dose reduction compared to the routine-dose CT (0.88 ± 0.83 mSv vs 3.01 ± 1.89 mSv). MBIR_STND and MBIR_RP20 with low-dose CT provided longer bronchial length measurements and were better in measurement variability and continuity and completeness of bronchial walls than ASIR in routine-dose CT (P < .05). MBIR_STND was better for subjective noise and MBIR_RP20 for showing distal branches_. CONCLUSIONS: MBIR_STND and MBIR_RP20 algorithms provide better airway quantification at 30% of the radiation dose, compared to ASIR at routine-dose CT.

Towards synchrotron phase-contrast lung imaging in patients - a proof-of-concept study on porcine lungs in a human-scale chest phantom

In-line free propagation phase-contrast synchrotron tomography of the lungs has been shown to provide superior image quality compared with attenuation-based computed tomography (CT) in small-animal studies. The present study was performed to prove the applicability on a human-patient scale using a chest phantom with ventilated fresh porcine lungs. Local areas of interest were imaged with a pixel size of 100 µm, yielding a high-resolution depiction of anatomical hallmarks of healthy lungs and artificial lung nodules. Details like fine spiculations into surrounding alveolar spaces were shown on a micrometre scale. Minor differences in artificial lung nodule density were detected by phase retrieval. Since we only applied a fraction of the X-ray dose used for clinical high-resolution CT scans, it is believed that this approach may become applicable to the detailed assessment of focal lung lesions in patients in the future.

Detection of artificial pulmonary lung nodules in ultralow-dose CT using an ex vivo lung phantom

Objectives

To assess the image quality of 3 different ultralow-dose CT protocols on pulmonary nodule depiction in a ventilated ex vivo-system.

Materials and methods

Four porcine lungs were inflated inside a dedicated chest phantom and prepared with n = 195 artificial nodules (0.5–1 mL). The artificial chest wall was filled with water to simulate the absorption of a human chest. Images were acquired with a 2x192-row detector CT using low-dose (reference protocol with a tube voltage of 120 kV) and 3 different ULD protocols (respective effective doses: 1mSv and 0.1mSv). A different tube voltage was used for each ULD protocol: 70kV, 100kV with tin filter (100kV_Sn) and 150kV with tin filter (150kV_Sn). Nodule delineation was assessed by two observers (scores 1–5, 1 = unsure, 5 = high confidence).

Results

The diameter of the 195 detected artificial nodules ranged from 0.9–21.5 mm (mean 7.84 mm ± 5.31). The best ULD scores were achieved using 100kV_Sn and 70 kV ULD protocols (4.14 and 4.06 respectively). Both protocols were not significantly different (p = 0.244).

The mean score of 3.78 in ULD 150kV_Sn was significantly lower compared to the 100kV_Sn ULD protocol (p = 0.008).

Conclusion

The results of this experiment, conducted in a realistic setting show the feasibility of ultralow-dose CT for the detection of pulmonary nodules.

Lung nodules: size still matters

The incidence of indeterminate pulmonary nodules has risen constantly over the past few years. Determination of lung nodule malignancy is pivotal, because the early diagnosis of lung cancer could lead to a definitive intervention. According to the current international guidelines, size and growth rate represent the main indicators to determine the nature of a pulmonary nodule. However, there are some limitations in evaluating and characterising nodules when only their dimensions are taken into account. There is no single method for measuring nodules, and intrinsic errors, which can determine variations in nodule measurement and in growth assessment, do exist when performing measurements either manually or with automated or semi-automated methods. When considering subsolid nodules the presence and size of a solid component is the major determinant of malignancy and nodule management, as reported in the latest guidelines. Nevertheless, other nodule morphological characteristics have been associated with an increased risk of malignancy. In addition, the clinical context should not be overlooked in determining the probability of malignancy. Predictive models have been proposed as a potential means to overcome the limitations of a sized-based assessment of the malignancy risk for indeterminate pulmonary nodules.

Generation of realistic virtual nodules based on three-dimensional spatial resolution in lung computed tomography: A pilot phantom study

PURPOSE

The aim of this feasibility study using phantoms was to propose a novel method for obtaining computer-generated realistic virtual nodules in lung computed tomography (CT).

