A Postreconstruction Harmonization Method for Multicenter Radiomic Studies in PET

Post-acquisition standardization of positron emission tomography images

Purpose

Tissue radiotracer activity measured from positron emission tomography (PET) images is an important biomarker that is clinically utilized for diagnosis, staging, prognostication, and treatment response assessment in patients with cancer and other clinical disorders. Using PET image values to define a normal range of metabolic activity for quantification purposes is challenging due to variations in patient-related factors and technical factors. Although the formulation of standardized uptake value (SUV) has compensated for some of these variabilities, significant non-standardness still persists. We propose an image processing method to substantially mitigate these variabilities.

Methods

The standardization method is similar for activity concentration (AC) PET and SUV PET images with some differences and consists of two steps. The calibration step is performed only once for each of AC PET or SUV PET, employs a set of images of normal subjects, and requires a reference object, while the transformation step is executed for each patient image to be standardized. In the calibration step, a standardized scale is determined along with 3 key image intensity landmarks defined on it including the minimum percentile intensitys min , median intensitys m , and high percentile intensitys max . s min ands m are estimated based on image intensities within the body region in the normal calibration image set. The optimal value of the maximum percentile β corresponding to the intensitys max is estimated via an optimization process by using the reference object to optimally separate the highly variable high uptake values from the normal uptake intensities. In the transformation step, the first two landmarks-the minimum percentile intensityp α ( I ) , and the median intensityp m ( I ) -are found for the given image I for the body region, and the high percentile intensityp β ( I ) is determined corresponding to the optimally estimated high percentile value β . Subsequently, intensities of I are mapped to the standard scale piecewise linearly for different segments. We employ three strategies for evaluation and comparison with other standardization methods: (i) comparing coefficient of variationC V O of mean intensity within test objects O across different normal test subjects before and after standardization; (ii) comparing mean absolute difference (MD O ) of mean intensity within test objects O across different subjects in repeat scans before and after standardization; (iii) comparingC V O of mean intensity across different normal subjects before and after standardization where the scans came from different brands of scanners.

Results

Our data set consisted of 84 FDG-PET/CT scans of the body torso including 38 normal subjects and two repeat-scans of 23 patients. We utilized one of two objects-liver and spleen-as a reference object and the other for testing. The proposed standardization method reducedC V O andMD O by a factor of 3-8 in comparison to other standardization methods and no standardization. Upon standardization by our method, the image intensities (both for AC and SUV) from two different brands of scanners become statistically indistinguishable, while without standardization, they differ significantly and by a factor of 3-9.

Conclusions

The proposed method is automatic, outperforms current standardization methods, and effectively overcomes the residual variation left over in SUV and inter-scanner variations.

Removal of site effects and enhancement of signal using dual projection independent component analysis for pooling multi-site MRI data

Combining magnetic resonance imaging (MRI) data from multi-site studies is a popular approach for constructing larger datasets to greatly enhance the reliability and reproducibility of neuroscience research. However, the scanner/site variability is a significant confound that complicates the interpretation of the results, so effective and complete removal of the scanner/site variability is necessary to realise the full advantages of pooling multi-site datasets. Independent component analysis (ICA) and general linear model (GLM) based harmonisation methods are the two primary methods used to eliminate scanner/site effects. Unfortunately, there are challenges with both ICA-based and GLM-based harmonisation methods to remove site effects completely when the signals of interest and scanner/site effects-related variables are correlated, which may occur in neuroscience studies. In this study, we propose an effective and powerful harmonisation strategy that implements dual projection (DP) theory based on ICA to remove the scanner/site effects more completely. This method can separate the signal effects correlated with site variables from the identified site effects for removal without losing signals of interest. Both simulations and vivo structural MRI datasets, including a dataset from Autism Brain Imaging Data Exchange II and a travelling subject dataset from the Strategic Research Program for Brain Sciences, were used to test the performance of a DP-based ICA harmonisation method. Results show that DP-based ICA harmonisation has superior performance for removing site effects and enhancing the sensitivity to detect signals of interest as compared with GLM-based and conventional ICA harmonisation methods.

