PET textural features stability and pattern discrimination power for radiomics analysis: An “ad-hoc” phantoms study

PurposeThe analysis of PET images by textural features, also known as radiomics, shows promising results in tumor characterization. However, radiomic metrics (RMs) analysis is currently not standardized and the impact of the whole processing chain still needs deep investigation. We characterized the impact on RM values of: i) two discretization methods, ii) acquisition statistics, and iii) reconstruction algorithm. The influence of tumor volume and standardized-uptake-value (SUV) on RM was also investigated.MethodsThe Chang-Gung-Image-Texture-Analysis (CGITA) software was used to calculate 39 RMs using phantom data. Thirty noise realizations were acquired to measure statistical effect size indicators for each RM. The parameter η² (fraction of variance explained by the nuisance factor) was used to assess the effect of categorical variables, considering η² < 20% and 20% < η² < 40% as representative of a “negligible” and a “small” dependence respectively. The Cohen’s d was used as discriminatory power to quantify the separation of two distributions.ResultsWe found the discretization method based on fixed-bin-number (FBN) to outperform the one based on fixed-bin-size in units of SUV (FBS), as the latter shows a higher SUV dependence, with 30 RMs showing η² > 20%. FBN was also less influenced by the acquisition and reconstruction setup: with FBN 37 RMs had η² < 40%, only 20 with FBS. Most RMs showed a good discriminatory power among heterogeneous PET signals (for FBN: 29 out of 39 RMs with d > 3).ConclusionsFor RMs analysis, FBN should be preferred. A group of 21 RMs was suggested for PET radiomics analysis.     

Robustness of CT radiomic features against image discretization and interpolation in characterizing pancreatic neuroendocrine neoplasms

PurposeTo explore the variation of the discriminative power of CT radiomic features (RF) against image discretization/interpolation in characterizing pancreatic neuro-endocrine (PanNEN) neoplasms.Materials and methodsThirty-nine PanNEN patients with pre-surgical high contrast CT available were considered. Image interpolation and discretization parameters were intentionally changed, including pixel size (0.73–2.19 mm²), slice thickness (2–5 mm) and binning (32–128 grey levels) and their combination generated 27 parameter’s set. The ability of 69 RF in discriminating post-surgically assessed tumor grade (>G1), positive nodes, metastases and vascular invasion was tested: AUC changes when changing the parameters were quantified for selected RF, significantly associated to each end-point. The analysis was repeated for the corresponding images with contrast medium and in a sub-group of 29/39 patients scanned on a single scanner.ResultsThe median tumor volume was 1.57 cm³ (16%-84% percentiles: 0.62–34.58 cm³). RF variability against discretization/interpolation parameters was large: only 21/69 RF showed %COV < 20%. Despite this variability, AUC changes were limited for all end-points: with typical AUC values around 0.75–0.85, AUC ranges for the 27 parameter’s set were on average 0.062 (1SD:0.037) for all end-points with maximum %COV equal to 5.5% (mean:2.3%). Performances significantly improved when excluding the 5 mm thickness case and fixing the binning to 64 (mean AUC range: 0.036, 1SD:0.019). Using contrast images or limiting the population to single-scanner patients had limited impact on AUC variability.ConclusionsThe discriminative power of CT RF for panNEN is relatively invariant against image interpolation/discretization within a large range of voxel sizes and binning.     

The effect of tube focal spot size and acquisition mode on task-based image quality performance of a GE revolution HD dual energy CT scanner

