Matching and Homogenizing Convolution Kernels for Quantitative Studies in Computed Tomography

Artificial intelligence-driven volumetric CT outcome score in cystic fibrosis: longitudinal and multicenter validation with/without modulators treatment

OBJECTIVES

Holistic segmentation of CT structural alterations with 3D deep learning has recently been described in cystic fibrosis (CF), allowing the measurement of normalized volumes of airway abnormalities (NOVAA-CT) as an automated quantitative outcome. Clinical validations are needed, including longitudinal and multicenter evaluations.

MATERIALS AND METHODS

The validation study was retrospective between 2010 and 2023. CF patients undergoing Elexacaftor/Tezacaftor/Ivacaftor (ETI) or corticosteroids for allergic broncho-pulmonary aspergillosis (ABPA) composed the monocenter ETI and ABPA groups, respectively. Patients from six geographically distinct institutions composed a multicenter external group. All patients had completed CT and pulmonary function test (PFT), with a second assessment at 1 year in case of ETI or ABPA treatment. NOVAA-CT quantified bronchiectasis, peribronchial thickening, bronchial mucus, bronchiolar mucus, collapse/consolidation, and their overall total abnormal volume (TAV). Two observers evaluated the visual Bhalla score.

RESULTS

A total of 139 CF patients (median age, 15 years [interquartile range: 13-25]) were evaluated. All correlations between NOVAA-CT to both PFT and Bhalla score were significant in the ETI (n = 60), ABPA (n = 20), and External groups (n = 59), such as the normalized TAV (ρ ≥ 0.76; p < 0.001). In both ETI and ABPA groups, there were significant longitudinal improvements in peribronchial thickening, bronchial mucus, bronchiolar mucus and collapse/consolidation (p ≤ 0.001). An additional reversibility in bronchiectasis volume was quantified with ETI (p < 0.001). Intraclass correlation coefficient of reproducibility was > 0.99.

CONCLUSION

NOVAA-CT automated scoring demonstrates validity, reliability and responsiveness for monitoring CF severity over an entire lung and quantifies therapeutic effects on lung structure at CT, such as the volumetric reversibility of airway abnormalities with ETI.

CLINICAL RELEVANCE STATEMENT

Normalized volume of airway abnormalities at CT automated 3D outcome enables objective, reproducible, and holistic monitoring of cystic fibrosis severity over an entire lung for management and endpoints during therapeutic trials.

KEY POINTS

Visual scoring methods lack sensitivity and reproducibility to assess longitudinal bronchial changes in cystic fibrosis (CF). AI-driven volumetric CT scoring correlates longitudinally to disease severity and reliably improves with Elexacaftor/Tezacaftor/Ivacaftor or corticosteroid treatments. AI-driven volumetric CT scoring enables reproducible monitoring of lung disease severity in CF and quantifies longitudinal structural therapeutic effects.

Preoperative CT-based radiomic prognostic index to predict the benefit of postoperative radiotherapy in patients with non-small cell lung cancer: a multicenter study

BACKGROUND

The value of postoperative radiotherapy (PORT) for patients with non-small cell lung cancer (NSCLC) remains controversial. A subset of patients may benefit from PORT. We aimed to identify patients with NSCLC who could benefit from PORT.

METHODS

Patients from cohorts 1 and 2 with pathological Tany N2 M0 NSCLC were included, as well as patients with non-metastatic NSCLC from cohorts 3 to 6. The radiomic prognostic index (RPI) was developed using radiomic texture features extracted from the primary lung nodule in preoperative chest CT scans in cohort 1 and validated in other cohorts. We employed a least absolute shrinkage and selection operator-Cox regularisation model for data dimension reduction, feature selection, and the construction of the RPI. We created a lymph-radiomic prognostic index (LRPI) by combining RPI and positive lymph node number (PLN). We compared the outcomes of patients who received PORT against those who did not in the subgroups determined by the LRPI.

RESULTS

In total, 228, 1003, 144, 422, 19, and 21 patients were eligible in cohorts 1-6. RPI predicted overall survival (OS) in all six cohorts: cohort 1 (HR = 2.31, 95% CI: 1.18-4.52), cohort 2 (HR = 1.64, 95% CI: 1.26-2.14), cohort 3 (HR = 2.53, 95% CI: 1.45-4.3), cohort 4 (HR = 1.24, 95% CI: 1.01-1.52), cohort 5 (HR = 2.56, 95% CI: 0.73-9.02), cohort 6 (HR = 2.30, 95% CI: 0.53-10.03). LRPI predicted OS (C-index: 0.68, 95% CI: 0.60-0.75) better than the pT stage (C-index: 0.57, 95% CI: 0.50-0.63), pT + PLN (C-index: 0.58, 95% CI: 0.46-0.70), and RPI (C-index: 0.65, 95% CI: 0.54-0.75). The LRPI was used to categorize individuals into three risk groups; patients in the moderate-risk group benefited from PORT (HR = 0.60, 95% CI: 0.40-0.91; p = 0.02), while patients in the low-risk and high-risk groups did not.