The usefulness of machine-learning-based evaluation of clinical and pretreatment 18F-FDG-PET/CT radiomic features for predicting prognosis in patients with laryngeal cancer

OBJECTIVE

To examine whether machine learning (ML) analyses involving clinical and 18F-FDG-PET-based radiomic features are helpful in predicting prognosis in patients with laryngeal cancer.

METHODS

This retrospective study included 49 patients with laryngeal cancer who underwent18F-FDG-PET/CT before treatment, and these patients were divided into the training (n = 34) and testing (n = 15) cohorts.Seven clinical (age, sex, tumor size, T stage, N stage, Union for International Cancer Control stage, and treatment) and 40 18F-FDG-PET-based radiomic features were used to predict disease progression and survival. Six ML algorithms (random forest, neural network, k-nearest neighbors, naïve Bayes, logistic regression, and support vector machine) were used for predicting disease progression. Two ML algorithms (cox proportional hazard and random survival forest [RSF] model) considering for time-to-event outcomes were used to assess progression-free survival (PFS), and prediction performance was assessed by the concordance index (C-index).

RESULTS

Tumor size, T stage, N stage, GLZLM_ZLNU, and GLCM_Entropy were the five most important features for predicting disease progression.In both cohorts, the naïve Bayes model constructed by these five features was the best performing classifier (training: AUC = 0.805; testing: AUC = 0.842). The RSF model using the five features (tumor size, GLZLM_ZLNU, GLCM_Entropy, GLRLM_LRHGE and GLRLM_SRHGE) exhibited the highest performance in predicting PFS (training: C-index = 0.840; testing: C-index = 0.808).

CONCLUSION

ML analyses involving clinical and 18F-FDG-PET-based radiomic features may help predict disease progression and survival in patients with laryngeal cancer.

ADVANCES IN KNOWLEDGE

ML approach using clinical and 18F-FDG-PET-based radiomic features has the potential to predict prognosis of laryngeal cancer.

ComBat Harmonization: Empirical Bayes versus fully Bayes approaches

Studying small effects or subtle neuroanatomical variation requires large-scale sample size data. As a result, combining neuroimaging data from multiple datasets is necessary. Variation in acquisition protocols, magnetic field strength, scanner build, and many other non-biologically related factors can introduce undesirable bias into studies. Hence, harmonization is required to remove the bias-inducing factors from the data. ComBat is one of the most common methods applied to features from structural images. ComBat models the data using a hierarchical Bayesian model and uses the empirical Bayes approach to infer the distribution of the unknown factors. The empirical Bayes harmonization method is computationally efficient and provides valid point estimates. However, it tends to underestimate uncertainty. This paper investigates a new approach, fully Bayesian ComBat, where Monte Carlo sampling is used for statistical inference. When comparing fully Bayesian and empirical Bayesian ComBat, we found Empirical Bayesian ComBat more effectively removed scanner strength information and was much more computationally efficient. Conversely, fully Bayesian ComBat better preserved biological disease and age-related information while performing more accurate harmonization on traveling subjects. The fully Bayesian approach generates a rich posterior distribution, which is useful for generating simulated imaging features for improving classifier performance in a limited data setting. We show the generative capacity of our model for augmenting and improving the detection of patients with Alzheimer's disease. Posterior distributions for harmonized imaging measures can also be used for brain-wide uncertainty comparison and more principled downstream statistical analysis.Code for our new fully Bayesian ComBat extension is available at https://github.com/batmanlab/BayesComBat.

Probabilistic estimation for across-batch compatibility enhancement for amyloid PET

INTRODUCTION

It is necessary to accurately account for systematic differences due to variability in scanners, radiotracers, and acquisition protocols in multisite studies combining amyloid imaging data.