PurposeTo assess the task-based performance of images obtained under different focal spot size and acquisition mode on a dual-energy CT scanner.MethodsAxial CT image series of the Catphan phantom were obtained using a tube focus at different sizes. Acquisitions were performed in standard single-energy, high resolution (HR) and dual-energy modes. Images were reconstructed using conventional and high definition (HD) kernels. Task-based transfer function at the 50% level (TTF_50%) for teflon, delrin, low density polyethylene (LDPE) and acrylic, as well as image noise and noise texture, were assessed across all focal spots and acquisition modes using Noise Power Spectrum (NPS) analysis. A non-prewhitening mathematical observer model was used to calculate detectability index ($d_{\mathit{NPW}}^{\prime }$).ResultsTTF_50% degraded with increasing focal spot size. TTF_50% ranged from 0.67 mm⁻¹ for teflon to 0.25 mm⁻¹ for acrylic. For standard kernel, image noise and NPS-determined average spatial frequency were 8.3 HU and 0.29 mm⁻¹, respectively in single-energy, 12.0 HU and 0.37 mm⁻¹ in HR, and 7.9 HU and 0.26 mm⁻¹ in dual-energy mode. For standard kernel, $d_{\mathit{NPW}}^{\prime }$ was 61 in single-energy and HR mode and reduced to 56 in dual-energy mode.ConclusionsThe task-based image quality assessment metrics have shown that spatial resolution is higher for higher image contrast materials and detectability is higher in the standard single-energy mode compared to HR and dual-energy mode. The results of the current study provide CT operators the required knowledge to characterize their CT system towards the optimization of its clinical performance.     

Radiomic Features Based Severity Prediction in Dementia MR Images Using Hybrid SSA-PSO Optimizer and Multi-class SVM Classifier

Objectives: In recent times, MR image is used to detect the dementia diagnostic differences in preclinical stages. Mild cognitive impairment (MCI) is characterized by slight cognitive deficits. This can be categorized into early and late mild cognitive impairment according to extent of episodic cognitive impairment. There is a higher risk of MCI subject to convert into Alzheimers disease. It is observed that there is no appropriate biomarker to find severity changes in dementia. Thus, this work aims to identify appropriate biomarker using radiomic and hybrid social algorithms.Materials: ADNI database is utilized for this study. Grey matter, cerebrospinal fluid, ventricle, hippocampus, brain stem and mid brain regions are examined to extract the radiomic features. This provides local and global tissue changes of these regions. The significant features are obtained using hybrid salp swarm and particle swarm optimization method (SSA-PSO). SVM is adopted to classify the normal and severity groups. The performance of work is validated clinically and statistically.Results: Results show that radiomic features capture anatomical changes for considered regions. The significant features from SSA-PSO show greater causal association and statistical significance for all considered regions. However, hippocampus achieves 88.5% of classification accuracy than other regions in the considered group. The inter class variations of hippocampus gives precise prognosis differences. From the clinical validation, it is also found that the obtained result show high statistical significance ($p<0.0001$) among the different severity.Conclusion: The proposed work shows promising results in using these biomarkers in detection of dementia and support clinical decisions.     

Comparison of radiomic features in diagnostic CT images with and without contrast enhancement in the delayed phase for NSCLC patients

PurposeTo compare radiomic features extracted from diagnostic computed tomography (CT) images with and without contrast enhancement in delayed phase for non-small cell lung cancer (NSCLC) patients.MethodsDiagnostic CT images from 269 tumors [non-contrast CT, 188 (dataset NE); contrast-enhanced CT, 81 (dataset CE)] were enrolled in this study. Eighteen first-order and seventy-five texture features were extracted by setting five bin width levels for CT values. Reproducible features were selected by the intraclass correlation coefficient (ICC). Radiomic features were compared between datasets NE and CE. Subgroup analyses were performed based on the CT acquisition period, exposure value, and patient characteristics.ResultsEighty features were considered reproducible (0.5 ≤ ICC). Twelve of the sixteen first-order features, independent of the bin width levels, were statistically different between datasets NE and CE (p < 0.05), and the p-values of two first-order features depending on the bin width levels were reduced with narrower bin widths. Sixteen out of sixty-two features showed a significant difference, regardless of the bin width (p < 0.05). There were significant differences between datasets NE and CE with older age, lighter body weight, better performance status, being a smoker, larger gross tumor volume, and tumor location at central region.ConclusionsContrast enhancement in the delayed phase of CT images for NSCLC patients affected some of the radiomic features and the variability of radiomic features due to contrast uptake may depend largely on the patient characteristics.     

