Effect of emphysema on AI software and human reader performance in lung nodule detection from low-dose chest CT.

BACKGROUND: Emphysema influences the appearance of lung tissue in computed tomography (CT). We evaluated whether this affects lung nodule detection by artificial intelligence (AI) and human readers (HR). METHODS: Individuals were selected from the "Lifelines" cohort who had undergone low-dose chest CT. Nodules in individuals without emphysema were matched to similar-sized nodules in individuals with at least moderate emphysema. AI results for nodular findings of 30-100 mm3 and 101-300 mm3 were compared to those of HR; two expert radiologists blindly reviewed discrepancies. Sensitivity and false positives (FPs)/scan were compared for emphysema and non-emphysema groups. RESULTS: Thirty-nine participants with and 82 without emphysema were included (n = 121, aged 61 ± 8 years (mean ± standard deviation), 58/121 males (47.9%)). AI and HR detected 196 and 206 nodular findings, respectively, yielding 109 concordant nodules and 184 discrepancies, including 118 true nodules. For AI, sensitivity was 0.68 (95% confidence interval 0.57-0.77) in emphysema versus 0.71 (0.62-0.78) in non-emphysema, with FPs/scan 0.51 and 0.22, respectively (p = 0.028). For HR, sensitivity was 0.76 (0.65-0.84) and 0.80 (0.72-0.86), with FPs/scan of 0.15 and 0.27 (p = 0.230). Overall sensitivity was slightly higher for HR than for AI, but this difference disappeared after the exclusion of benign lymph nodes. FPs/scan were higher for AI in emphysema than in non-emphysema (p = 0.028), while FPs/scan for HR were higher than AI for 30-100 mm3 nodules in non-emphysema (p = 0.009). CONCLUSIONS: AI resulted in more FPs/scan in emphysema compared to non-emphysema, a difference not observed for HR. RELEVANCE STATEMENT: In the creation of a benchmark dataset to validate AI software for lung nodule detection, the inclusion of emphysema cases is important due to the additional number of FPs. KEY POINTS: • The sensitivity of nodule detection by AI was similar in emphysema and non-emphysema. • AI had more FPs/scan in emphysema compared to non-emphysema. • Sensitivity and FPs/scan by the human reader were comparable for emphysema and non-emphysema. • Emphysema and non-emphysema representation in benchmark dataset is important for validating AI.

CT-based emphysema characterization per lobe: A proof of concept.

PURPOSE: The Fleischner society criteria are global criteria to visually evaluate and classify pulmonary emphysema on CT. It may group heterogeneous disease severity within the same category, potentially obscuring clinically relevant differences in emphysema severity. This proof-of-concept study proposes to split emphysema into more categories and to assess each lobe separately, and applies this to two general population-based cohort samples to assess what information such an extension adds. METHOD: From a consecutive sample in two general population-based cohorts with low-dose chest CT, 117 participants with more than a trace of emphysema were included. Two independent readers performed an extended per-lobe classification and assessed overall severity semi-quantitatively. An emphysema sum score was determined by adding the severity score of all lobes. Inter-reader agreement was quantified with Krippendorff Alpha. RESULTS: Based on Fleischner society criteria, 69 cases had mild to severe centrilobular emphysema, and 90 cases had mild or moderate paraseptal emphysema (42 had both types of emphysema). The emphysema sum score was significantly different between mild (10.7 ± 4.3, range 2-22), moderate (20.1 ± 3.1, range: 15-24), and severe emphysema (23.6 ± 3.4, range: 17-28, p < 0.001), but ranges showed significant overlap. Inter-reader agreement for the extended classification and sum score was substantial (alpha 0.79 and 0.85, respectively). Distribution was homogenous across lobes in never-smokers, yet heterogenous in current smokers, with upper-lobe predominance. CONCLUSIONS: The proposed emphysema evaluation method adds information to the original Fleischner society classification. Individuals in the same Fleischner category have diverse emphysema sum scores, and lobar emphysema distribution differs between smoking groups.

Effect of Chest Computed Tomography Kernel Use on Emphysema Score in Severe Chronic Obstructive Pulmonary Disease Patients Evaluated for Lung Volume Reduction.

