PET Radiomic Features Research Articles

Feasibility of Using 18F-FDG PET/CT Radiomics and Machine Learning to Detect Drug-Induced Interstitial Lung Disease

Background: Bleomycin is an oncolytic and antibiotic agent used to treat various human cancers because of its antitumor activity. Unfortunately, up to 46% of the patients treated with bleomycin develop drug-induced interstitial lung disease (DIILD) and potentially life-threatening interstitial pulmonary fibrosis. Tools and biomarkers for predicting and detecting DIILD are limited. Therefore, we aimed to evaluate the feasibility of 18F-FDG PET/CT, PET radiomics, and machine learning in distinguishing DIILD in an explorative pilot study. Methods: Eighteen Hodgkin’s lymphoma (HL) patients, of whom 10 developed DIILD after treatment with bleomycin, were retrospectively included. Five diffuse large B-cell lymphoma (DLBCL) patients were included as a control group since they were not treated with bleomycin. All patients underwent 18F-FDG PET/CT scans before (baseline) and during treatment (interim). Structural changes were assessed by changes in Hounsfield Units (HUs). The 18F-FDG PET scans were used to assess metabolic changes by examining the feasibility of 504 radiomics features, including the mean activity of the lungs (SUVmean). A Random Forest (RF) classifier evaluated the identification and prediction of DIILD based on PET radiomics features. Results: HL patients who developed DIILD showed a significant increase in standard SUV metrics (SUVmean; p = 0.012, median increase 37.4%), and in some regional PET radiomics features (texture strength; p = 0.009, median increase 101.6% and zone distance entropy; p = 0.019, median increase 18.5%), while this was not found in HL patients who did not develop DIILD and DLBCL patients. The RF classifier correctly identified DIILD in 72.2% of the patients and predicted the development of DIILD correctly in 50% of the patients. There were no significant differences in HUs over time within all three patient groups. Conclusions: Our explorative longitudinal pilot study suggests that certain regional 18F-FDG PET radiomics features can effectively identify DIILD in HL patients treated with bleomycin, as significant longitudinal increases were observed in SUVmean, texture strength, and zone distance entropy after the development of DIILD. The metabolic activity of these features did not significantly increase over time in DLBCL patients and HL patients who did not develop DIILD. This indicates that 18F-FDG PET radiomics, with and without machine learning, might serve as potential biomarkers for detecting DIILD.

NIMG-42. ANALYSIS OF TUMOR RELAPSE PROBABILITY AND OVERALL SURVIVAL PREDICTION FOLLOWING TEMOZOLOMIDE CHEMORADIATION USING FET PET INCLUDING RADIOMICS IN PATIENTS WITH GLIOBLASTOMA

Abstract BACKGROUND Early after surgery and completion of first-line concomitant chemoradiation with temozolomide, the evaluation of tumor relapse probability and overall survival prediction is of considerable interest for the management of patients with glioblastoma METHODS Sixty-three newly diagnosed patients with glioblastoma (age range, 19-82) who received static and dynamic PET imaging using the radiolabeled amino acid O-(2-[18F]fluoroethyl)-L-tyrosine (FET) after surgery or biopsy and completion of concomitant chemoradiation with temozolomide were evaluated. Static FET PET parameters such as maximum and mean tumor-to-brain ratios (TBRmax, TBRmean) and metabolic tumor volumes (MTV), and the dynamic FET PET parameter time-to-peak (TTP) were obtained. Additionally, FET PET radiomics features (n=1,303) were extracted for each patient and a test-retest analysis was performed to identify robust features prior to feature selection. The prognostic value of FET PET parameters and radiomics features was evaluated using receiver-operating-characteristic (ROC) analyses with regard to a significantly prolonged progression-free and overall survival (PFS, OS). Subsequently, univariate and multivariate survival estimates were performed to assess the prognostic value of these parameters to predict a significantly longer PFS and OS RESULTS ROC analyses revealed that the parameter TBRmax was a powerful parameter to predict both a significantly longer PFS (20.3 vs. 8.9 months; P=0.002; threshold, 2.75) and OS (34.9 vs. 18.7 months; P=0.005; threshold, 2.75). MTV had a similar prognostic value for both PFS (16.7 vs. 7.2 months; P=0.002; threshold, 34.0 mL) and OS (30.6 vs. 14.8 months; P=0.002; threshold, 32.6 mL). After feature selection, 3 of 15 selected radiomics features predicted a significantly longer PFS and OS in univariate analyses. TBRmax, MTV, and one of 15 radiomics features remained significant in multivariate survival analysis (all P≤0.03) CONCLUSION Our results suggest that FET PET and radiomics parameters were highly prognostic in this group of patients at an early stage of first-line therapy.