Exploration of temporal stability and prognostic power of radiomic features based on electronic portal imaging device images

PurposeWe aimed to explore the temporal stability of radiomic features in the presence of tumor motion and the prognostic powers of temporally stable features.MethodsWe selected single fraction dynamic electronic portal imaging device (EPID) (n = 275 frames) and static digitally reconstructed radiographs (DRRs) of 11 lung cancer patients, who received stereotactic body radiation therapy (SBRT) under free breathing. Forty-seven statistical radiomic features, which consisted of 14 histogram-based features and 33 texture features derived from the graylevel co-occurrence and graylevel run-length matrices, were computed. The temporal stability was assessed by using a multiplication of the intra-class correlation coefficients (ICCs) between features derived from the EPID and DRR images at three quantization levels. The prognostic powers of the features were investigated using a different database of lung cancer patients (n = 221) based on a Kaplan-Meier survival analysis.ResultsFifteen radiomic features were found to be temporally stable for various quantization levels. Among these features, seven features have shown potentials for prognostic prediction in lung cancer patients.ConclusionsThis study suggests a novel approach to select temporally stable radiomic features, which could hold prognostic powers in lung cancer patients.     

Combined Metaheuristic Algorithm and Radiomics Strategy for the Analysis of Neuroanatomical Structures in Schizophrenia and Schizoaffective Disorders

ObjectivesSchizophrenia (SZ) is the most chronic disabling psychotic brain disorder. It is characterized by delusions and auditory hallucinations, as well as impairments in memory. Schizoaffective (SA) signs are co-morbid with SZ and are characterized by symptoms of SZ and mood disorder. Various researches suggest that SZ and SA share a number of equally severe cognitive deficits, but the pathophysiology has not yet been addressed in a comprehensive way. In this work, the heterogeneity in whole brain, ventricle and cerebellum region from psychotic MR brain images is examined using Machine learning and radiomic features.Materials and methodsT1 weighted MR brain images are obtained from Schizconnect database for the analysis. The shape prior level set method is used to segment the ventricle and cerebellum structures. The radiomic features which include shape and texture are extracted from these regions to discriminate the SZ and SA subjects. The performance of these features is evaluated with Binary Particle Swarm Optimization (BPSO) based Fuzzy Support Vector Machine (FSVM) classifier.ResultsThe shape constrained Level Set method is able to better segment ventricles and cerebellum regions from the images. The significant features that are extracted from whole brain, ventricle and cerebellum are identified by the BPSO based FSVM. The combination of radiomic features extracted from cerebellum region achieved high classification accuracy (90.09%) using metaheuristic algorithm. The extracted features from cerebellum are correlated with PANSS score. The causal analysis shows that there is an association been the tissue texture variation in identifying the disease severity. The symmetry analysis shows that left brain mean area is larger than the right side area. In particular SA has low cerebellum area compared to SZ. The radiomic features such as Hermite, Laws and tensor extracted from the left cerebellum show a significant texture variation in all the considered subjects (p<0.0001).ConclusionsThe results are clinically relevant in discriminating the pattern change in the structure, hence this biomarker and frame work could be used for the severity study of psychotic disorders.     

Influence of inter-observer delineation variability on radiomic features of the parotid gland

PurposeThis study aimed to quantify the variability in the values of radiomic features extracted from a right parotid gland (RPG) delineated by a series of independent observers.MethodsThis was a secondary analysis of anonymous data from a delineation workshop. Inter-observer variability of the RPG from 40 participants was quantified using DICE similarity coefficient (DSC) and Hausdorff distance (HD). An additional contour was generated using Varian SmartSegmentation. Radiomic features extracted include four shape features, six histogram features, and 32 texture features. The absolute mean paired percentage difference (PPD) in feature values from the expert and participants were ranked . Feature robustness was classified using pre- determined thresholds.Results63% of participants achieved a DSC > 0.7, the auto- segmentation DSC was 0.76. The average HD for the participants was 16.16 mm ± 0.66 mm, and 15.16 mm for the auto-segmentation. 48% (n = 20) and 33% (n = 14) of features were deemed to be robust with a mean absolute PPD < 5%, for the auto-segmentation and manual delineations respectively; the majority of which were from the grey-run length matrix family. 7% (n = 3) of features from the auto- segmentation and 10% (n = 4) from the manual contours were deemed to be unstable with a mean absolute PPD > 50%. The value of the most robust feature was not related to DSC and HD.ConclusionInter-observer delineation variability affects the value of the radiomic features extracted from the RPG. This study identifies the radiomic features least sensitive to these uncertainties. Further investigation of the clinical relevance of these features in prediction of xerostomia is warranted.     