METHODS

In the proposed methodology, pulmonary nodule images obtained with a CT scanner are deconvolved with the point spread function (PSF) in the scan plane and slice sensitivity profile (SSP) measured for the scanner; the resultant images are referred to as nodule-like object functions. Next, by convolving the nodule-like object function with the PSF and SSP of another (target) scanner, the virtual nodule can be generated so that it has the characteristics of the spatial resolution of the target scanner. To validate the methodology, the authors applied physical nodules of 5-, 7- and 10-mm-diameter (uniform spheres) included in a commercial CT test phantom. The nodule-like object functions were calculated from the sphere images obtained with two scanners (Scanner A and Scanner B); these functions were referred to as nodule-like object functions A and B, respectively. From these, virtual nodules were generated based on the spatial resolution of another scanner (Scanner C). By investigating the agreement of the virtual nodules generated from the nodule-like object functions A and B, the equivalence of the nodule-like object functions obtained from different scanners could be assessed. In addition, these virtual nodules were compared with the real (true) sphere images obtained with Scanner C. As a practical validation, five types of laboratory-made physical nodules with various complicated shapes and heterogeneous densities, similar to real lesions, were used. The nodule-like object functions were calculated from the images of these laboratory-made nodules obtained with Scanner A. From them, virtual nodules were generated based on the spatial resolution of Scanner C and compared with the real images of laboratory-made nodules obtained with Scanner C.

RESULTS

Good agreement of the virtual nodules generated from the nodule-like object functions A and B of the phantom spheres was found, suggesting the validity of the nodule-like object functions. The virtual nodules generated from the nodule-like object function A of the phantom spheres were similar to the real images obtained with Scanner C; the root mean square errors (RMSEs) between them were 10.8, 11.1, and 12.5 Hounsfield units (HU) for 5-, 7-, and 10-mm-diameter spheres, respectively. The equivalent results (RMSEs) using the nodule-like object function B were 15.9, 16.8, and 16.5 HU, respectively. These RMSEs were small considering the high contrast between the sphere density and background density (approximately 674 HU). The virtual nodules generated from the nodule-like object functions of the five laboratory-made nodules were similar to the real images obtained with Scanner C; the RMSEs between them ranged from 6.2 to 8.6 HU in five cases.

CONCLUSIONS

The nodule-like object functions calculated from real nodule images would be effective to generate realistic virtual nodules. The proposed method would be feasible for generating virtual nodules that have the characteristics of the spatial resolution of the CT system used in each institution, allowing for site-specific nodule generation.

Influence of exposure parameters and iterative reconstruction on automatic airway segmentation and analysis on MDCT-An ex vivo phantom study

OBJECTIVES

To evaluate the influence of exposure parameters and raw-data-based iterative reconstruction (IR) on computer-aided segmentation and quantitative analysis of the tracheobronchial tree on multidetector computed tomography (MDCT).

MATERIAL AND METHODS

10 porcine heart-lung-explants were mounted inside a dedicated chest phantom. MDCT was performed at 120kV and 80kV with 120, 60, 30 and 12 mAs each. All scans were reconstructed with filtered back projection (FBP) or IR, resulting in a total of 160 datasets. The maximum number of detected airway segments, most peripheral airway generation detected, generation-specific airway wall thickness (WT), total diameter (TD) and normalized wall thickness (pi10) were compared.

RESULTS

The number of detected airway segments decreased slightly with dose (324.8±118 at 120kV/120mAs vs. 288.9±130 at 80kV/30mAs with FBP, p<0.05) and was not changed by IR. The 20th generation was constantly detected as most peripheral. WT did not change significantly with exposure parameters and reconstruction algorithm across all generations: range 1st generation 2.4-2.7mm, 5th 1.0-1.1mm, and 10th 0.7mm with FBP; 1st 2.3-2.4mm, 5th 1.0-1.1mm, and 10th 0.7-0.8mm with IR. pi10 was not affected as well (range 0.32-0.34mm).

CONCLUSIONS

Exposure parameters and IR had no relevant influence on measured airway parameters even for WT <1mm. Thus, no systematic errors would be expected using automatic airway analysis with low-dose MDCT and IR.