Enhancing Survival Outcome Predictions in Metastatic Non-Small Cell Lung Cancer Through PET Radiomics Analysis

(1) Background: Advanced-stage lung cancer poses significant management challenges. The goal of this study was to identify crucial clinical and PET radiomics features that enable prognostic stratification for predicting outcomes. (2) Methods: PET radiomics features of the primary lung lesions were extracted from 99 patients with stage IVB NSCLC, and the robustness of these PET radiomics features was evaluated against uncertainties stemming from extraction parameters and contour variation. We trained three survival risk models (clinical, radiomics, and a composite) through a penalized Cox model framework. We also created a Balanced Random Forest classification predictive model, using the selected features, to predict 1-year survival. (3) Results: We identified 367 common PET radiomics features that exhibited robustness to perturbations introduced by contour variation and extraction parameters. Our findings indicated that both the radiomics and the composite model outperformed the clinical model in stratifying the risk for survival with statistical significance. In predicting 1-year survival, the radiomics model and the composite model also achieved better predicting accuracies compared to the clinical model. (4) Conclusions: Robust PET radiomics analysis successfully facilitated the stratification of patient risk for survival outcomes and predicted 1-year survival in stage IVB NSCLC.

Machine learning predicts conventional imaging metastasis-free survival (MFS) for oligometastatic castration-sensitive prostate cancer (omCSPC) using prostate-specific membrane antigen (PSMA) PET radiomics

PurposeThis study investigated imaging biomarkers derived from PSMA-PET acquired pre- and post-metastasis-directed therapy (MDT) to predict 2-year metastasis-free survival (MFS), which provides valuable early response assessment to improve patient outcomes. Materials/MethodsAn international cohort of 117 oligometastatic castration-sensitive prostate cancer (omCSPC) patients, comprising 34 from John Hopkins Hospital (JHH) and 83 from Baskent University (BU), were treated with stereotactic ablative radiation therapy (SABR) MDT with both pre- and post-MDT PSMA-PET/CT scans acquired. PET radiomic features were analyzed from a CT-PET fusion defined gross tumor volume ((GTV) or zone 1), and a 5 mm expansion ring area outside the GTV (zone 2). A total of 1748 PET radiomic features were extracted from these zones. The six most significant features selected using the Chi2 method, along with five clinical parameters (age, Gleason score, number of total lesions, untreated lesions, and pre-MDT prostate-specific antigen (PSA)) were extracted as inputs to the models. Various machine learning models, including Random Forest, Decision Tree, Support Vector Machine, and Naïve Bayesian, were employed for 2-year MFS prediction and tested using leave-one-out and cross-institution validation. ResultsSix radiomic features, including Total Energy, Entropy, and Standard Deviation from pre-PSMA-PET zone 1, Total Energy and Contrast from post-PSMA-PET zone 1, and Entropy from pre-PSMA-PET zone 2, along with five clinical parameters were selected for predicting 2-year MFS. In a leave-one-out test with all the patients, random forest achieved an accuracy of 80 % and an AUC of 0.82 in predicting 2-year MFS. In cross-institution validation, the model correctly predicted 2-year MFS events with an accuracy of 75 % and an AUC of 0.77 for patients from JHH, and an accuracy of 78 % and an AUC of 0.80 for BU patients, respectively. ConclusionOur study demonstrated the promise of using pre- and post-MDT PSMA-PET-based imaging biomarkers for MFS prediction for omCSPC patients.

Does consensus contour improve robustness and accuracy in 18F-FDG PET radiomic features?

PurposeThe purpose of our study is to validate the robustness and accuracy of consensus contour in 2-deoxy-2-[18\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^{18}$$\\end{document}F]fluoro-D-glucose (18\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^{18}$$\\end{document}F-FDG) PET radiomic features.Methods225 nasopharyngeal carcinoma (NPC) and 13 extended cardio-torso (XCAT) simulated data were enrolled. All segmentation were performed with four segmentation methods under two different initial masks, respectively. Consensus contour (ConSeg) was then developed using the majority vote rule. 107 radiomic features were extracted by Pyradiomics based on segmentation and the intraclass correlation coefficient (ICC) was calculated for each feature between masks or among segmentation, respectively. In XCAT ICC between segmentation and simulated ground truth were also calculated to access the accuracy.ResultsICC varied with the dataset, segmentation method, initial mask and feature type. ConSeg presented higher ICC for radiomic features in robustness tests and similar ICC in accuracy tests, compared with the average of four segmentation results. Higher ICC were also generally observed in irregular initial masks compared with rectangular masks in both robustness and accuracy tests. Furthermore, 19 features (17.76%) had ICC ≥ 0.75 in both robustness and accuracy tests for any of the segmentation methods or initial masks. The dataset was observed to have a large impact on the correlation relationships between radiomic features, but not the segmentation method or initial mask.ConclusionsThe consensus contour combined with irregular initial mask could improve the robustness and accuracy in radiomic analysis to some extent. The correlation relationships between radiomic features and feature clusters largely depended on the dataset, but not segmentation method or initial mask.