Multiparametric MRI-based radiomics analysis for the prediction of breast tumor regression patterns after neoadjuvant chemotherapy

ObjectivesBreast cancers show different regression patterns after neoadjuvant chemotherapy. Certain regression patterns are associated with more reliable margins in breast-conserving surgery. Our study aims to establish a nomogram based on radiomic features and clinicopathological factors to predict regression patterns in breast cancer patients.MethodsWe retrospectively reviewed 144 breast cancer patients who received neoadjuvant chemotherapy and underwent definitive surgery in our center from January 2016 to December 2019. Tumor regression patterns were categorized as type 1 (concentric regression + pCR) and type 2 (multifocal residues + SD + PD) based on pathological results. We extracted 1158 multidimensional features from 2 sequences of MRI images. After feature selection, machine learning was applied to construct a radiomic signature. Clinical characteristics were selected by backward stepwise selection. The combined prediction model was built based on both the radiomic signature and clinical factors. The predictive performance of the combined prediction model was evaluated.ResultsTwo radiomic features were selected for constructing the radiomic signature. Combined with two significant clinical characteristics, the combined prediction model showed excellent prediction performance, with an area under the receiver operating characteristic curve of 0.902 (95% confidence interval 0.8343–0.9701) in the primary cohort and 0.826 (95% confidence interval 0.6774–0.9753) in the validation cohort.ConclusionsOur study established a unique model combining a radiomic signature and clinicopathological factors to predict tumor regression patterns prior to the initiation of NAC. The early prediction of type 2 regression offers the opportunity to modify preoperative treatments or aids in determining surgical options.     

Deep radiomic model based on the sphere–shell partition for predicting treatment response to chemotherapy in lung cancer

BackgroundThe prognosis of chemotherapy is important in clinical decision-making for non-small cell lung cancer (NSCLC) patients.ObjectivesTo develop a model for predicting treatment response to chemotherapy in NSCLC patients from pre-chemotherapy CT images.Materials and MethodsThis retrospective multicenter study enrolled 485 patients with NSCLC who received chemotherapy alone as a first-line treatment. Two integrated models were developed using radiomic and deep-learning-based features. First, we partitioned pre-chemotherapy CT images into spheres and shells with different radii around the tumor (0–3, 3–6, 6–9, 9–12, 12–15 mm) containing intratumoral and peritumoral regions. Second, we extracted radiomic and deep-learning-based features from each partition. Third, using radiomic features, five sphere–shell models, one feature fusion model, and one image fusion model were developed. Finally, the model with the best performance was validated in two cohorts.ResultsAmong the five partitions, the model of 9–12 mm achieved the highest area under the curve (AUC) of 0.87 (95% confidence interval: 0.77–0.94). The AUC was 0.94 (0.85–0.98) for the feature fusion model and 0.91 (0.82–0.97) for the image fusion model. For the model integrating radiomic and deep-learning-based features, the AUC was 0.96 (0.88–0.99) for the feature fusion method and 0.94 (0.85–0.98) for the image fusion method. The best-performing model had an AUC of 0.91 (0.81–0.97) and 0.89 (0.79–0.93) in two validation sets, respectively.ConclusionsThis integrated model can predict the response to chemotherapy in NSCLC patients and assist physicians in clinical decision-making.     