The effect of radiation dose reduction on computer-aided detection (CAD) performance in a low-dose lung cancer screening population

PURPOSE

Lung cancer screening with low-dose CT has recently been approved for reimbursement, heralding the arrival of such screening services worldwide. Computer-aided detection (CAD) tools offer the potential to assist radiologists in detecting nodules in these screening exams. In lung screening, as in all CT exams, there is interest in further reducing radiation dose. However, the effects of continued dose reduction on CAD performance are not fully understood. In this work, we investigated the effect of reducing radiation dose on CAD lung nodule detection performance in a screening population.

METHODS

The raw projection data files were collected from 481 patients who underwent low-dose screening CT exams at our institution as part of the National Lung Screening Trial (NLST). All scans were performed on a multidetector scanner (Sensation 64, Siemens Healthcare, Forchheim Germany) according to the NLST protocol, which called for a fixed tube current scan of 25 effective mAs for standard-sized patients and 40 effective mAs for larger patients. The raw projection data were input to a reduced-dose simulation software to create simulated reduced-dose scans corresponding to 50% and 25% of the original protocols. All raw data files were reconstructed at the scanner with 1 mm slice thickness and B50 kernel. The lungs were segmented semi-automatically, and all images and segmentations were input to an in-house CAD algorithm trained on higher dose scans (75-300 mAs). CAD findings were compared to a reference standard generated by an experienced reader. Nodule- and patient-level sensitivities were calculated along with false positives per scan, all of which were evaluated in terms of the relative change with respect to dose. Nodules were subdivided based on size and solidity into categories analogous to the LungRADS assessment categories, and sub-analyses were performed.

RESULTS

From the 481 patients in this study, 82 had at least one nodule (prevalence of 17%) and 399 did not (83%). A total of 118 nodules were identified. Twenty-seven nodules (23%) corresponded to LungRADS category 4 based on size and composition, while 18 (15%) corresponded to LungRADS category 3 and 73 (61%) corresponded to LungRADS category 2. For solid nodules ≥8 mm, patient-level median sensitivities were 100% at all three dose levels, and mean sensitivities were 72%, 63%, and 63% at original, 50%, and 25% dose, respectively. Overall mean patient-level sensitivities for nodules ranging from 3 to 45 mm were 38%, 37%, and 38% at original, 50%, and 25% dose due to the prevalence of smaller nodules and nonsolid nodules in our reference standard. The mean false-positive rates were 3, 5, and 13 per case.

CONCLUSIONS

CAD sensitivity decreased very slightly for larger nodules as dose was reduced, indicating that reducing the dose to 50% of original levels may be investigated further for use in CT screening. However, the effect of dose was small relative to the effect of the nodule size and solidity characteristics. The number of false positives per scan increased substantially at 25% dose, illustrating the importance of tuning CAD algorithms to very challenging, high-noise screening exams.

Effects of Iterative Reconstruction Algorithms on Computer-assisted Detection (CAD) Software for Lung Nodules in Ultra-low-dose CT for Lung Cancer Screening

RATIONALE AND OBJECTIVES

This study aimed to evaluate the effects of iterative reconstruction (IR) algorithms on computer-assisted detection (CAD) software for lung nodules in ultra-low-dose computed tomography (ULD-CT) for lung cancer screening.

MATERIALS AND METHODS

We selected 85 subjects who underwent both a low-dose CT (LD-CT) scan and an additional ULD-CT scan in our lung cancer screening program for high-risk populations. The LD-CT scans were reconstructed with filtered back projection (FBP; LD-FBP). The ULD-CT scans were reconstructed with FBP (ULD-FBP), adaptive iterative dose reduction 3D (AIDR 3D; ULD-AIDR 3D), and forward projected model-based IR solution (FIRST; ULD-FIRST). CAD software for lung nodules was applied to each image dataset, and the performance of the CAD software was compared among the different IR algorithms.

RESULTS

The mean volume CT dose indexes were 3.02 mGy (LD-CT) and 0.30 mGy (ULD-CT). For overall nodules, the sensitivities of CAD software at 3.0 false positives per case were 78.7% (LD-FBP), 9.3% (ULD-FBP), 69.4% (ULD-AIDR 3D), and 77.8% (ULD-FIRST). Statistical analysis showed that the sensitivities of ULD-AIDR 3D and ULD-FIRST were significantly higher than that of ULD-FBP (P < .001).