CONCLUSIONS

We developed preoperative CT-based radiomic and lymph-radiomic prognostic indexes capable of predicting OS and the benefits of PORT for patients with NSCLC.

Rank acquisition impact on radiomics estimation (AсquIRE) in chest CT imaging: A retrospective multi-site, multi-use-case study

BACKGROUND

Radiomics is a method within medical image analysis that involves the extraction of quantitative data from radiologic scans, often in conjunction with machine learning algorithms to phenotype disease appearance, prognosticate disease outcome, and predict treatment response. However, variance in CT scanner acquisition parameters, such as convolution kernels or pixel spacing, can impact radiomics texture feature values.

PURPOSE

The extent to which the parameters influence radiomics features continues to be an active area of investigation. In this study, we describe a novel approach, Acquisition Impact on Radiomics Estimation (AcquIRE), to rank the impact of CT acquisition parameters on radiomic texture features.

METHODS

In this work, we used three chest CT imaging datasets (n = 749 patients) from nine sites comprising: i) lung granulomas and adenocarcinomas (D1) (10 and 52 patients, respectively); ii) minimal and frank invasive adenocarcinoma (D2) (74 and 145 patients); and iii) early-stage NSCLC patients (D3) (315 patients). Datasets D2 and D3 were collected from four sites each, and D1 from a single site. For each patient, 744 texture features and nine acquisition parameters were extracted and utilized to evaluate which parameters impact radiomic features the most. The AcquIRE method establishes a relative assessment between acquisition parameters and radiomic texture featuresa through the creation of a classification model, which is then utilized to assess the rank of the acquisition parameters.

RESULTS

Across the use cases, CT software version and convolution kernel parameters were found to have the most variance. In D1, it was observed that the Haralick texture feature family was the least affected by variations in acquisition parameters, while the Gabor feature family was the most impacted. However, in datasets D2 and D3, the Gabor features were found to be the least affected. Our findings suggest that the impact on radiomic parameters is as much a function of the problem in question as it is acquisition parameters.

CONCLUSIONS

The software version and convolution kernel parameters impacted the radiomics feature the most.

Multi-organ segmentation of CT via convolutional neural network: impact of training setting and scanner manufacturer

Objective. Automated organ segmentation on CT images can enable the clinical use of advanced quantitative software devices, but model performance sensitivities must be understood before widespread adoption can occur. The goal of this study was to investigate performance differences between Convolutional Neural Networks (CNNs) trained to segment one (single-class) versus multiple (multi-class) organs, and between CNNs trained on scans from a single manufacturer versus multiple manufacturers.Methods. The multi-class CNN was trained on CT images obtained from 455 whole-body PET/CT scans (413 for training, 42 for testing) taken with Siemens, GE, and Phillips PET/CT scanners where 16 organs were segmented. The multi-class CNN was compared to 16 smaller single-class CNNs trained using the same data, but with segmentations of only one organ per model. In addition, CNNs trained on Siemens-only (N = 186) and GE-only (N = 219) scans (manufacturer-specific) were compared with CNNs trained on data from both Siemens and GE scanners (manufacturer-mixed). Segmentation performance was quantified using five performance metrics, including the Dice Similarity Coefficient (DSC).Results. The multi-class CNN performed well compared to previous studies, even in organs usually considered difficult auto-segmentation targets (e.g., pancreas, bowel). Segmentations from the multi-class CNN were significantly superior to those from smaller single-class CNNs in most organs, and the 16 single-class models took, on average, six times longer to segment all 16 organs compared to the single multi-class model. The manufacturer-mixed approach achieved minimally higher performance over the manufacturer-specific approach.Significance. A CNN trained on contours of multiple organs and CT data from multiple manufacturers yielded high-quality segmentations. Such a model is an essential enabler of image processing in a software device that quantifies and analyzes such data to determine a patient's treatment response. To date, this activity of whole organ segmentation has not been adopted due to the intense manual workload and time required.

Deep learning-based harmonization of CT reconstruction kernels towards improved clinical task performance

OBJECTIVES

To develop a deep learning-based harmonization framework, assessing whether it can improve performance of radiomics models given different kernels in different clinical tasks and additionally generalize to mitigate the effects of new/unobserved kernels on radiomics features.

METHODS

Patient data with 2 reconstruction kernels and phantom data with 22 reconstruction kernels were included. Eighty-five patients were studied for lymph node metastasis (LNM) prediction, and 164 patients for differential diagnosis between lung cancer (LC) and pulmonary tuberculosis (TB). Two convolutional neural network (CNN) models were developed to convert images (i) from B70f to B30f (CNNa) and (ii) from B30f to B70f (CNNb). Model performance between the two kernels was evaluated using AUC and compared with other well-known harmonization methods. Patient-normalized feature difference (PNFD) was used to identify the incompatible kernels (i.e., kernel with median PNFD > 1) with baseline (B30f/B70f), and measure the ability of the CNN models to convert the non-comparable kernels.

