Test-Retest Repeatability of Patlak Slopes versus Standardized Uptake Values for Hypermetabolic Lesions and Normal Organs in an Oncologic PET/CT Population

PURPOSE

We aimed to determine the test-retest repeatability of quantitative metrics based on the Patlak slope (PS) versus the standardized uptake value (SUV) among lesions and normal organs on oncologic [18F]FDG-PET/CT.

PROCEDURES

This prospective, single-center study enrolled adults undergoing standard-of-care oncologic [18F]FDG-PET/CTs. Early (35-50 min post-injection) and late (75-90 min post-injection) SUV and PS images were reconstructed from dynamic whole-body PET data. Repeat imaging occurred within 7 days. Relevant quantitative metrics were extracted from lesions and normal organs. Repeatability was assessed via mean test-retest percent changes [T-RT %Δ], within-subject coefficients of variation (wCVs), and intra-class correlation coefficients (ICCs).

RESULTS

Nine subjects (mean age, 61.7 ± 6.2 years; 6 females) completed the test-retest protocol. Four subjects collectively had 17 [18F]FDG-avid lesions. Lesion wCVs were higher (i.e., worse repeatability) for PS-early-max (16.2%) and PS-early-peak (15.6%) than for SUV-early-max (8.9%) and SUV-early-peak (8.1%), with similar early metric ICCs (0.95-0.98). Lesion wCVs were similar for PS-late-max (8.5%) and PS-late-peak (6.4%) relative to SUV-late-max (9.7%) and SUV-late-peak (7.2%), with similar late metric ICCs (0.93-0.98). There was a significant bias toward higher retest SUV and PS values in the lesion analysis (T-RT %Δ [95% CI]: SUV-late-max, 10.0% [2.6%, 17.0%]; PS-late-max, 20.4% [14.3%, 26.4%]) but not in the normal organ analysis.

CONCLUSIONS

Among [18F]FDG-avid lesions, the repeatability of PS-based metrics is similar to equivalent SUV-based metrics at late post-injection time points, indicating that PS-based metrics may be suitable for tracking response to oncologic therapies. However, further validation is required in light of our study's limitations, including small sample size and bias toward higher retest values for some metrics.

Standardized Radiomics Analysis of Clinical Myocardial Perfusion Stress SPECT Images to Identify Coronary Artery Calcification

PURPOSE

Myocardial perfusion (MP) stress single-photon emission computed tomography (SPECT) is an established diagnostic test for patients suspected of coronary artery disease (CAD). Meanwhile, coronary artery calcification (CAC) scoring obtained from diagnostic CT is a highly sensitive test, offering incremental diagnostic information in identifying patients with significant CAD yet normal MP stress SPECT (MPSS) scans. However, after decades of wide utilization of MPSS, CAC is not commonly reimbursed (e.g. by the CMS), nor widely deployed in community settings. We studied the potential of complementary information deduced from the radiomics analysis of normal MPSS scans in predicting the CAC score.

METHODS

We collected data from 428 patients with normal (non-ischemic) MPSS (99mTc-sestamibi; consensus reading). A nuclear medicine physician verified iteratively reconstructed images (attenuation-corrected) to be free from fixed perfusion defects and artifactual attenuation. Three-dimensional images were automatically segmented into four regions of interest (ROIs), including myocardium and three vascular segments (left anterior descending [LAD]-left circumference [LCX]-right coronary artery [RCA]). We used our software package, standardized environment for radiomics analysis (SERA), to extract 487 radiomic features in compliance with the image biomarker standardization initiative (IBSI). Isotropic cubic voxels were discretized using fixed bin-number discretization (eight schemes). We first performed blind-to-outcome feature selection focusing on a priori usefulness, dynamic range, and redundancy of features. Subsequently, we performed univariate and multivariate machine learning analyses to predict CAC scores from i) selected radiomic features, ii) 10 clinical features, and iii) combined radiomics + clinical features. Univariate analysis invoked Spearman correlation with Benjamini-Hotchberg false-discovery correction. The multivariate analysis incorporated stepwise linear regression, where we randomly selected a 15% test set and divided the other 85% of data into 70% training and 30% validation sets. Training started from a constant (intercept) model, iteratively adding/removing features (stepwise regression), invoking the Akaike information criterion (AIC) to discourage overfitting. Validation was run similarly, except that the training output model was used as the initial model. We randomized training/validation sets 20 times, selecting the best model using log-likelihood for evaluation in the test set. Assessment in the test set was performed thoroughly by running the entire operation 50 times, subsequently employing Fisher's method to verify the significance of independent tests.