BACKGROUND: Chest computed tomography (CT) emphysema quantification is a vital diagnostic tool in patient evaluation for bronchoscopic lung volume reduction (BLVR). Smooth kernels for CT image reconstruction are generally recommended for quantitative analyses. This recommendation is not always followed, which may affect quantification of emphysema extent and eventually, treatment decisions. OBJECTIVE: The main goal is to demonstrate the influence of CT reconstruction kernels on emphysema quantification in patients with severe COPD, considered for BLVR. METHODS: Chest CT scans were acquired with one multi-detector CT system and reconstructed using three different kernels: smooth, medium smooth, and sharp. Other parameters were kept constant. Emphysema scores (ESs), meaning the percentage of voxels below -950 Hounsfield units, were calculated and compared to the smooth reference kernel using paired t tests. Bland-Altman plots were made to assess the biases and limits of agreement between kernels. RESULTS: Ninety-eight COPD patient CT scans were analyzed. The sharp kernel had a systematic bias of 6.2% and limits of agreement of 16.6% to -4.2% compared to the smooth kernel. The medium smooth kernel had a systematic bias of 5.7% and limits of agreement of 9.2% and 2.2% compared to the smooth kernel. The ES differed, for a single patient, up to 18% for different kernels. CONCLUSIONS: Chest CT kernel reconstruction can lead to a significant difference in emphysema severity quantification. This may cause invalid treatment selection in COPD patients evaluated for BLVR. Standardization of a smooth CT kernel setting and/or normalization to a standard kernel is strongly recommended.

Association between Chest CT-defined Emphysema and Lung Cancer: A Systematic Review and Meta-Analysis.

Background Given the different methods of assessing emphysema, controversy exists as to whether it is associated with lung cancer. Purpose To perform a systematic review and meta-analysis of the association between chest CT-defined emphysema and the presence of lung cancer. Materials and Methods The PubMed, Embase, and Cochrane databases were searched up to July 15, 2021, to identify studies on the association between emphysema assessed visually or quantitatively with CT and lung cancer. Associations were determined by emphysema severity (trace, mild, or moderate to severe, assessed visually and quantitatively) and subtype (centrilobular and paraseptal, assessed visually). Overall and stratified pooled odds ratios (ORs) with their 95% CIs were obtained. Results Of the 3343 screened studies, 21 studies (107 082 patients) with 26 subsets were included. The overall pooled ORs for lung cancer given the presence of emphysema were 2.3 (95% CI: 2.0, 2.6; I2 = 35%; 19 subsets) and 1.02 (95% CI: 1.01, 1.02; six subsets) per 1% increase in low attenuation area. Studies with visual (pooled OR, 2.3; 95% CI: 1.9, 2.6; I2 = 48%; 12 subsets) and quantitative (pooled OR, 2.2; 95% CI: 1.8, 2.8; I2 = 3.7%; eight subsets) assessments yielded comparable results for the dichotomous assessment. Based on six studies (1716 patients), the pooled ORs for lung cancer increased with emphysema severity and were higher for visual assessment (2.5, 3.7, and 4.5 for trace, mild, and moderate to severe, respectively) than for quantitative assessment (1.9, 2.2, and 2.5) based on point estimates. Compared with no emphysema, only centrilobular emphysema (three studies) was associated with lung cancer (pooled OR, 2.2; 95% CI: 1.5, 3.2; P < .001). Conclusion Both visual and quantitative CT assessments of emphysema were associated with a higher odds of lung cancer, which also increased with emphysema severity. Regarding subtype, only centrilobular emphysema was significantly associated with lung cancer. Clinical trial registration no. CRD42021262163 © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Hunsaker in this issue.

COPD identification and grading based on deep learning of lung parenchyma and bronchial wall in chest CT images.

OBJECTIVE: Chest CT can display the main pathogenic factors of chronic obstructive pulmonary disease (COPD), emphysema and airway wall remodeling. This study aims to establish deep convolutional neural network (CNN) models using these two imaging markers to diagnose and grade COPD. METHODS: Subjects who underwent chest CT and pulmonary function test (PFT) from one hospital (n = 373) were retrospectively included as the training cohort, and subjects from another hospital (n = 226) were used as the external test cohort. According to the PFT results, all subjects were labeled as Global Initiative for Chronic Obstructive Lung Disease (GOLD) Grade 1, 2, 3, 4 or normal. Two DenseNet-201 CNNs were trained using CT images of lung parenchyma and bronchial wall to generate two corresponding confidence levels to indicate the possibility of COPD, then combined with logistic regression analysis. Quantitative CT was used for comparison. RESULTS: In the test cohort, CNN achieved an area under the curve of 0.899 (95%CI: 0.853-0.935) to determine the existence of COPD, and an accuracy of 81.7% (76.2-86.7%), which was significantly higher than the accuracy 68.1% (61.6%-74.2%) using quantitative CT method (p < 0.05). For three-way (normal, GOLD 1-2, and GOLD 3-4) and five-way (normal, GOLD 1, 2, 3, and 4) classifications, CNN reached accuracies of 77.4 and 67.9%, respectively. CONCLUSION: CNN can identify emphysema and airway wall remodeling on CT images to infer lung function and determine the existence and severity of COPD. It provides an alternative way to detect COPD using the extensively available chest CT. ADVANCES IN KNOWLEDGE: CNN can identify the main pathological changes of COPD (emphysema and airway wall remodeling) based on CT images, to infer lung function and determine the existence and severity of COPD. CNN reached an area under the curve of 0.853 to determine the existence of COPD in the external test cohort. The CNN approach provides an alternative and effective way for early detection of COPD using extensively used chest CT, as an important alternative to pulmonary function test.