Preliminary study on the ability of the machine learning models based on 18F-FDG PET/CT to differentiate between mass-forming pancreatic lymphoma and pancreatic carcinoma

PurposeThe objective of this study was to preliminarily assess the ability of metabolic parameters and radiomics derived from 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) to distinguish mass-forming pancreatic lymphoma from pancreatic carcinoma using machine learning. MethodsA total of 88 lesions from 86 patients diagnosed as mass-forming pancreatic lymphoma or pancreatic carcinoma were included and randomly divided into a training set and a validation set at a 4-to-1 ratio. The segmentation of regions of interest was performed using ITK-SNAP software, PET metabolic parameters and radiomics features were extracted using 3Dslicer and PYTHON. Following the selection of optimal metabolic parameters and radiomics features, Logistic regression (LR), support vector machine (SVM), and random forest (RF) models were constructed for PET metabolic parameters, CT radiomics, PET radiomics, and PET/CT radiomics. Model performance was assessed in terms of area under the curve (AUC), accuracy, sensitivity, and specificity in both the training and validation sets. ResultsStrong discriminative ability observed in all models, with AUC values ranging from 0.727 to 0.978. The highest performance exhibited by the combined PET and CT radiomics features. AUC values for PET/CT radiomics models in the training set were LR 0.994, SVM 0.994, RF 0.989. In the validation set, AUC values were LR 0.909, SVM 0.883, RF 0.844. ConclusionMachine learning models utilizing the metabolic parameters and radiomics of 18F-FDG PET/CT show promise in distinguishing between pancreatic carcinoma and mass-forming pancreatic lymphoma. Further validation on a larger cohort is necessary before practical implementation in clinical settings.

Histology-Based Radiomics for [18F]FDG PET Identifies Tissue Heterogeneity in Pancreatic Cancer.

Radiomics features can reveal hidden patterns in a tumor but usually lack an underlying biologic rationale. In this work, we aimed to investigate whether there is a correlation between radiomics features extracted from [18F]FDG PET images and histologic expression patterns of a glycolytic marker, monocarboxylate transporter-4 (MCT4), in pancreatic cancer. Methods: A cohort of pancreatic ductal adenocarcinoma patients (n = 29) for whom both tumor cross sections and [18F]FDG PET/CT scans were available was used to develop an [18F]FDG PET radiomics signature. By using immunohistochemistry for MCT4, we computed density maps of MCT4 expression and extracted pathomics features. Cluster analysis identified 2 subgroups with distinct MCT4 expression patterns. From corresponding [18F]FDG PET scans, radiomics features that associate with the predefined MCT4 subgroups were identified. Results: Complex heat map visualization showed that the MCT4-high/heterogeneous subgroup was correlating with a higher MCT4 expression level and local variation. This pattern linked to a specific [18F]FDG PET signature, characterized by a higher SUVmean and SUVmax and second-order radiomics features, correlating with local variation. This MCT4-based [18F]FDG PET signature of 7 radiomics features demonstrated prognostic value in an independent cohort of pancreatic cancer patients (n = 71) and identified patients with worse survival. Conclusion: Our cross-modal pipeline allows the development of PET scan signatures based on immunohistochemical analysis of markers of a particular biologic feature, here demonstrated on pancreatic cancer using intratumoral MCT4 expression levels to select [18F]FDG PET radiomics features. This study demonstrated the potential of radiomics scores to noninvasively capture intratumoral marker heterogeneity and identify a subset of pancreatic ductal adenocarcinoma patients with a poor prognosis.

Multicenter PET image harmonization using generative adversarial networks

PurposeTo improve reproducibility and predictive performance of PET radiomic features in multicentric studies by cycle-consistent generative adversarial network (GAN) harmonization approaches.MethodsGAN-harmonization was developed to harmonize whole-body PET scans to perform image style and texture translation between different centers and scanners. GAN-harmonization was evaluated by application to two retrospectively collected open datasets and different tasks. First, GAN-harmonization was performed on a dual-center lung cancer cohort (127 female, 138 male) where the reproducibility of radiomic features in healthy liver tissue was evaluated. Second, GAN-harmonization was applied to a head and neck cancer cohort (43 female, 154 male) acquired from three centers. Here, the clinical impact of GAN-harmonization was analyzed by predicting the development of distant metastases using a logistic regression model incorporating first-order statistics and texture features from baseline 18F-FDG PET before and after harmonization.ResultsImage quality remained high (structural similarity: left kidney ≥\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\ge$$\\end{document} 0.800, right kidney ≥\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\ge$$\\end{document} 0.806, liver ≥\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\ge$$\\end{document} 0.780, lung ≥\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\ge$$\\end{document} 0.838, spleen ≥\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\ge$$\\end{document} 0.793, whole-body ≥\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\ge$$\\end{document} 0.832) after image harmonization across all utilized datasets. Using GAN-harmonization, inter-site reproducibility of radiomic features in healthy liver tissue increased at least by ≥\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\ge$$\\end{document} 5 ± 14% (first-order), ≥\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\ge$$\\end{document} 16 ± 7% (GLCM), ≥\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\ge$$\\end{document} 19 ± 5% (GLRLM), ≥\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\ge$$\\end{document} 16 ± 8% (GLSZM), ≥\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\ge$$\\end{document} 17 ± 6% (GLDM), and ≥\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\ge$$\\end{document} 23 ± 14% (NGTDM). In the head and neck cancer cohort, the outcome prediction improved from AUC 0.68 (95% CI 0.66–0.71) to AUC 0.73 (0.71–0.75) by application of GAN-harmonization.ConclusionsGANs are capable of performing image harmonization and increase reproducibility and predictive performance of radiomic features derived from different centers and scanners.