RESULTS

Unsupervised feature selection significantly reduced 8×487 features to 56. In univariate analysis, no feature survived the false-discovery rate (FDR) to directly correlate with CAC scores. Applying Fisher's method to the multivariate regression results demonstrated combining radiomics with the clinical features to enhance the significance of the prediction model across all cardiac segments.  Conclusions: Our standardized and statistically robust multivariate analysis demonstrated significant prediction of the CAC score for all cardiac segments when combining MPSS radiomic features with clinical features, suggesting radiomics analysis can add diagnostic or prognostic value to standard MPSS for wide clinical usage.

Deep learning and radiomics framework for PSMA-RADS classification of prostate cancer on PSMA PET

BACKGROUND

Accurate classification of sites of interest on prostate-specific membrane antigen (PSMA) positron emission tomography (PET) images is an important diagnostic requirement for the differentiation of prostate cancer (PCa) from foci of physiologic uptake. We developed a deep learning and radiomics framework to perform lesion-level and patient-level classification on PSMA PET images of patients with PCa.

METHODS

This was an IRB-approved, HIPAA-compliant, retrospective study. Lesions on [¹⁸F]DCFPyL PET/CT scans were assigned to PSMA reporting and data system (PSMA-RADS) categories and randomly partitioned into training, validation, and test sets. The framework extracted image features, radiomic features, and tissue type information from a cropped PET image slice containing a lesion and performed PSMA-RADS and PCa classification. Performance was evaluated by assessing the area under the receiver operating characteristic curve (AUROC). A t-distributed stochastic neighbor embedding (t-SNE) analysis was performed. Confidence and probability scores were measured. Statistical significance was determined using a two-tailed t test.

RESULTS

PSMA PET scans from 267 men with PCa had 3794 lesions assigned to PSMA-RADS categories. The framework yielded AUROC values of 0.87 and 0.90 for lesion-level and patient-level PSMA-RADS classification, respectively, on the test set. The framework yielded AUROC values of 0.92 and 0.85 for lesion-level and patient-level PCa classification, respectively, on the test set. A t-SNE analysis revealed learned relationships between the PSMA-RADS categories and disease findings. Mean confidence scores reflected the expected accuracy and were significantly higher for correct predictions than for incorrect predictions (P < 0.05). Measured probability scores reflected the likelihood of PCa consistent with the PSMA-RADS framework.

CONCLUSION

The framework provided lesion-level and patient-level PSMA-RADS and PCa classification on PSMA PET images. The framework was interpretable and provided confidence and probability scores that may assist physicians in making more informed clinical decisions.

Voxel-based partial volume correction of PET images via subtle MRI guided non-local means regularization

PURPOSE

Positron emission tomography (PET) images tend to be significantly degraded by the partial volume effect (PVE) resulting from the limited spatial resolution of the reconstructed images. Our purpose is to propose a partial volume correction (PVC) method to tackle this issue.

METHODS

In the present work, we explore a voxel-based PVC method under the least squares framework (LS) employing anatomical non-local means (NLMA) regularization. The well-known non-local means (NLM) filter utilizes the high degree of information redundancy that typically exists in images, and is typically used to directly reduce image noise by replacing each voxel intensity with a weighted average of its non-local neighbors. Here we explore NLM as a regularization term within iterative-deconvolution model to perform PVC. Further, an anatomical-guided version of NLM was proposed that incorporates MRI information into NLM to improve resolution and suppress image noise. The proposed approach makes subtle usage of the accompanying MRI information to define a more appropriate search space within the prior model. To optimize the regularized LS objective function, we used the Gauss-Seidel (GS) algorithm with the one-step-late (OSL) technique.

RESULTS

After the import of NLMA, the visual and quality results are all improved. With a visual check, we notice that NLMA reduce the noise compared to other PVC methods. This is also validated in bias-noise curve compared to non-MRI-guided PVC framework. We can see NLMA gives better bias-noise trade-off compared to other PVC methods.

CONCLUSIONS

Our efforts were evaluated in the base of amyloid brain PET imaging using the BrainWeb phantom and in vivo human data. We also compared our method with other PVC methods. Overall, we demonstrated the value of introducing subtle MRI-guidance in the regularization process, the proposed NLMA method resulting in promising visual as well as quantitative performance improvements.