METHODS

We propose Probabilistic Estimation for Across-batch Compatibility Enhancement (PEACE), a fully Bayesian multimodal extension of the widely used ComBat harmonization model, and we apply it to harmonize regional amyloid positron emission tomography data from two scanners.

RESULTS

Simulations show that PEACE recovers true harmonized values better than ComBat, even for unimodal data. PEACE harmonization of multiscanner regional amyloid imaging data yields results that agree better with longitudinal data compared to ComBat, without removing the known biological effects of age or apolipoprotein E genotype.

DISCUSSION

PEACE outperforms ComBat in both unimodal and bimodal contexts, is applicable to multisite amyloid imaging data, and holds promise for the harmonization of other neuroimaging data over ComBat.

HIGHLIGHTS

We introduce PEACE, a fully Bayesian multimodal extension of ComBat harmonization.Simulations show that PEACE recovers true harmonized values better than ComBat.PEACE accurately harmonizes multiscanner regional amyloid imaging data.

Additional Value of PET and CT Image-Based Features in the Detection of Occult Lymph Node Metastases in Lung Cancer: A Systematic Review of the Literature

Lung cancer represents the second most common malignancy worldwide and lymph node (LN) involvement serves as a crucial prognostic factor for tailoring treatment approaches. Invasive methods, such as mediastinoscopy and endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA), are employed for preoperative LN staging. Among the preoperative non-invasive diagnostic methods, computed tomography (CT) and, recently, positron emission tomography (PET)/CT with fluorine-18-fludeoxyglucose ([¹⁸F]FDG) are routinely recommended by several guidelines; however, they can both miss pathologically proven LN metastases, with an incidence up to 26% for patients staged with [¹⁸F]FDG PET/CT. These undetected metastases, known as occult LN metastases (OLMs), are usually cases of micro-metastasis or small LN metastasis (shortest radius below 10 mm). Hence, it is crucial to find novel approaches to increase their discovery rate. Radiomics is an emerging field that seeks to uncover and quantify the concealed information present in biomedical images by utilising machine or deep learning approaches. The extracted features can be integrated into predictive models, as numerous reports have emphasised their usefulness in the staging of lung cancer. However, there is a paucity of studies examining the detection of OLMs using quantitative features derived from images. Hence, the objective of this review was to investigate the potential application of PET- and/or CT-derived quantitative radiomic features for the identification of OLMs.

Variability in PET image quality and quantification measured with a permanently filled 68Ge-phantom: a multi-center study

BACKGROUND

This study evaluated, as a snapshot, the variability in quantification and image quality (IQ) of the clinically utilized PET [¹⁸F]FDG whole-body protocols in Finland using a NEMA/IEC IQ phantom permanently filled with ⁶⁸Ge.

METHODS

The phantom was imaged on 14 PET-CT scanners, including a variety of models from two major vendors. The variability of the recovery coefficients (RC_max, RC_mean and RC_peak) of the hot spheres as well as percent background variability (PBV), coefficient of variation of the background (COV_BG) and accuracy of corrections (AOC) were studied using images from clinical and standardized protocols with 20 repeated measurements. The ranges of the RCs were also compared to the limits of the EARL ¹⁸F standards 2 accreditation (EARL2). The impact of image noise on these parameters was studied using averaged images (AVIs).

RESULTS

The largest variability in RC values of the routine protocols was found for the RC_max with a range of 68% and with 10% intra-scanner variability, decreasing to 36% when excluding protocols with suspected cross-calibration failure or without point-spread-function (PSF) correction. The RC ranges of individual hot spheres in routine or standardized protocols or AVIs fulfilled the EARL2 ranges with two minor exceptions, but fulfilling the exact EARL2 limits for all hot spheres was variable. RC_peak was less dependent on averaging and reconstruction parameters than RC_max and RC_mean. The PBV, COV_BG and AOC varied between 2.3-11.8%, 9.6-17.8% and 4.8-32.0%, respectively, for the routine protocols. The RC ranges, PBV and COV_BG were decreased when using AVIs. With AOC, when excluding routine protocols without PSF correction, the maximum value dropped to 15.5%.