RESULTS

For LC versus pulmonary TB diagnosis, AUCs of CNNa vs. others were 0.85 vs. 0.54-0.74 (p = 0.0001-0.0003), and for CNNb vs. others: 0.87 vs. 0.54-0.86 (p = 0.0001-0.55). For LNM prediction, AUCs of CNNa vs. others were 0.68 vs. 0.56-0.61 (p = 0.10-0.39), and for CNNb vs. others: 0.78 vs. 0.70-0.73 (p = 0.07-0.40). After CNN harmonization, 17 of 20 (85%) of investigated unknown kernels produced comparable radiomics feature values relative to baseline (median PNFD from 1.10-2.31 to 0.23-1.13).

CONCLUSION

The CNN harmonization effectively improved performance of radiomics models between reconstruction kernels in different clinical tasks, and reduced feature differences between unknown kernels vs. baseline.

KEY POINTS

• The soft (B30f) and sharp (B70f) kernels strongly affect radiomics reproducibility and generalizability. • The convolutional neural network (CNN) harmonization methods performed better than location-scale (ComBat and centering-scaling) and matrix factorization harmonization methods (based on singular value decomposition (SVD) and independent component analysis (ICA)) in both clinical tasks. • The CNN harmonization methods improve feature reproducibility not only between specific kernels (B30f and B70f) from the same scanner, but also between unobserved kernels from different scanners of different vendors.

Evaluation of deep-learning image reconstruction for chest CT examinations at two different dose levels

AIMS

The aims of the present study were to, for both a full-dose protocol and an ultra-low dose (ULD) protocol, compare the image quality of chest CT examinations reconstructed using TrueFidelity (Standard kernel) with corresponding examinations reconstructed using ASIR-V (Lung kernel) and to evaluate if post-processing using an edge-enhancement filter affects the noise level, spatial resolution and subjective image quality of clinical images reconstructed using TrueFidelity.

METHODS

A total of 25 patients were examined with both a full-dose protocol and an ULD protocol using a GE Revolution APEX CT system (GE Healthcare, Milwaukee, USA). Three different reconstructions were included in the study: ASIR-V 40%, DLIR-H, and DLIR-H with additional post-processing using an edge-enhancement filter (DLIR-H + E2). Five observers assessed image quality in two separate visual grading characteristics (VGC) studies. The results from the studies were statistically analyzed using VGC Analyzer. Quantitative evaluations were based on determination of two-dimensional power spectrum (PS), contrast-to-noise ratio (CNR), and spatial resolution in the reconstructed patient images.

RESULTS

For both protocols, examinations reconstructed using TrueFidelity were statistically rated equal to or significantly higher than examinations reconstructed using ASIR-V 40%, but the ULD protocol benefitted more from TrueFidelity. In general, no differences in observer ratings were found between DLIR-H and DLIR-H + E2. For the three investigated image reconstruction methods, ASIR-V 40% showed highest noise and spatial resolution and DLIR-H the lowest, while the CNR was highest in DLIR-H and lowest in ASIR-V 40%.

CONCLUSION

The use of TrueFidelity for image reconstruction resulted in higher ratings on subjective image quality than ASIR-V 40%. The benefit of using TrueFidelity was larger for the ULD protocol than for the full-dose protocol. Post-processing of the TrueFidelity images using an edge-enhancement filter resulted in higher image noise and spatial resolution but did not affect the subjective image quality.

Impact of image filtering and assessment of volume-confounding effects on CT radiomic features and derived survival models in non-small cell lung cancer

Background

No evidence supports the choice of specific imaging filtering methodologies in radiomics. As the volume of the primary tumor is a well-recognized prognosticator, our purpose is to assess how filtering may impact the feature/volume dependency in computed tomography (CT) images of non-small cell lung cancer (NSCLC), and if such impact translates into differences in the performance of survival modeling. The role of lesion volume in model performances was also considered and discussed.

Methods

Four-hundred seventeen CT images NSCLC patients were retrieved from the NSCLC-Radiomics public repository. Pre-processing and features extraction were implemented using Pyradiomics v3.0.1. Features showing high correlation with volume across original and filtered images were excluded. Cox proportional hazards (PH) with least absolute shrinkage and selection operator (LASSO) regularization and CatBoost models were built with and without volume, and their concordance (C-) indices were compared using Wilcoxon signed-ranked test. The Mann Whitney U test was used to assess model performances after stratification into two groups based on low- and high-volume lesions.