Computed tomography-based radiomic model at node level for the prediction of normal-sized lymph node metastasis in cervical cancer

PurposeRadiomic models have been demonstrated to have acceptable discrimination capability for detecting lymph node metastasis (LNM). We aimed to develop a computed tomography–based radiomic model and validate its usefulness in the prediction of normal-sized LNM at node level in cervical cancer.MethodsA total of 273 LNs of 219 patients from 10 centers were evaluated in this study. We randomly divided the LNs from the 2 centers with the largest number of LNs into the training and internal validation cohorts, and the rest as the external validation cohort. Radiomic features were extracted from the arterial and venous phase images. We trained an artificial neural network (ANN) to develop two single-phase models. A radiomic model reflecting the features of two-phase images was also built for directly predicting LNM in cervical cancer. Moreover, four state-of-the-art methods were used for comparison. The performance of all models was assessed using the area under the receiver operating characteristic curve (AUC).ResultsAmong the models we built, the models combining the features of two phases surpassed the single-phase models, and the models generated by ANN had better performance than the others. We found that the radiomic model achieved the highest AUCs of 0.912 and 0.859 in the training and internal validation cohorts, respectively. In the external validation cohort, the AUC of the radiomic model was 0.800.ConclusionWe constructed a radiomic model that exhibited great ability in the prediction of LNM. The application of the model could optimize clinical staging and decision-making.     

Evaluation of radiomic texture feature error due to MRI acquisition and reconstruction: A simulation study utilizing ground truth

The purpose of this study was to examine the dependence of image texture features on MR acquisition parameters and reconstruction using a digital MR imaging phantom. MR signal was simulated in a parallel imaging radiofrequency coil setting as well as a single element volume coil setting, with varying levels of acquisition noise, three acceleration factors, and four image reconstruction algorithms. Twenty-six texture features were measured on the simulated images, ground truth images, and clinical brain images. Subtle algorithm-dependent errors were observed on reconstructed phantom images, even in the absence of added noise. Sources of image error include Gibbs ringing at image edge gradients (tissue interfaces) and well-known artifacts due to high acceleration; two of the iterative reconstruction algorithms studied were able to mitigate these image errors. The difference of the texture features from ground truth, and their variance over reconstruction algorithm and parallel imaging acceleration factor, were compared to the clinical “effect size”, i.e., the feature difference between high- and low-grade tumors on T1- and T2-weighted brain MR images of twenty glioma patients. The measured feature error (difference from ground truth) was small for some features, but substantial for others. The feature variance due to reconstruction algorithm and acceleration factor were generally smaller than the clinical effect size. Certain texture features may be preserved by MR imaging, but adequate precautions need to be taken regarding their validity and reliability. We present a general simulation framework for assessing the robustness and accuracy of radiomic textural features under various MR acquisition/reconstruction scenarios.     

Diagnosis of Distant Metastasis of Lung Cancer: Based on Clinical and Radiomic Features

OBJECTIVES: To analyze the distant metastasis possibility based on computed tomography (CT) radiomic features in patients with lung cancer. METHODS: This was a retrospective analysis of 348 patients with lung cancer enrolled between 2014 and February 2015. A feature set containing clinical features and 485 radiomic features was extracted from the pretherapy CT images. Feature selection via concave minimization (FSV) was used to select effective features. A support vector machine (SVM) was used to evaluate the predictive ability of each feature. RESULTS: Four radiomic features and three clinical features were obtained by FSV feature selection. Classification accuracy by the proposed SVM with SGD method was 71.02%, and the area under the curve was 72.84% with only the radiomic features extracted from CT. After the addition of clinical features, 89.09% can be achieved. CONCLUSION: The radiomic features of the pretherapy CT images may be used as predictors of distant metastasis. And it also can be used in combination with the patient's gender and tumor T and N phase information to diagnose the possibility of distant metastasis in lung cancer.     

Exploring the variability of radiomic features of lung cancer lesions on unenhanced and contrast-enhanced chest CT imaging