Machine learning techniques based on 18F-FDG PET radiomics features of temporal regions for the classification of temporal lobe epilepsy patients from healthy controls.

This study aimed to investigate the clinical application of 18F-FDG PET radiomics features for temporal lobe epilepsy and to create PET radiomics-based machine learning models for differentiating temporal lobe epilepsy (TLE) patients from healthy controls. A total of 347 subjects who underwent 18F-FDG PET scans from March 2014 to January 2020 (234 TLE patients: 25.50 ± 8.89 years, 141 male patients and 93 female patients; and 113 controls: 27.59 ± 6.94 years, 48 male individuals and 65 female individuals) were allocated to the training (n = 248) and test (n = 99) sets. All 3D PET images were registered to the Montreal Neurological Institute template. PyRadiomics was used to extract radiomics features from the temporal regions segmented according to the Automated Anatomical Labeling (AAL) atlas. The least absolute shrinkage and selection operator (LASSO) and Boruta algorithms were applied to select the radiomics features significantly associated with TLE. Eleven machine-learning algorithms were used to establish models and to select the best model in the training set. The final radiomics features (n = 7) used for model training were selected through the combinations of the LASSO and the Boruta algorithms with cross-validation. All data were randomly divided into a training set (n = 248) and a testing set (n = 99). Among 11 machine-learning algorithms, the logistic regression (AUC 0.984, F1-Score 0.959) model performed the best in the training set. Then, we deployed the corresponding online website version (https://wane199.shinyapps.io/TLE_Classification/), showing the details of the LR model for convenience. The AUCs of the tuned logistic regression model in the training and test sets were 0.981 and 0.957, respectively. Furthermore, the calibration curves demonstrated satisfactory alignment (visually assessed) for identifying the TLE patients. The radiomics model from temporal regions can be a potential method for distinguishing TLE. Machine learning-based diagnosis of TLE from preoperative FDG PET images could serve as a useful preoperative diagnostic tool.

Impact of 18F-FDG PET Intensity Normalization on Radiomic Features of Oropharyngeal Squamous Cell Carcinomas and Machine Learning-Generated Biomarkers.

We aimed to investigate the effects of 18F-FDG PET voxel intensity normalization on radiomic features of oropharyngeal squamous cell carcinoma (OPSCC) and machine learning-generated radiomic biomarkers. Methods: We extracted 1,037 18F-FDG PET radiomic features quantifying the shape, intensity, and texture of 430 OPSCC primary tumors. The reproducibility of individual features across 3 intensity-normalized images (body-weight SUV, reference tissue activity ratio to lentiform nucleus of brain and cerebellum) and the raw PET data was assessed using an intraclass correlation coefficient (ICC). We investigated the effects of intensity normalization on the features' utility in predicting the human papillomavirus (HPV) status of OPSCCs in univariate logistic regression, receiver-operating-characteristic analysis, and extreme-gradient-boosting (XGBoost) machine-learning classifiers. Results: Of 1,037 features, a high (ICC ≥ 0.90), medium (0.90 > ICC ≥ 0.75), and low (ICC < 0.75) degree of reproducibility across normalization methods was attained in 356 (34.3%), 608 (58.6%), and 73 (7%) features, respectively. In univariate analysis, features from the PET normalized to the lentiform nucleus had the strongest association with HPV status, with 865 of 1,037 (83.4%) significant features after multiple testing corrections and a median area under the receiver-operating-characteristic curve (AUC) of 0.65 (interquartile range, 0.62-0.68). Similar tendencies were observed in XGBoost models, with the lentiform nucleus-normalized model achieving the numerically highest average AUC of 0.72 (SD, 0.07) in the cross validation within the training cohort. The model generalized well to the validation cohorts, attaining an AUC of 0.73 (95% CI, 0.60-0.85) in independent validation and 0.76 (95% CI, 0.58-0.95) in external validation. The AUCs of the XGBoost models were not significantly different. Conclusion: Only one third of the features demonstrated a high degree of reproducibility across intensity-normalization techniques, making uniform normalization a prerequisite for interindividual comparability of radiomic markers. The choice of normalization technique may affect the radiomic features' predictive value with respect to HPV. Our results show trends that normalization to the lentiform nucleus may improve model performance, although more evidence is needed to draw a firm conclusion.

Evaluation of bone marrow invasion on the machine learning of 18 F-FDG PET texture analysis in lower gingival squamous cell carcinoma.