Results

Radiomic models significantly outperformed models built on only clinical variables and volume. However, the exclusion/inclusion of volume did not generally alter the performances of radiomic models. Overall, performances were not substantially affected by the choice of either imaging filter (overall C-index 0.539-0.590 for Cox PH and 0.589-0.612 for CatBoost). The separation of patients with high-volume lesions resulted in significantly better performances in 2/10 and 7/10 cases for Cox PH and CatBoost models, respectively. Both low- and high-volume models performed significantly better with the inclusion of radiomic features (P<0.0001), but the improvement was largest in the high-volume group (+10.2% against +8.7% improvement for CatBoost models and +10.0% against +5.4% in Cox PH models).

Conclusions

Radiomic features complement well-known prognostic factors such as volume, but their volume-dependency is high and should be managed with vigilance. The informative content of radiomic features may be diminished in small lesion volumes, which could limit the applicability of radiomics in early-stage NSCLC, where tumors tend to be small. Our results also suggest an advantage of CatBoost models over the Cox PH models.

Toward the determination of sensitive and reliable whole-lung computed tomography features for robust standard radiomics and delta-radiomics analysis in a nonhuman primate model of coronavirus disease 2019

Purpose

We propose a method to identify sensitive and reliable whole-lung radiomic features from computed tomography (CT) images in a nonhuman primate model of coronavirus disease 2019 (COVID-19). Criteria used for feature selection in this method may improve the performance and robustness of predictive models.

Approach

Fourteen crab-eating macaques were assigned to two experimental groups and exposed to either severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) or a mock inoculum. High-resolution CT scans were acquired before exposure and on several post-exposure days. Lung volumes were segmented using a deep-learning methodology, and radiomic features were extracted from the original image. The reliability of each feature was assessed by the intraclass correlation coefficient (ICC) using the mock-exposed group data. The sensitivity of each feature was assessed using the virus-exposed group data by defining a factor R that estimates the excess of variation above the maximum normal variation computed in the mock-exposed group. R and ICC were used to rank features and identify non-sensitive and unstable features.

Results

Out of 111 radiomic features, 43% had excellent reliability ( ICC > 0.90 ), and 55% had either good ( ICC > 0.75 ) or moderate ( ICC > 0.50 ) reliability. Nineteen features were not sensitive to the radiological manifestations of SARS-CoV-2 exposure. The sensitivity of features showed patterns that suggested a correlation with the radiological manifestations.

Conclusions

Features were quantified and ranked based on their sensitivity and reliability. Features to be excluded to create more robust models were identified. Applicability to similar viral pneumonia studies is also possible.

Influence of a Deep Learning Noise Reduction on the CT Values, Image Noise and Characterization of Kidney and Ureter Stones

Deep-learning (DL) noise reduction techniques in computed tomography (CT) are expected to reduce the image noise while maintaining the clinically relevant information in reduced dose acquisitions. This study aimed to assess the size, attenuation, and objective image quality of reno-ureteric stones denoised using DL-software in comparison to traditionally reconstructed low-dose abdominal CT-images and evaluated its clinical impact. In this institutional review-board-approved retrospective study, 45 patients with renal and/or ureteral stones were included. All patients had undergone abdominal CT between August 2019 and October 2019. CT-images were reconstructed using the following three methods: filtered back-projection, iterative reconstruction, and PixelShine (DL-software) with both sharp and soft kernels. Stone size, CT attenuation, and objective image quality (signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR)) were evaluated and compared using Bonferroni-corrected Friedman tests. Objective image quality was measured in six regions-of-interest. Stone size ranged between 4.4 × 3.1−4.4 × 3.2 mm (sharp kernel) and 5.1 × 3.8−5.6 × 4.2 mm (soft kernel). Mean attenuation ranged between 704−717 Hounsfield Units (HU) (soft kernel) and 915−1047 HU (sharp kernel). Differences in measured stone sizes were ≤1.3 mm. DL-processed images resulted in significantly higher CNR and SNR values (p < 0.001) by decreasing image noise significantly (p < 0.001). DL-software significantly improved objective image quality while maintaining both correct stone size and CT-attenuation values. Therefore, the clinical impact of stone assessment in denoised image data sets remains unchanged. Through the relevant noise suppression, the software additionally offers the potential to further reduce radiation exposure.

Repeatability and Reproducibility of Computed Tomography Radiomics for Pulmonary Nodules: A Multicenter Phantom Study

BACKGROUND

Radiomics can yield minable information from medical images, which can facilitate computer-aided diagnosis. However, the lack of repeatability and reproducibility of radiomic features (RFs) may hinder their generalizability in clinical applications.

OBJECTIVES

The aims of this study were to explore 3 main sources of variability in RFs, investigate their influencing magnitudes and patterns, and identify a subset of robust RFs for further studies.