Next-Generation Radiogenomics Sequencing for Prediction of EGFR and KRAS Mutation Status in NSCLC Patients Using Multimodal Imaging and Machine Learning Algorithms

PURPOSE

Considerable progress has been made in the assessment and management of non-small cell lung cancer (NSCLC) patients based on mutation status in the epidermal growth factor receptor (EGFR) and Kirsten rat sarcoma viral oncogene (KRAS). At the same time, NSCLC management through KRAS and EGFR mutation profiling faces challenges. In the present work, we aimed to evaluate a comprehensive radiomics framework that enabled prediction of EGFR and KRAS mutation status in NSCLC patients based on radiomic features from low-dose computed tomography (CT), contrast-enhanced diagnostic quality CT (CTD), and positron emission tomography (PET) imaging modalities and use of machine learning algorithms.

METHODS

Our study involved NSCLC patients including 150 PET, low-dose CT, and CTD images. Radiomic features from original and preprocessed (including 64 bin discretizing, Laplacian-of-Gaussian (LOG), and Wavelet) images were extracted. Conventional clinically used standard uptake value (SUV) parameters and metabolic tumor volume (MTV) were also obtained from PET images. Highly correlated features were pre-eliminated, and false discovery rate (FDR) correction was performed with the resulting q-values reported for univariate analysis. Six feature selection methods and 12 classifiers were then used for multivariate prediction of gene mutation status (provided by polymerase chain reaction (PCR)) in patients. We performed 10-fold cross-validation for model tuning to improve robustness, and our developed models were assessed on an independent validation set with 68 patients (common in all three imaging modalities). The average area under the receiver operator characteristic curve (AUC) was utilized for performance evaluation.

RESULTS

The best predictive power for conventional PET parameters was achieved by SUV_peak (AUC 0.69, p value = 0.0002) and MTV (AUC 0.55, p value = 0.0011) for EGFR and KRAS, respectively. Univariate analysis of extracted radiomics features improved AUC performance to 0.75 (q-value 0.003, Short-Run Emphasis feature of GLRLM from LOG preprocessed image of PET with sigma value 1.5) and 0.71 (q-value 0.00005, Large Dependence Low Gray-Level Emphasis feature of GLDM in LOG preprocessed image of CTD with sigma value 5) for EGFR and KRAS, respectively. Furthermore, multivariate machine learning-based AUC performances were significantly improved to 0.82 for EGFR (LOG preprocessed image of PET with sigma 3 with variance threshold (VT) feature selector and stochastic gradient descent (SGD) classifier (q-value = 4.86E-05) and 0.83 for KRAS (LOG preprocessed image of CT with sigma 3.5 with select model (SM) feature selector and SGD classifier (q-value = 2.81E-09).

CONCLUSION

Our work demonstrated that non-invasive and reliable radiomics analysis can be successfully used to predict EGFR and KRAS mutation status in NSCLC patients. We demonstrated that radiomic features extracted from different image-feature sets could be used for EGFR and KRAS mutation status prediction in NSCLC patients and showed improved predictive power relative to conventional image-derived metrics.

A physics-guided modular deep-learning based automated framework for tumor segmentation in PET

An important need exists for reliable positron emission tomography (PET) tumor-segmentation methods for tasks such as PET-based radiation-therapy planning and reliable quantification of volumetric and radiomic features. To address this need, we propose an automated physics-guided deep-learning-based three-module framework to segment PET images on a per-slice basis. The framework is designed to help address the challenges of limited spatial resolution and lack of clinical training data with known ground-truth tumor boundaries in PET. The first module generates PET images containing highly realistic tumors with known ground-truth using a new stochastic and physics-based approach, addressing lack of training data. The second module trains a modified U-net using these images, helping it learn the tumor-segmentation task. The third module fine-tunes this network using a small-sized clinical dataset with radiologist-defined delineations as surrogate ground-truth, helping the framework learn features potentially missed in simulated tumors. The framework was evaluated in the context of segmenting primary tumors in 18F-fluorodeoxyglucose (FDG)-PET images of patients with lung cancer. The framework's accuracy, generalizability to different scanners, sensitivity to partial volume effects (PVEs) and efficacy in reducing the number of training images were quantitatively evaluated using Dice similarity coefficient (DSC) and several other metrics. The framework yielded reliable performance in both simulated (DSC: 0.87 (95% confidence interval (CI): 0.86, 0.88)) and patient images (DSC: 0.73 (95% CI: 0.71, 0.76)), outperformed several widely used semi-automated approaches, accurately segmented relatively small tumors (smallest segmented cross-section was 1.83 cm2), generalized across five PET scanners (DSC: 0.74 (95% CI: 0.71, 0.76)), was relatively unaffected by PVEs, and required low training data (training with data from even 30 patients yielded DSC of 0.70 (95% CI: 0.68, 0.71)). In conclusion, the proposed automated physics-guided deep-learning-based PET-segmentation framework yielded reliable performance in delineating tumors in FDG-PET images of patients with lung cancer.