Lower gingival squamous cell carcinoma (LGSCC) has the potential to invade the alveolar bone. Traditionally, the diagnosis of LGSCC relied on morphological imaging, but inconsistencies between these assessments and surgical findings have been observed. This study aimed to assess the correlation between LGSCC bone marrow invasion and PET texture features and to enhance diagnostic accuracy by using machine learning. A retrospective analysis of 159 LGSCC patients with pretreatment 18 F-fluorodeoxyglucose (FDG) PET/computed tomography (CT) examination from 2009 to 2017 was performed. We extracted radiomic features from the PET images, focusing on pathologic bone marrow invasion detection. Extracted features underwent the least absolute shrinkage and selection operator algorithm-based selection and were then used for machine learning via the XGBoost package to distinguish bone marrow invasion presence. Receiver operating characteristic curve analysis was performed. From the 159 patients, 88 qualified for further analysis (59 men; average age, 69.2 years), and pathologic bone marrow invasion was identified in 69 (78%) of these patients. Three significant radiological features were identified: Gray level co-occurrence matrix_Correlation, INTENSITY-BASED_IntensityInterquartileRange, and MORPHOLOGICAL_SurfaceToVolumeRatio. An XGBoost machine-learning model, using PET radiomic features to detect bone marrow invasion, yielded an area under the curve value of 0.83. Our findings highlighted the potential of 18 F-FDG PET radiomic features, combined with machine learning, as a promising avenue for improving LGSCC diagnosis and treatment. Using 18 F-FDG PET texture features may provide a robust and accurate method for determining the presence or absence of bone marrow invasion in LGSCC patients.

An assessment of PET and CMR radiomic features for the detection of cardiac sarcoidosis.

Visual interpretation of PET and CMR may fail to identify cardiac sarcoidosis (CS) with high specificity. This study aimed to evaluate the role of [18F]FDG PET and late gadolinium enhancement (LGE)-CMR radiomic features in differentiating CS from another cause of myocardial inflammation, in this case patients with cardiac-related clinical symptoms following COVID-19. [18F]FDG PET and LGE-CMR were treated separately in this work. There were 35 post-COVID-19 (PC) and 40 CS datasets. Regions of interest were delineated manually around the entire left ventricle for the PET and LGE-CMR datasets. Radiomic features were then extracted. The ability of individual features to correctly identify image data as CS or PC was tested to predict the clinical classification of CS vs. PC using Mann-Whitney U-tests and logistic regression. Features were retained if the P-value was <0.00053, the AUC was >0.5, and the accuracy was >0.7. After applying the correlation test, uncorrelated features were used as a signature (joint features) to train machine learning classifiers. For LGE-CMR analysis, to further improve the results, different classifiers were used for individual features besides logistic regression, and the results of individual features of each classifier were screened to create a signature that included all features that followed the previously mentioned criteria and used it them as input for machine learning classifiers. The Mann-Whitney U-tests and logistic regression were trained on individual features to build a collection of features. For [18F]FDG PET analysis, the maximum target-to-background ratio (TBRmax ) showed a high area under the curve (AUC) and accuracy with small P-values (<0.00053), but the signature performed better (AUC 0.98 and accuracy 0.91). For LGE-CMR analysis, the Gray Level Dependence Matrix (gldm)-Dependence Non-Uniformity showed good results with small error bars (accuracy 0.75 and AUC 0.87). However, by applying a Support Vector Machine classifier to individual LGE-CMR features and creating a signature, a Random Forest classifier displayed better AUC and accuracy (0.91 and 0.84, respectively). Using radiomic features may prove useful in identifying individuals with CS. Some features showed promising results in differentiating between PC and CS. By automating the analysis, the patient management process can be accelerated and improved.

Noninvasive Assessment of HER2 Expression Status in Gastric Cancer Using 18F-FDG Positron Emission Tomography/Computed Tomography-Based Radiomics: A Pilot Study.

Purpose: Immunohistochemistry (IHC) is the main method to detect human epidermal growth factor receptor 2 (HER2) expression levels. However, IHC is invasive and cannot reflect HER2 expression status in real time. The aim of this study was to construct and verify three types of radiomics models based on 18F-fuorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) imaging and to evaluate the predictive ability of these radiomics models for the expression status of HER2 in patients with gastric cancer (GC). Patients and Methods: A total of 118 patients with GC were enrolled in this study. 18F-FDG PET/CT imaging was performed prior to surgery. The LIFEx software package was applied to extract PET and CT radiomics features. The minimum absolute contraction and selection operator (least absolute shrinkage and selection operator [LASSO]) algorithm was used to select the best radiomics features. Three machine learning methods, logistic regression (LR), support vector machine (SVM), and random forest (RF) models, were constructed and verified. The Synthetic Minority Oversampling Technique (SMOTE) was applied to address data imbalance. Results: In the training and test sets, the area under the curve (AUC) values of the LR, SVM, and RF models were 0.809, 0.761, 0.861 and 0.628, 0.993, 0.717, respectively, and the Brier scores were 0.118, 0.214, and 0.143, respectively. Among the three models, the LR and RF models exhibited extremely good prediction performance. The AUC values of the three models significantly improved after SMOTE balanced the data. Conclusions: 18F-FDG PET/CT-based radiomics models, especially LR and RF models, demonstrate good performance in predicting HER2 expression status in patients with GC and can be used to preselect patients who may benefit from HER2-targeted therapy.

The role of FDG PET/CT radiomics in the prediction of pathological response to neoadjuvant treatment in patients with esophageal cancer.