CONCLUSION

The maximum variability of the RC values for the [¹⁸F]FDG whole-body protocols was about 60%. The RC ranges of properly cross-calibrated scanners with PSF correction fitted to the EARL2 RC ranges for individual sphere sizes, but fulfilling the exact RC limits would have needed further optimization. RC_peak was the most robust RC measure. Besides COV_BG, also RCs and PVB were sensitive to image noise.

Predictive value of 18F-FDG PET/CT-based radiomics model for neoadjuvant chemotherapy efficacy in breast cancer: a multi-scanner/center study with external validation

PURPOSE

To develop and validate the predictive value of an 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) model for breast cancer neoadjuvant chemotherapy (NAC) efficacy based on the tumor-to-liver ratio (TLR) radiomic features and multiple data pre-processing methods.

METHODS

One hundred and ninety-three breast cancer patients from multiple centers were retrospectively included in this study. According to the endpoint of NAC, we divided the patients into pathological complete remission (pCR) and non-pCR groups. All patients underwent 18F-FDG PET/CT imaging before NAC treatment, and CT and PET images volume of interest (VOI) segmentation by manual segmentation and semi-automated absolute threshold segmentation, respectively. Then, feature extraction of VOI was performed with the pyradiomics package. A total of 630 models were created based on the source of radiomic features, the elimination of the batch effect approach, and the discretization method. The differences in data pre-processing approaches were compared and analyzed to identify the best-performing model, which was further tested by the permutation test.

RESULTS

A variety of data pre-processing methods contributed in varying degrees to the improvement of model effects. Among them, TLR radiomic features and Combat and Limma methods that eliminate batch effects could enhance the model prediction overall, and data discretization could be used as a potential method that can further optimize the model. A total of seven excellent models were selected and then based on the AUC of each model in the four test sets and their standard deviations, we selected the optimal model. The optimal model predicted AUC between 0.7 and 0.77 for the four test groups, with p-values less than 0.05 for the permutation test.

CONCLUSION

It is necessary to enhance the predictive effect of the model by eliminating confounding factors through data pre-processing. The model developed in this way is effective in predicting the efficacy of NAC for breast cancer.

Prognostic Value of [18F]FDG PET Radiomics to Detect Peritoneal and Distant Metastases in Locally Advanced Gastric Cancer-A Side Study of the Prospective Multicentre PLASTIC Study

AIM

To improve identification of peritoneal and distant metastases in locally advanced gastric cancer using [¹⁸F]FDG-PET radiomics.

METHODS

[¹⁸F]FDG-PET scans of 206 patients acquired in 16 different Dutch hospitals in the prospective multicentre PLASTIC-study were analysed. Tumours were delineated and 105 radiomic features were extracted. Three classification models were developed to identify peritoneal and distant metastases (incidence: 21%): a model with clinical variables, a model with radiomic features, and a clinicoradiomic model, combining clinical variables and radiomic features. A least absolute shrinkage and selection operator (LASSO) regression classifier was trained and evaluated in a 100-times repeated random split, stratified for the presence of peritoneal and distant metastases. To exclude features with high mutual correlations, redundancy filtering of the Pearson correlation matrix was performed (r = 0.9). Model performances were expressed by the area under the receiver operating characteristic curve (AUC). In addition, subgroup analyses based on Lauren classification were performed.

RESULTS

None of the models could identify metastases with low AUCs of 0.59, 0.51, and 0.56, for the clinical, radiomic, and clinicoradiomic model, respectively. Subgroup analysis of intestinal and mixed-type tumours resulted in low AUCs of 0.67 and 0.60 for the clinical and radiomic models, and a moderate AUC of 0.71 in the clinicoradiomic model. Subgroup analysis of diffuse-type tumours did not improve the classification performance.