MATERIALS AND METHODS

A chest phantom with nodules was scanned with different computed tomography (CT) scanners repeatedly with varying acquisition and reconstruction parameters (April-May 2019) to evaluate 3 sources of variability: test-retest, inter-CT, and intra-CT protocol variability. The robustness of the RFs was measured using the concordance correlation coefficient, dynamic range, and intraclass correlation coefficient (ICC). The influencing magnitudes and patterns were analyzed using the Friedman test and Spearman rank correlation coefficient. Stable and informative RFs were selected, and their redundancy was eliminated using hierarchical clustering. Clinical validation was also performed to verify the clinical effectiveness and potential enhancement of the generalizability of radiomics research.

RESULTS

A total of 1295 RFs that showed all 3 sources of variability were included. The reconstruction kernel and the iteration level showed the greatest (ICC, 0.35 ± 0.31) and the least (ICC, 0.63 ± 0.27) influence on magnitudes. The different sources of variability showed relatively consistent patterns of influence (false discovery rate <0.001). Finally, we obtained a subset of 19 stable, informative, and nonredundant RFs under all 3 sources of variability. These RFs exhibited clinical effectiveness and showed better prediction performance than unstable RFs in the validation dataset (P = 0.017, Delong test).

CONCLUSIONS

The stability of RFs was affected to different degrees by test-retest and differences in CT manufacturers and models and CT acquisition and reconstruction parameters, but the influences of these factors showed relatively consistent patterns. We also obtained a subset of 19 stable, informative, and nonredundant RFs that should be preferably used to enhance the generalizability of further radiomics research.

Correction for Systematic Bias in Radiomics Measurements Due to Variation in Imaging Protocols

RATIONALE AND OBJECTIVES

The accuracy of measured radiomics features is affected by CT imaging protocols. This study aims to ascertain if applying bias corrections can improve the classification performance of the radiomics features.

MATERIALS AND METHODS

A cohort of 144 Non-Small Cell Lung Cancer patient CT images was used to calculate radiomics features for use in predictive models of patient pathological stage. Simulation models of the tumors, matched to patient lesion qualities of size, contrast, and degree of spiculation, were used to both create and assess protocol-specific correction factors. The usefulness of correction was first assessed by applying the corrections to simulated lesion phantoms with known properties using a corrected paired Student's t-test. The sensitivity of radiomics features to correction factors was assessed by applying a library of possible theoretical correction factors to the uncorrected radiomics from the patient data. The data were then used to assess the effect of the correction on prediction performance (AUC) from a logistic regression radiomics model across the patient cases.

RESULTS

The correction factors were shown to reduce the bias of radiomics features, caused by protocols, provided that the correction factors were derived from lesion models with similar properties. The sensitivity of the radiomics features to changes due to protocol effects was on average 89% among all features. The corrections applied to patient data resulted in a small increase of 0.0074 in AUC that was not statistically significant (p=0.60).

CONCLUSION

Protocol-specific correction factors can be applied to radiomics studies to control for biases introduced by different imaging protocols. The correction factors should ideally be lesion-specific, derived using lesion models that echo patient lesion characteristics in terms of size, contrast, and degree of spiculation. Small corrections in the 10% range offers only a small improvement in the predictability of radiomics.

A radiomics approach for lung nodule detection in thoracic CT images based on the dynamic patterns of morphological variation

OBJECTIVES

To propose and evaluate a set of radiomic features, called morphological dynamics features, for pulmonary nodule detection, which were rooted in the dynamic patterns of morphological variation and needless precise lesion segmentation.

MATERIALS AND METHODS

Two datasets were involved, namely, university hospital (UH) and LIDC datasets, comprising 72 CT scans (360 nodules) and 888 CT scans (2230 nodules), respectively. Each nodule was annotated by multiple radiologists. Denoted the category of nodules identified by at least k radiologists as ALk. A nodule detection algorithm, called CAD-MD algorithm, was proposed based on the morphological dynamics radiomic features, characterizing a lesion by ten sets of the same features with different values extracted from ten different thresholding results. Each nodule candidate was classified by a two-level classifier, including ten decision trees and a random forest, respectively. The CAD-MD algorithm was compared with a deep learning approach, the N-Net, using the UH dataset.

RESULTS

On the AL1 and AL2 of the UH dataset, the AUC of the AFROC curves were 0.777 and 0.851 for the CAD-MD algorithm and 0.478 and 0.472 for the N-Net, respectively. The CAD-MD algorithm achieved the sensitivities of 84.4% and 91.4% with 2.98 and 3.69 FPs/scan and the N-Net 74.4% and 80.7% with 3.90 and 4.49 FPs/scan, respectively. On the LIDC dataset, the CAD-MD algorithm attained the sensitivities of 87.6%, 89.2%, 92.2%, and 95.0% with 4 FPs/scan for AL1-AL4, respectively.