AI-Driven Model for Automatic Emphysema Detection in Low-Dose Computed Tomography Using Disease-Specific Augmentation.

The objective of this study is to evaluate the feasibility of a disease-specific deep learning (DL) model based on minimum intensity projection (minIP) for automated emphysema detection in low-dose computed tomography (LDCT) scans. LDCT scans of 240 individuals from a population-based cohort in the Netherlands (ImaLife study, mean age ± SD = 57 ± 6 years) were retrospectively chosen for training and internal validation of the DL model. For independent testing, LDCT scans of 125 individuals from a lung cancer screening cohort in the USA (NLST study, mean age ± SD = 64 ± 5 years) were used. Dichotomous emphysema diagnosis based on radiologists' annotation was used to develop the model. The automated model included minIP processing (slab thickness range: 1 mm to 11 mm), classification, and detection maps generation. The data-split for the pipeline evaluation involved class-balanced and imbalanced settings. The proposed DL pipeline showed the highest performance (area under receiver operating characteristics curve) for 11 mm slab thickness in both the balanced (ImaLife = 0.90 ± 0.05) and the imbalanced dataset (NLST = 0.77 ± 0.06). For ImaLife subcohort, the variation in minIP slab thickness from 1 to 11 mm increased the DL model's sensitivity from 75 to 88% and decreased the number of false-negative predictions from 10 to 5. The minIP-based DL model can automatically detect emphysema in LDCTs. The performance of thicker minIP slabs was better than that of thinner slabs. LDCT can be leveraged for emphysema detection by applying disease specific augmentation.

Improved precision of noise estimation in CT with a volume-based approach.

Assessment of image noise is a relevant issue in computed tomography (CT). Noise is routinely measured by the standard deviation of density values (Hounsfield units, HU) within a circular region of interest (ROI). We explored the effect of a spherical volume of interest (VOI) on noise measurements. Forty-nine chronic obstructive pulmonary disease patients underwent CT with clinical protocol (regular dose [RD], volumetric CT dose index [CTDIvol] 3.04 mGy, 64-slice unit), and ultra-low dose (ULD) protocol (median CTDIvol 0.38 mGy, dual-source unit). Noise was measured in 27 1-cm2 ROIs and 27 0.75-cm3 VOIs inside the trachea. Median true noise was 21 HU (range 17-29) for RD-CT and 33 HU (26-39) for ULD-CT. The VOI approach resulted in a lower mean distance between limits of agreement compared to ROI: 5.9 versus 10.0 HU for RD-CT (-40%); 4.7 versus 9.9 HU for ULD-CT (-53%). Mean systematic bias barely changed: -1.6 versus -0.9HU for RD-CT; 0.0 to 0.4HU for ULD-CT. The average measurement time was 6.8 s (ROI) versus 9.7 (VOI), independent of dose level. For chest CT, measuring noise with a VOI-based instead of a ROI-based approach reduces variability by 40-53%, without a relevant effect on systematic bias and measurement time.

Potential for dose reduction in CT emphysema densitometry with post-scan noise reduction: a phantom study.

OBJECTIVE: The aim of this phantom study was to investigate the effect of scan parameters and noise suppression techniques on the minimum radiation dose for acceptable image quality for CT emphysema densitometry. METHODS: The COPDGene phantom was scanned on a third generation dual-source CT system with 16 scan setups (CTDIvol 0.035-10.680 mGy). Images were reconstructed at 1.0/0.7 mm slice thickness/increment, with three kernels (one soft, two hard), filtered backprojection and three grades of third-generation iterative reconstruction (IR). Additionally, deep learning-based noise suppression software was applied. Main outcomes: overlap in area of the normalized histograms of CT density for the emphysema insert and lung material, and the radiation dose required for a maximum of 4.3% overlap (defined as acceptable image quality). RESULTS: In total, 384 scan reconstructions were analyzed. Decreasing radiation dose resulted in an exponential increase of the overlap in normalized histograms of CT density. The overlap was 11-91% for the lowest dose setting (CTDIvol 0.035mGy). The soft kernel reconstruction showed less histogram overlap than hard filter kernels. IR and noise suppression also reduced overlap. Using intermediate grade IR plus noise suppression software allowed for 85% radiation dose reduction while maintaining acceptable image quality. CONCLUSION: CT density histogram overlap can quantify the degree of discernibility of emphysema and healthy lung tissue. Noise suppression software, IR, and soft reconstruction kernels substantially decrease the dose required for acceptable image quality. ADVANCES IN KNOWLEDGE: Noise suppression software, IR, and soft reconstruction kernels allow radiation dose reduction by 85% while still allowing differentiation between emphysema and normal lung tissue.