PurposeThe aim of this methods work is to explore the different behavior of radiomic features resulting by using or not the contrast medium in chest CT imaging of non-small cell lung cancer.MethodsChest CT scans, unenhanced and contrast-enhanced, of 17 patients were selected from images collected as part of the staging process. The major T1-T3 lesion was contoured through a semi-automatic approach. These lesions formed the lesion phantoms to study features behavior. The stability of 94 features of the 3D-Slicer package Radiomics was analyzed. Feature discrimination power was quantified by means of Gini's coefficient. Correlation between distance matrices was evaluated through Mantel statistic. Heatmap, cluster and silhouette plots were applied to find well-structured partitions of lesions.ResultsThe Gini's coefficient evidenced a low discrimination power, <0.05, for four features and a large discrimination power, around 0.8, for five features. About 90% of features was affected by the contrast medium, masking tumor lesions variability; thirteen features only were found stable. On 8178 combinations of stable features, only one group of four features produced the same partition of lesions with the silhouette width greater than 0.51, both on unenhanced and contrast-enhanced images.ConclusionsGini’s coefficient highlighted the features discrimination power in both CT series. Many features were sensitive to the use of the contrast medium, masking the lesions intrinsic variability. Four stable features produced, on both series, the same partition of cancer lesions with reasonable structure; this may merit being objects of further validation studies and interpretative investigations.     

Homological radiomics analysis for prognostic prediction in lung cancer patients

PurposeThis study explored a novel homological analysis method for prognostic prediction in lung cancer patients.Materials and methodsThe potential of homology-based radiomic features (HFs) was investigated by comparing HFs to conventional wavelet-based radiomic features (WFs) and combined radiomic features consisting of HFs and WFs (HWFs), using training (n = 135) and validation (n = 70) datasets, and Kaplan–Meier analysis. A total of 13,824 HFs were derived through homology-based texture analysis using Betti numbers, which represent the topologically invariant morphological characteristics of lung cancer. The prognostic potential of HFs was evaluated using statistically significant differences (p-values, log-rank test) to compare the survival curves of high- and low-risk patients. Those patients were stratified into high- and low-risk groups using the medians of the radiomic scores of signatures constructed with an elastic-net-regularized Cox proportional hazard model. Furthermore, deep learning (DL) based on AlexNet was utilized to compare HFs by stratifying patients into the two groups using a network that was pre-trained with over one million natural images from an ImageNet database.ResultsFor the training dataset, the p-values between the two survival curves were 6.7 × 10⁻⁶ (HF), 5.9 × 10⁻³ (WF), 7.4 × 10⁻⁶ (HWF), and 1.1 × 10⁻³ (DL). The p-values for the validation dataset were 3.4 × 10⁻⁵ (HF), 6.7 × 10⁻¹ (WF), 1.7 × 10⁻⁷ (HWF), and 1.2 × 10⁻¹ (DL).ConclusionThis study demonstrates the excellent potential of HFs for prognostic prediction in lung cancer patients.     

Early risk-assessment of patients with nasopharyngeal carcinoma: the added prognostic value of MR-based radiomics

ObjectivesTo assess the additive prognostic value of MR-based radiomics in predicting progression-free survival (PFS) in patients with nasopharyngeal carcinoma (NPC)MethodsPatients newly diagnosed with non-metastatic NPC between June 2006 and October 2019 were retrospectively included and randomly grouped into training and test cohorts (7:3 ratio). Radiomic features (n=213) were extracted from T2-weighted and contrast-enhanced T1-weighted MRI. The patients were staged according to the 8^th edition of American Joint Committee on Cancer Staging Manual. The least absolute shrinkage and selection operator was used to select the relevant radiomic features. Univariate and multivariate Cox proportional hazards analyses were conducted for PFS, yielding three different survival models (clinical, stage, and radiomic). The integrated time-dependent area under the curve (iAUC) for PFS was calculated and compared among different combinations of survival models, and the analysis of variance was used to compare the survival models. The prognostic performance of all models was validated using a test set with integrated Brier scores.ResultsThis study included 81 patients (training cohort=57; test cohort=24), and the mean PFS was 57.5 ± 43.6 months. In the training cohort, the prognostic performances of survival models improved significantly with the addition of radiomics to the clinical (iAUC, 0.72–0.80; p=0.04), stage (iAUC, 0.70–0.79; p=0.001), and combined models (iAUC, 0.76–0.81; p<0.001). In the test cohort, the radiomics and combined survival models were robustly validated for their ability to predict PFS.ConclusionIntegration of MR-based radiomic features with clinical and stage variables improved the prediction PFS in patients diagnosed with NPC.