CONCLUSION

The morphological dynamics radiomic features might serve as an effective set of radiomic features for lung nodule detection.

KEY POINTS

• Texture features varied with such CT system settings as reconstruction kernels of CT images, CT scanner models, and parameter settings, and so on. • Shape and first-order statistics were shown to be the most robust features against variation in CT imaging parameters. • The morphological dynamics radiomic features, which mainly characterized the dynamic patterns of morphological variation, were shown to be effective for lung nodule detection.

Computed tomographic assessment of retrograde urohydropropulsion in male dogs and prediction of stone composition using Hounsfield unit in dogs and cats

Background

Persistent uroliths after a cystotomy in dogs are a common cause of surgical failure.

Objectives

This study examined the following: the success rate of retrograde urohydropropulsion in male dogs using non-enhanced computed tomography (CT), whether the CT mean beam attenuation values in Hounsfield Units (mHU) measured in vivo could predict the urolithiasis composition and whether the selected reconstruction kernel may influence the measured mHU.

Methods

All dogs and cats that presented with lower urinary tract uroliths and had a non-enhanced CT preceding surgery were included. In male dogs, CT was performed after retrograde urohydropropulsion to detect the remaining urethral calculi. The percentage and location of persistent calculi were recorded. The images were reconstructed using three kernels, from smooth to ultrasharp, and the calculi mHU were measured.

Results

Sixty-five patients were included in the study. The success rate of retrograde urohydropropulsion in the 45 male dogs was 55.6% and 86.7% at the first and second attempts, respectively. The predominant components of the calculi were cystine (20), struvite (15), calcium oxalate (8), and urate (7). The convolution kernel influenced the mHU values (p < 0.05). The difference in mHU regarding the calculus composition was better assessed using the smoother kernel. A mHU greater than 1,000 HU was predictive of calcium oxalate calculi.

Conclusions

Non-enhanced CT is useful for controlling the success of retrograde urohydropropulsion. The mHU could allow a prediction of the calculus composition, particularly for calcium oxalate, which may help determine the therapeutic strategy.

The Discriminative Power and Stability of Radiomics Features With Computed Tomography Variations: Task-Based Analysis in an Anthropomorphic 3D-Printed CT Phantom

OBJECTIVES

The aims of this study were to determine the stability of radiomics features against computed tomography (CT) parameter variations and to study their discriminative power concerning tissue classification using a 3D-printed CT phantom based on real patient data.

MATERIALS AND METHODS

A radiopaque 3D phantom was developed using real patient data and a potassium iodide solution paper-printing technique. Normal liver tissue and 3 lesion types (benign cyst, hemangioma, and metastasis) were manually annotated in the phantom. The stability and discriminative power of 86 radiomics features were assessed in measurements taken from 240 CT series with 8 parameter variations of reconstruction algorithms, reconstruction kernels, slice thickness, and slice spacing. Pairwise parameter group and pairwise tissue class comparisons were performed using Wilcoxon signed rank tests.

RESULTS

In total, 19,264 feature stability tests and 8256 discriminative power tests were performed. The 8 CT parameter variation pairwise group comparisons had statistically significant differences on average in 78/86 radiomics features. On the other hand, 84% of the univariate radiomics feature tests had a successful and statistically significant differentiation of the 4 classes of liver tissue. The 86 radiomics features were ranked according to the cumulative sum of successful stability and discriminative power tests.

CONCLUSIONS

The differences in radiomics feature values obtained from different types of liver tissue are generally greater than the intraclass differences resulting from CT parameter variations.

Harmonization of in-plane resolution in CT using multiple reconstructions from single acquisitions

PURPOSE

To providea methodology that removes the spatial variability of in-plane resolution using different CT reconstructions. The methodology does not require any training, sinogram, or specific reconstruction method.

METHODS

The methodology is formulated as a reconstruction problem. The desired sharp image is modeled as an unobservable variable to be estimated from an arbitrary number of observations with spatially variant resolution. The methodology comprises three steps: (1) density harmonization, which removes the density variability across reconstructions; (2) point spread function (PSF) estimation, which estimates a spatially variant PSF with arbitrary shape; (3) deconvolution, which is formulated as a regularized least squares problem. The assessment was performed with CT scans of phantoms acquired with three different Siemens scanners (Definition AS, Definition AS+, Drive). Four low-dose acquisitions reconstructed with backprojection and iterative methods were used for the resolution harmonization. A sharp, high-dose (HD) reconstruction was used as a validation reference. The different factors affecting the in-plane resolution (radial, angular, and longitudinal) were studied with regression analysis of the edge decay (between 10% and 90% of the edge spread function (ESF) amplitude).