The Image Biomarker Standardization Initiative: Standardized Quantitative Radiomics for High-Throughput Image-based Phenotyping

Background Radiomic features may quantify characteristics present in medical imaging. However, the lack of standardized definitions and validated reference values have hampered clinical use. Purpose To standardize a set of 174 radiomic features. Materials and Methods Radiomic features were assessed in three phases. In phase I, 487 features were derived from the basic set of 174 features. Twenty-five research teams with unique radiomics software implementations computed feature values directly from a digital phantom, without any additional image processing. In phase II, 15 teams computed values for 1347 derived features using a CT image of a patient with lung cancer and predefined image processing configurations. In both phases, consensus among the teams on the validity of tentative reference values was measured through the frequency of the modal value and classified as follows: less than three matches, weak; three to five matches, moderate; six to nine matches, strong; 10 or more matches, very strong. In the final phase (phase III), a public data set of multimodality images (CT, fluorine 18 fluorodeoxyglucose PET, and T1-weighted MRI) from 51 patients with soft-tissue sarcoma was used to prospectively assess reproducibility of standardized features. Results Consensus on reference values was initially weak for 232 of 302 features (76.8%) at phase I and 703 of 1075 features (65.4%) at phase II. At the final iteration, weak consensus remained for only two of 487 features (0.4%) at phase I and 19 of 1347 features (1.4%) at phase II. Strong or better consensus was achieved for 463 of 487 features (95.1%) at phase I and 1220 of 1347 features (90.6%) at phase II. Overall, 169 of 174 features were standardized in the first two phases. In the final validation phase (phase III), most of the 169 standardized features could be excellently reproduced (166 with CT; 164 with PET; and 164 with MRI). Conclusion A set of 169 radiomics features was standardized, which enabled verification and calibration of different radiomics software. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Kuhl and Truhn in this issue.

Multi-Level Multi-Modality Fusion Radiomics: Application to PET and CT Imaging for Prognostication of Head and Neck Cancer

To characterize intra-tumor heterogeneity comprehensively, we propose a multi-level fusion strategy to combine PET and CT information at the image-, matrix-and feature-levels towards improved prognosis. Specifically, we developed fusion radiomics in the context of 3 prognostic outcomes in a multi-center setting (4 centers) involving 296 head & neck cancer patients. Eight clinical parameters were first utilized to build a (1) clinical model. We also built models by extracting 127 radiomics features from (2) PET images alone; (3-8) PET and CT images fused via wavelet-based fusion (WF) using CT-weights of 0.2, 0.4, 0.6 and 0.8, gradient transfer fusion (GTF), and guided filtering-based fusion (GFF); (9) fused matrices (sumMat); (10-11) fused features constructed via feature averaging (avgFea) and feature concatenation (conFea); and finally, (12) CT images alone; above models were also expanded to include both clinical and radiomics features. Seven variations of training and testing partitions were investigated. Highest performance in 5, 6 and 5 partitions was achieved by image-level fusion strategies for RFS, MFS and OS prediction, respectively. Among all partitions, WF0.6 and WF0.8 showed significantly higher performance than CT model for RFS (C-index: 0.60 ± 0.04 vs. 0.56 ± 0.03, p-value: 0.015) and MFS (C-index: 0.71 ± 0.13 vs. 0.62 ± 0.08, p-value: 0.020) predictions, respectively. In partition CER 23 vs. 14, WF0.6 significantly outperformed Clinical model for RFS prediction (C-index: 0.67 vs. 0.53, p-value: 0.003); both avgFea and WF0.6 showed C-index of 0.64 and significantly higher than that of PET only (C-index: 0.51, p-value: 0.018 and 0.031, respectively) for OS prediction. Fusion radiomics modeling showed varying improvements compared to single modality models for different outcome predictions in different partitions, highlighting the importance of generalizing radiomics models. Image-level fusion holds potential to capture more useful characteristics.

Prognostic modeling for patients with colorectal liver metastases incorporating FDG PET radiomic features

OBJECTIVE

We aimed to improve prediction of outcome for patients with colorectal liver metastases, via prognostic models incorporating PET-derived measures, including radiomic features that move beyond conventional standard uptake value (SUV) measures.

PATIENTS AND METHODS

A range of parameters including volumetric and heterogeneity measures were derived from FDG PET images of 52 patients with colorectal intrahepatic-only metastases (29 males and 23 females; mean age 62.9 years [SD 9.8; range 32-82]). The patients underwent PET/CT imaging as part of the clinical workup prior to final decision on treatment. Univariate and multivariate models were implemented, which included statistical considerations (to discourage false discovery and overfitting), to predict overall survival (OS), progression-free survival (PFS) and event-free survival (EFS). Kaplan-Meier survival analyses were performed, where the subjects were divided into high-risk and low-risk groups, from which the hazard ratios (HR) were computed via Cox proportional hazards regression.