Attainment of a complete histopathological response following neoadjuvant therapy has been associated with favorable long-term survival outcomes in esophageal cancer patients. We investigated the ability of 18F-fluorodeoxyglucose positron emission tomography/computed tomography (FDG PET/CT) radiomic features to predict the pathological response to neoadjuvant treatment in patients with esophageal cancer. A retrospective review of medical records of patients with locally advanced resectable esophageal or esophagogastric junctional cancers. Included patients had a baseline FDG PET/CT scan and underwent Chemoradiotherapy for Oesophageal Cancer Followed by Surgery Study (CROSS) protocol followed by surgery. Four demographic variables and 107 PET radiomic features were extracted and analyzed using univariate and multivariate analyses to predict response to neoadjuvant therapy. Overall, 53 FDG-avid primary esophageal cancer lesions were segmented and radiomic features were extracted. Seventeen radiomic features and 2 non-radiomics variables were found to exhibit significant differences between neoadjuvant therapy responders and non-responders. An unsupervised hierarchical clustering analysis using these 19 variables classified patients in a manner significantly associated with response to neoadjuvant treatment (p < 0.01). Our findings highlight the potential of FDG PET/CT radiomic features as a predictor for the response to neoadjuvant therapy in esophageal cancer patients. The combination of these radiomic features with select non-radiomic variables provides a model for stratifying patients based on their likelihood to respond to neoadjuvant treatment.

Development and validation of clinical-radiomics analysis for preoperative prediction of IDH mutation status and WHO grade in diffuse gliomas: a consecutive L-[methyl-11C] methionine cohort study with two PET scanners.

Determination of isocitrate dehydrogenase (IDH) genotype is crucial in the stratification of diagnosis and prognostication in diffuse gliomas. We sought to build and validate radiomics models and clinical features incorporated nomogram for preoperative prediction of IDH mutation status and WHO grade of diffuse gliomas with L-[methyl-11C] methionine ([11C]MET) PET/CT imaging according to the 2016 WHO classification of tumors of the central nervous system. Consecutive 178 preoperative [11C]MET PET/CT images were retrospectively studied for radiomics analysis. One hundred six patients from PET scanner 1 were used as training dataset, and 72 patients from PET scanner 2 were used for validation dataset. [11C]MET PET and integrated CT radiomics features were extracted, respectively; three independent predictive models were built based on PET features, CT features, and combined PET/CT features, respectively. The SelectKBest method, Spearman correlation analysis, Least Absolute Shrinkage and Selection Operator (LASSO) regression, and machine learning algorithms were applied for feature selection and model building. After filtering the satisfactory predictive model, key clinical features were incorporated for the nomogram establishment. The combined [11C]MET PET/CT radiomics model, which consisted of four PET features and eight integrated CT features, was significantly associated with IDH genotype (p < 0.0001 for both training and validation datasets). Nomogram based on the [11C]MET PET/CT radiomics score, patients' age, and dichotomous tumor location status showed satisfactory discrimination capacity, and the AUC was 0.880 (95% CI, 0.726-0.998) in the training dataset and 0.866 (95% CI, 0.777-0.956) in the validation dataset. In IDH stratified WHO grade prediction, the final radiomics model consists of four PET features and two CT features had reasonable and stable differential efficacy of WHO grade II and III patients from grade IV patients in IDH-wildtype patients, and the AUC was 0.820 (95% CI, 0.541-1.000) in the training dataset and 0.766 (95% CI, 0.612-0.921) in the validation dataset. [11C]MET PET radiomics features could benefit non-invasive IDH genotype prediction, and integrated CT radiomics features could enhance the efficacy. Radiomics and clinical features incorporation could establish satisfactory nomogram for clinical application. This non-invasive predictive investigation based on our consecutive cohort from two PET scanners could provide the perspective to observe the differential efficacy and the stability of radiomics-based investigation in untreated diffuse gliomas.

Artificial Intelligence Derived Changes between Baseline and Interim FDG-PET/CT Radiomics Features Are Associated with Survival Outcomes in Diffuse Large B-Cell Lymphoma (DLBCL)