CONCLUSIONS

The performance of CAD software in ULD-CT was improved by using IR algorithms. In particular, the performance of CAD in ULD-FIRST was almost equivalent to that in LD-FBP.

A method for evaluating the performance of computer-aided detection of pulmonary nodules in lung cancer CT screening: detection limit for nodule size and density

OBJECTIVE

We propose the application of virtual nodules to evaluate the performance of computer-aided detection (CAD) of lung nodules in cancer screening using low-dose CT.

METHODS

The virtual nodules were generated based on the spatial resolution measured for a CT system used in an institution providing cancer screening and were fused into clinical lung images obtained at that institution, allowing site specificity. First, we validated virtual nodules as an alternative to artificial nodules inserted into a phantom. In addition, we compared the results of CAD analysis between the real nodules (n = 6) and the corresponding virtual nodules. Subsequently, virtual nodules of various sizes and contrasts between nodule density and background density (ΔCT) were inserted into clinical images (n = 10) and submitted for CAD analysis.

RESULTS

In the validation study, 46 of 48 virtual nodules had the same CAD results as artificial nodules (kappa coefficient = 0.913). Real nodules and the corresponding virtual nodules showed the same CAD results. The detection limits of the tested CAD system were determined in terms of size and density of peripheral lung nodules; we demonstrated that a nodule with a 5-mm diameter was detected when the nodule had a ΔCT > 220 HU.

CONCLUSION

Virtual nodules are effective in evaluating CAD performance using site-specific scan/reconstruction conditions. Advances in knowledge: Virtual nodules can be an effective means of evaluating site-specific CAD performance. The methodology for guiding the detection limit for nodule size/density might be a useful evaluation strategy.

Advanced imaging tools in pulmonary nodule detection and surveillance

Lung cancer is a leading cause of death worldwide. The National Lung Screening Trial has demonstrated that lung cancer screening can reduce lung cancer specific and all cause mortality. With approval of national coverage for lung cancer screening, it is expected that an increase in exams related to pulmonary nodule detection and surveillance will ensue. Advanced imaging technologies for nodule detection and surveillance will be more important than ever. While computed tomography (CT) remains the modality of choice, other emerging modalities such as magnetic resonance imaging provides viable alternatives to CT.

Effect of radiation dose reduction and iterative reconstruction on computer-aided detection of pulmonary nodules: Intra-individual comparison

OBJECTIVE

To evaluate the effect of radiation dose reduction and iterative reconstruction (IR) on the performance of computer-aided detection (CAD) for pulmonary nodules.

METHODS

In this prospective study twenty-five patients were included who were scanned for pulmonary nodule follow-up. Image acquisition was performed at routine dose and three reduced dose levels in a single session by decreasing mAs-values with 45%, 60% and 75%. Tube voltage was fixed at 120 kVp for patients ≥ 80 kg and 100 kVp for patients < 80 kg. Data were reconstructed with filtered back projection (FBP), iDose(4) (levels 1,4,6) and IMR (levels 1-3). All noncalcified solid pulmonary nodules ≥ 4 mm identified by two radiologists in consensus served as the reference standard. Subsequently, nodule volume was measured with CAD software and compared to the reference consensus. The numbers of true-positives, false-positives and missed pulmonary nodules were evaluated as well as the sensitivity.

RESULTS

Median effective radiation dose was 2.2 mSv at routine dose and 1.2, 0.9 and 0.6 mSv at respectively 45%, 60% and 75% reduced dose. A total of 28 pulmonary nodules were included. With FBP at routine dose, 89% (25/28) of the nodules were correctly identified by CAD. This was similar at reduced dose levels with FBP, iDose(4) and IMR. CAD resulted in a median number of false-positives findings of 11 per scan with FBP at routine dose (93% of the CAD marks) increasing to 15 per scan with iDose(4) (95% of the CAD marks) and 26 per scan (96% of the CAD marks) with IMR at the lowest dose level.

CONCLUSION

CAD can identify pulmonary nodules at submillisievert dose levels with FBP, hybrid and model-based IR. However, the number of false-positive findings increased using hybrid and especially model-based IR at submillisievert dose while dose reduction did not affect the number of false-positives with FBP.