RESULTS

Results showed that the in-plane resolution improves remarkably and the spatial variability is substantially reduced without compromising the noise characteristics. The modulated transfer function (MTF) also confirmed a pronounced increase in resolution. The resolution improvement was also tested by measuring the wall thickness of tubes simulating airways. In all scanners, the resolution harmonization obtained better performance than the HD, sharp reconstruction used as a reference (up to 50 percentage points). The methodology was also evaluated in clinical scans achieving a noise reduction and a clear improvement in thin-layered structures. The estimated ESF and MTF confirmed the resolution improvement.

CONCLUSION

We propose a versatile methodology to reduce the spatial variability of in-plane resolution in CT scans by leveraging different reconstructions available in clinical studies. The methodology does not require any sinogram, training, or specific reconstruction, and it is not limited to a fixed number of input images. Therefore, it can be easily adopted in multicenter studies and clinical practice. The results obtained with our resolution harmonization methodology evidence its suitability to reduce the spatially variant in-plane resolution in clinical CT scans without compromising the reconstruction's noise characteristics. We believe that the resolution increase achieved by our methodology may contribute in more accurate and reliable measurements of small structures such as vasculature, airways, and wall thickness.

Standardization of histogram- and GLCM-based radiomics in the presence of blur and noise

Radiomics have been extensively investigated as quantitative biomarkers that can enhance the utility of imaging studies and aid the clinical decision making process. A major challenge to the clinical translation of radiomics is their variability as a result of different imaging and reconstruction protocols. In this work, we present a novel radiomics standardization framework capable of modeling and recovering the underlying radiomic feature in images that have been corrupted by the effects of spatial resolution and noise. We focus on two classes of radiomics based on pixel value distributions - i.e., histograms and gray-level co-occurrence matrices. We developed a model that predicts these distributions in the presence of system blur and noise, and used that model to invert these physical effects and recover the underlying distributions. Specifically, the effect of blur on histogram and GLCM is highly image-dependent, while additive noise convolves the histogram/GLCM of the noiseless image with those of the noise. The recovery method therefore consists of two deconvolution operations: the first in the image domain to remove the effect of system blur, the second in the histogram/GLCM domain to remove the effect of noise. The performance of the proposed recovery strategy was investigated using a set of texture phantoms and an emulated CT imaging chain with a range of realistic blur and noise levels. The proposed method was able to obtain histogram and GLCM estimates that closely resemble the ground truth. The method performed well across imaging conditions and significantly lowered the variability associated with different imaging protocols. This improvement also translated to better classification accuracy, where recovered radiomic values results in greater separation of radiomic clusters for two different texture phantoms as compared to values derived from the original blurred and noisy images. In summary, the novel radiomics standardization framework demonstrates high potential for mitigating radiomic variability as a result of the imaging system and can potentially be integrated as a preprocessing step towards more robust and reproducible radiomic models.

Minimizing acquisition-related radiomics variability by image resampling and batch effect correction to allow for large-scale data analysis

OBJECTIVE

To identify CT-acquisition parameters accounting for radiomics variability and to develop a post-acquisition CT-image correction method to reduce variability and improve radiomics classification in both phantom and clinical applications.

METHODS

CT-acquisition protocols were prospectively tested in a phantom. The multi-centric retrospective clinical study included CT scans of patients with colorectal/renal cancer liver metastases. Ninety-three radiomics features of first order and texture were extracted. Intraclass correlation coefficients (ICCs) between CT-acquisition protocols were evaluated to define sources of variability. Voxel size, ComBat, and singular value decomposition (SVD) compensation methods were explored for reducing the radiomics variability. The number of robust features was compared before and after correction using two-proportion z test. The radiomics classification accuracy (K-means purity) was assessed before and after ComBat- and SVD-based correction.

RESULTS

Fifty-three acquisition protocols in 13 tissue densities were analyzed. Ninety-seven liver metastases from 43 patients with CT from two vendors were included. Pixel size, reconstruction slice spacing, convolution kernel, and acquisition slice thickness are relevant sources of radiomics variability with a percentage of robust features lower than 80%. Resampling to isometric voxels increased the number of robust features when images were acquired with different pixel sizes (p < 0.05). SVD-based for thickness correction and ComBat correction for thickness and combined thickness-kernel increased the number of reproducible features (p < 0.05). ComBat showed the highest improvement of radiomics-based classification in both the phantom and clinical applications (K-means purity 65.98 vs 73.20).

CONCLUSION

CT-image post-acquisition processing and radiomics normalization by means of batch effect correction allow for standardization of large-scale data analysis and improve the classification accuracy.

KEY POINTS

• The voxel size (accounting for the pixel size and slice spacing), slice thickness, and convolution kernel are relevant sources of CT-radiomics variability. • Voxel size resampling increased the mean percentage of robust CT-radiomics features from 59.50 to 89.25% when comparing CT scans acquired with different pixel sizes and from 71.62 to 82.58% when the scans were acquired with different slice spacings. • ComBat batch effect correction reduced the CT-radiomics variability secondary to the slice thickness and convolution kernel, improving the capacity of CT-radiomics to differentiate tissues (in the phantom application) and the primary tumor type from liver metastases (in the clinical application).