Background: Relapsed DLBCL occurs in up to 40% of patients and has a poor prognosis. Current prognostic tools such as the international prognostic index (IPI) and newer forms (NCCN-IPI) only include baseline clinical features. PET/CT features have not yet been incorporated into prognostic risk models despite being essential for staging and response assessment. Deep learning computer algorithms have consistently demonstrated the ability to analyze imaging data and identify features that are not readily apparent to the human eye. Using a deep learning computer vision model that automatically extracted radiomic features from PET/CTs, we examined the association of changes between baseline PET and interim PET (PET2) radiomic features on survival outcomes and compared the prognostic performance with NCCN-IPI. Methods: Adult patients with newly diagnosed diffuse large B-cell lymphoma (DLBCL) between 2005 and 2015 and enrolled in the prospective MER cohort were included. Native DICOM images from FDG PET/CT scans at baseline and interim PET (PET2, after 2-3 cycles of chemotherapy) were preprocessed for regularity and then segmented using the nnU-Net architecture with an external lymphoma-specific model. Feature extraction and standard uptake value (SUV) characterization were accomplished with PyRadiomics and NiBabel. Our computational method included no manual segmentation correction. Individual radiomics features (n=115) and the % change above (Δ High) and below (Δ low) 50 th quartile from baseline to PET2 were evaluated for association with event free survival (EFS) and overall survival (OS) using a univariate Cox regression model and by Kaplan Meier analysis. EFS was defined as time from PET2 to first systemic relapse or death from any cause. OS was defined as time from PET2 to death from any cause. Additionally, Spearman and Pearson correlation matrices were used to guide feature selection while minimizing collinearity. We then assessed the association of the highest scoring radiomics features with EFS in a multivariate cox regression model. Results: A total of 506 patients had baseline and PET2 scans available for radiomic feature extraction. Median age at diagnosis was 61 years (range 18-90), 43% were females. NCCN-IPI composition and outcomes of our cohort was comparable to the previously published NCCN cohort, with patient distribution and 5-year OS as follows: Low (9%, 5-y OS 83%), Low-intermediate (38%, 5-y OS 49%), High-intermediate (41%, 5-y OS 31%), High (11%, 5-y OS 12%). At a median follow up of 81 months,there were 308 events, and272 patients had died.Several radiomics features were associated with EFS including change in tumor surface area (SA Δ) (HR: 1.75, 95% CI 1.36-2.24, p&amp;lt;0.005, median of 26,927 mm 2, Figure 1A) and change in total metabolic tumor volume (TMTV Δ) (HR: 1.64, 95% CI 1.27-2.10, p&amp;lt;0.005, median of 120,473 mm 3, Figure 1B). As a reference, NCCN-IPI univariate Cox regression analysis of EFS produced a Harrell's C-statistic (c) of 0.624. Single radiomics features of SA Δ and TMTV Δ individually produced unadjusted c-statistics of 0.616 and 0.614 respectively. Combinations of radiomics features (SA Δ, TMTV Δ and NCCN-IPI classification) demonstrated cumulative benefit (c=0.652). Conclusion: Automatic PET-CT radiomic feature extraction using deep learning models is feasible among patients with newly diagnosed DLBCL. The change in tumor SA and TMTV appear to be promising radiomic biomarkers. A low decline in either tumor SA or TMTV by PET2 is associated with higher risk of relapse and poor survival outcomes. As individual characteristics, these features are similar prognostic indicators of EFS as the NCCN-IPI. Further investigation into risk stratification when combining radiomics with clinical and genomic variables is warranted. Figure 1. Event free survival for DLBCL patients with A) a change in tumor surface area by PET2 less than 50% percentile (SA_Δ Low) and greater than 50 th percentile (SA_Δ High) and B) a change in TMTV by PET2 less than 50% percentile (TMTV_Δ Low) and greater than 50 th percentile (TMTV_Δ High).

Prognostic Value of 18F-FDG PET Radiomics Features at Baseline in PET-Guided Consolidation Strategy in Diffuse Large B-Cell Lymphoma: A Machine-Learning Analysis from the GAINED Study.

The results of the GA in Newly Diagnosed Diffuse Large B-Cell Lymphoma (GAINED) study demonstrated the success of an 18F-FDG PET-driven approach to allow early identification-for intensification therapy-of diffuse large B-cell lymphoma patients with a high risk of relapse. Besides, some works have reported the prognostic value of baseline PET radiomics features (RFs). This work investigated the added value of such biomarkers on survival of patients involved in the GAINED protocol. Methods: Conventional PET features and RFs were computed from 18F-FDG PET at baseline and extracted using different volume definitions (patient level, largest lesion, and hottest lesion). Clinical features and the consolidation treatment information were also considered in the model. Two machine-learning pipelines were trained with 80% of patients and tested on the remaining 20%. The training was repeated 100 times to highlight the test set variability. For the 2-y progression-free survival (PFS) outcome, the pipeline included a data augmentation and an elastic net logistic regression model. Results for different feature groups were compared using the mean area under the curve (AUC). For the survival outcome, the pipeline included a Cox univariate model to select the features. Then, the model included a split between high- and low-risk patients using the median of a regression score based on the coefficients of a penalized Cox multivariate approach. The log-rank test P values over the 100 loops were compared with a Wilcoxon signed-ranked test. Results: In total, 545 patients were included for the 2-y PFS classification and 561 for survival analysis. Clinical features alone, consolidation features alone, conventional PET features, and RFs extracted at patient level achieved an AUC of, respectively, 0.65 ± 0.07, 0.64 ± 0.06, 0.60 ± 0.07, and 0.62 ± 0.07 (0.62 ± 0.07 for the largest lesion and 0.54 ± 0.07 for the hottest). Combining clinical features with the consolidation features led to the best AUC (0.72 ± 0.06). Adding conventional PET features or RFs did not improve the results. For survival, the log-rank P values of the model involving clinical and consolidation features together were significantly smaller than all combined-feature groups (P < 0.007). Conclusion: The results showed that a concatenation of multimodal features coupled with a simple machine-learning model does not seem to improve the results in terms of 2-y PFS classification and PFS prediction for patient treated according to the GAINED protocol.