RESULTS

Commonly-invoked SUV metrics performed relatively poorly for different prediction tasks (SUVmax HR = 1.48, 0.83 and 1.16; SUVpeak HR = 2.05, 1.93, and 1.64, for OS, PFS and EFS, respectively). By contrast, the number of liver metastases and metabolic tumor volume (MTV) each performed well (with respective HR values of 2.71, 2.61 and 2.42, and 2.62, 1.96 and 2.29, for OS, PFS and EFS). Total lesion glycolysis (TLG) also resulted in similar performance as MTV. Multivariate prognostic modeling incorporating different features (including those quantifying intra-tumor heterogeneity) resulted in further enhanced prediction. Specifically, HR values of 4.29, 4.02 and 3.20 (p-values = 0.00004, 0.0019 and 0.0002) were obtained for OS, PFS and EFS, respectively.

CONCLUSIONS

PET-derived measures beyond commonly invoked SUV parameters hold significant potential towards improved prediction of clinical outcome in patients with liver metastases, especially when utilizing multivariate models.

Practical no-gold-standard evaluation framework for quantitative imaging methods: application to lesion segmentation in positron emission tomography

Recently, a class of no-gold-standard (NGS) techniques have been proposed to evaluate quantitative imaging methods using patient data. These techniques provide figures of merit (FoMs) quantifying the precision of the estimated quantitative value without requiring repeated measurements and without requiring a gold standard. However, applying these techniques to patient data presents several practical difficulties including assessing the underlying assumptions, accounting for patient-sampling-related uncertainty, and assessing the reliability of the estimated FoMs. To address these issues, we propose statistical tests that provide confidence in the underlying assumptions and in the reliability of the estimated FoMs. Furthermore, the NGS technique is integrated within a bootstrap-based methodology to account for patient-sampling-related uncertainty. The developed NGS framework was applied to evaluate four methods for segmenting lesions from F-Fluoro-2-deoxyglucose positron emission tomography images of patients with head-and-neck cancer on the task of precisely measuring the metabolic tumor volume. The NGS technique consistently predicted the same segmentation method as the most precise method. The proposed framework provided confidence in these results, even when gold-standard data were not available. The bootstrap-based methodology indicated improved performance of the NGS technique with larger numbers of patient studies, as was expected, and yielded consistent results as long as data from more than 80 lesions were available for the analysis.

A novel metric for quantification of homogeneous and heterogeneous tumors in PET for enhanced clinical outcome prediction

Oncologic PET images provide valuable information that can enable enhanced prognosis of disease. Nonetheless, such information is simplified significantly in routine clinical assessment to meet workflow constraints. Examples of typical FDG PET metrics include: (i) SUVmax, (2) total lesion glycolysis (TLG), and (3) metabolic tumor volume (MTV). We have derived and implemented a novel metric for tumor quantification, inspired in essence by a model of generalized equivalent uniform dose as used in radiation therapy. The proposed metric, denoted generalized effective total uptake (gETU), is attractive as it encompasses the abovementioned commonly invoked metrics, and generalizes them, for both homogeneous and heterogeneous tumors, using a single parameter a. We evaluated this new metric for improved overall survival (OS) prediction on two different baseline FDG PET/CT datasets: (a) 113 patients with squamous cell cancer of the oropharynx, and (b) 72 patients with locally advanced pancreatic adenocarcinoma. Kaplan-Meier survival analysis was performed, where the subjects were subdivided into two groups using the median threshold, from which the hazard ratios (HR) were computed in Cox proportional hazards regression. For the oropharyngeal cancer dataset, MTV, TLG, SUVmax, SUVmean and SUVpeak produced HR values of 1.86, 3.02, 1.34, 1.36 and 1.62, while the proposed gETU metric for a  = 0.25 (greater emphasis on volume information) enabled significantly enhanced OS prediction with HR  =  3.94. For the pancreatic cancer dataset, MTV, TLG, SUVmax, SUVmean and SUVpeak resulted in HR values of 1.05, 1.25, 1.42, 1.45 and 1.52, while gETU at a  = 3.2 (greater emphasis on SUV information) arrived at an improved HR value of 1.61. Overall, the proposed methodology allows placement of differing degrees of emphasis on tumor volume versus uptake for different types of tumors to enable enhanced clinical outcome prediction.