Impact of CT convolution kernel on robustness of radiomic features for different lung diseases and tissue types

OBJECTIVES

In this study, we aimed to assess the impact of different CT reconstruction kernels on the stability of radiomic features and the transferability between different diseases and tissue types. Three lung diseases were evaluated, i.e. non-small cell lung cancer (NSCLC), malignant pleural mesothelioma (MPM) and interstitial lung disease related to systemic sclerosis (SSc-ILD) as well as four different tissue types, i.e. primary tumor, largest involved lymph node ipsilateral and contralateral lung.

METHODS

Pre-treatment non-contrast enhanced CT scans from 23 NSCLC, 10 MPM and 12 SSc-ILD patients were collected retrospectively. For each patient, CT scans were reconstructed using smooth and sharp kernel in filtered back projection. The regions of interest (ROIs) were contoured on the smooth kernel-based CT and transferred to the sharp kernel-based CT. The voxels were resized to the largest voxel dimension of each cohort. In total, 1386 features were analyzed. Feature stability was assessed using the intraclass correlation coefficient. Features above the stability threshold >0.9 were considered stable.

RESULTS

We observed a strong impact of the reconstruction method on stability of the features (at maximum 26% of the 1386 features were stable). Intensity features were the most stable followed by texture and wavelet features. The wavelet features showed a positive correlation between percentage of stable features and size of the ROI (R2 = 0.79, p = 0.005). Lymph node radiomics showed poorest stability (<10%) and lung radiomics the largest stability (26%). Robustness analysis done on the contralateral lung could to a large extent be transferred to the ipsilateral lung, and the overlap of stable lung features between different lung diseases was more than 50%. However, results of robustness studies cannot be transferred between tissue types, which was investigated in NSCLC and MPM patients; the overlap of stable features for lymph node and lung, as well as for primary tumor and lymph node was very small in both disease types.

CONCLUSION

The robustness of radiomic features is strongly affected by different reconstruction kernels. The effect is largely influenced by the tissue type and less by the disease type.

ADVANCES IN KNOWLEDGE

The study presents to our knowledge the most complete analysis on the impact of convolution kernel on the robustness of CT-based radiomics for four relevant tissue types in three different lung diseases. .

Minimizing acquisition-related radiomics variability by image resampling and batch effect correction to allow for large-scale data analysis

Objective

To identify CT-acquisition parameters accounting for radiomics variability and to develop a post-acquisition CT-image correction method to reduce variability and improve radiomics classification in both phantom and clinical applications.

Methods

CT-acquisition protocols were prospectively tested in a phantom. The multi-centric retrospective clinical study included CT scans of patients with colorectal/renal cancer liver metastases. Ninety-three radiomics features of first order and texture were extracted. Intraclass correlation coefficients (ICCs) between CT-acquisition protocols were evaluated to define sources of variability. Voxel size, ComBat, and singular value decomposition (SVD) compensation methods were explored for reducing the radiomics variability. The number of robust features was compared before and after correction using two-proportion z test. The radiomics classification accuracy (K-means purity) was assessed before and after ComBat- and SVD-based correction.

Results

Fifty-three acquisition protocols in 13 tissue densities were analyzed. Ninety-seven liver metastases from 43 patients with CT from two vendors were included. Pixel size, reconstruction slice spacing, convolution kernel, and acquisition slice thickness are relevant sources of radiomics variability with a percentage of robust features lower than 80%. Resampling to isometric voxels increased the number of robust features when images were acquired with different pixel sizes (p < 0.05). SVD-based for thickness correction and ComBat correction for thickness and combined thickness–kernel increased the number of reproducible features (p < 0.05). ComBat showed the highest improvement of radiomics-based classification in both the phantom and clinical applications (K-means purity 65.98 vs 73.20).

Conclusion

CT-image post-acquisition processing and radiomics normalization by means of batch effect correction allow for standardization of large-scale data analysis and improve the classification accuracy.

Key Points

• The voxel size (accounting for the pixel size and slice spacing), slice thickness, and convolution kernel are relevant sources of CT-radiomics variability.

• Voxel size resampling increased the mean percentage of robust CT-radiomics features from 59.50 to 89.25% when comparing CT scans acquired with different pixel sizes and from 71.62 to 82.58% when the scans were acquired with different slice spacings.

• ComBat batch effect correction reduced the CT-radiomics variability secondary to the slice thickness and convolution kernel, improving the capacity of CT-radiomics to differentiate tissues (in the phantom application) and the primary tumor type from liver metastases (in the clinical application).

Electronic supplementary material

The online version of this article (10.1007/s00330-020-07174-0) contains supplementary material, which is available to authorized users.