Impact of respiratory motion on 18 F-FDG PET radiomics stability: Clinical evaluation with a digital PET scanner.

18 F-FDG PET quantitative features are susceptible to respiratory motion. However, studies using clinical patient data to explore the impact of respiratory motion on 18 F-FDG PET radiomic features are limited. In this study, we investigated the impact of respiratory motion on radiomics stability with clinical 18 F-FDG PET images using a data-driven gating (DDG) algorithm on the digital PET scanner. A total of 101 patients who underwent oncological 18 F-FDG PET scans were retrospectively included. A DDG algorithm combined with a motion compensation technique was used to extract the PET images with respiratory motion correction. 18 F-FDG-avid lesions from the thorax to the upper abdomen were analyzed on the non-DDG and DDG PET images. The lesions were segmented with a 40% threshold of the maximum standardized uptake. A total of 725 radiomic features were computed from the segmented lesions, including first-order, shape, texture, and wavelet features. The intraclass correlation coefficient (ICC) and coefficient of variation (COV) were calculated to evaluate feature stability. An ICC above 0.9 and a COV below 5% were considered high stability. In total, 168 lesions with and without respiratory motion correction were analyzed. Our results indicated that most 18 F-FDG PET radiomic features are sensitive to respiratory motion. Overall, only 27 out of 725 (3.72%) radiomic features were identified as highly stable, including one from the first-order features (entropy), one from the shape features (sphericity), four from the gray-level co-occurrence matrix features (normalized and unnormalized inverse difference moment, joint entropy, and sum entropy), one from the gray-level run-length matrix features (run entropy), and 20 from the wavelet filter-based features. Respiratory motion has a significant impact on 18 F-FDG PET radiomics stability. The highly stable features identified in our study may serve as potential candidates for further applications, such as machine learning modeling.

Prognostic 18F-FDG Radiomic Features in Advanced High-Grade Serous Ovarian Cancer.

High-grade serous ovarian cancer (HGSOC) is an aggressive disease with different clinical outcomes and poor prognosis. This could be due to tumor heterogeneity. The 18F-FDG PET radiomic parameters permit addressing tumor heterogeneity. Nevertheless, this has been not well studied in ovarian cancer. The aim of our work was to assess the prognostic value of pretreatment 18F-FDG PET radiomic features in patients with HGSOC. A review of 36 patients diagnosed with advanced HGSOC between 2016 and 2020 in our center was performed. Radiomic features were obtained from pretreatment 18F-FDGPET. Disease-free survival (DFS) and overall survival (OS) were calculated. Optimal cutoff values with receiver operating characteristic curve/median values were used. A correlation between radiomic features and DFS/OS was made. The mean DFS was 19.6 months and OS was 37.1 months. Total Lesion Glycolysis (TLG), GLSZM_ Zone Size Non-Uniformity (GLSZM_ZSNU), and GLRLM_Run Length Non-Uniformity (GLRLM_RLNU) were significantly associated with DFS. The survival-curves analysis showed a significant difference of DSF in patients with GLRLM_RLNU > 7388.3 versus patients with lower values (19.7 months vs. 31.7 months, p = 0.035), maintaining signification in the multivariate analysis (p = 0.048). Moreover, Intensity-based Kurtosis was associated with OS (p = 0.027). Pretreatment 18F-FDG PET radiomic features GLRLM_RLNU, GLSZM_ZSNU, and Kurtosis may have prognostic value in patients with advanced HGSOC.

Exploring published and novel pre-treatment CT and PET radiomics to stratify risk of progression among early-stage non-small cell lung cancer patients treated with stereotactic radiation

PurposeDisease progression after definitive stereotactic body radiation therapy (SBRT) for early-stage non-small cell lung cancer (NSCLC) occurs in 20–40% of patients. Here, we explored published and novel pre-treatment CT and PET radiomics features to identify patients at risk of progression. Materials/MethodsPublished CT and PET features were identified and explored along with 15 other CT and PET features in 408 consecutively treated early-stage NSCLC patients having CT and PET < 3 months pre-SBRT (training/set-aside validation subsets: n = 286/122). Features were associated with progression-free survival (PFS) using bootstrapped Cox regression (Bonferroni-corrected univariate predictor: p ≤ 0.002) and only non-strongly correlated predictors were retained (|Rs|<0.70) in forward-stepwise multivariate analysis. ResultsTumor diameter and SUVmax were the two most frequently reported features associated with progression/survival (in 6/20 and 10/20 identified studies). These two features and 12 of the 15 additional features (CT: 6; PET: 6) were candidate PFS predictors. A re-fitted model including diameter and SUVmax presented with the best performance (c-index: 0.78; log-rank p-value < 0.0001). A model built with the two best additional features (CTspiculation1 and SUVentropy) had a c-index of 0.75 (log-rank p-value < 0.0001). ConclusionsA re-fitted pre-treatment model using the two most frequently published features – tumor diameter and SUVmax – successfully stratified early-stage NSCLC patients by PFS after receiving SBRT.