CONCLUSION

Overall, [¹⁸F]FDG-PET-based radiomics did not contribute to the preoperative identification of peritoneal and distant metastases in patients with locally advanced gastric carcinoma. In intestinal and mixed-type tumours, the classification performance of the clinical model slightly improved with the addition of radiomic features, but this slight improvement does not outweigh the laborious radiomic analysis.

Artificial intelligence-driven radiomics study in cancer: the role of feature engineering and modeling

Modern medicine is reliant on various medical imaging technologies for non-invasively observing patients' anatomy. However, the interpretation of medical images can be highly subjective and dependent on the expertise of clinicians. Moreover, some potentially useful quantitative information in medical images, especially that which is not visible to the naked eye, is often ignored during clinical practice. In contrast, radiomics performs high-throughput feature extraction from medical images, which enables quantitative analysis of medical images and prediction of various clinical endpoints. Studies have reported that radiomics exhibits promising performance in diagnosis and predicting treatment responses and prognosis, demonstrating its potential to be a non-invasive auxiliary tool for personalized medicine. However, radiomics remains in a developmental phase as numerous technical challenges have yet to be solved, especially in feature engineering and statistical modeling. In this review, we introduce the current utility of radiomics by summarizing research on its application in the diagnosis, prognosis, and prediction of treatment responses in patients with cancer. We focus on machine learning approaches, for feature extraction and selection during feature engineering and for imbalanced datasets and multi-modality fusion during statistical modeling. Furthermore, we introduce the stability, reproducibility, and interpretability of features, and the generalizability and interpretability of models. Finally, we offer possible solutions to current challenges in radiomics research.

Development and External Validation of a PET Radiomic Model for Prognostication of Head and Neck Cancer

AIM

To build and externally validate an [¹⁸F]FDG PET radiomic model to predict overall survival in patients with head and neck squamous cell carcinoma (HNSCC).

METHODS

Two multicentre datasets of patients with operable HNSCC treated with preoperative afatinib who underwent a baseline and evaluation [¹⁸F]FDG PET/CT scan were included (EORTC: n = 20, Unicancer: n = 34). Tumours were delineated, and radiomic features were extracted. Each cohort served once as a training and once as an external validation set for the prediction of overall survival. Supervised feature selection was performed using variable hunting with variable importance, selecting the top two features. A Cox proportional hazards regression model using selected radiomic features and clinical characteristics was fitted on the training dataset and validated in the external validation set. Model performances are expressed by the concordance index (C-index).

RESULTS

In both models, the radiomic model surpassed the clinical model with validation C-indices of 0.69 and 0.79 vs. 0.60 and 0.67, respectively. The model that combined the radiomic features and clinical variables performed best, with validation C-indices of 0.71 and 0.82.

CONCLUSION

Although assessed in two small but independent cohorts, an [¹⁸F]FDG-PET radiomic signature based on the evaluation scan seems promising for the prediction of overall survival for HNSSC treated with preoperative afatinib. The robustness and clinical applicability of this radiomic signature should be assessed in a larger cohort.

Repeatability of radiomics studies in colorectal cancer: a systematic review

BACKGROUND

Recently, radiomics has been widely used in colorectal cancer, but many variable factors affect the repeatability of radiomics research. This review aims to analyze the repeatability of radiomics studies in colorectal cancer and to evaluate the current status of radiomics in the field of colorectal cancer.

METHODS

The included studies in this review by searching from the PubMed and Embase databases. Then each study in our review was evaluated using the Radiomics Quality Score (RQS). We analyzed the factors that may affect the repeatability in the radiomics workflow and discussed the repeatability of the included studies.

RESULTS

A total of 188 studies was included in this review, of which only two (2/188, 1.06%) studies controlled the influence of individual factors. In addition, the median score of RQS was 11 (out of 36), range-1 to 27.

CONCLUSIONS

The RQS score was moderately low, and most studies did not consider the repeatability of radiomics features, especially in terms of Intra-individual, scanners, and scanning parameters. To improve the generalization of the radiomics model, it is necessary to further control the variable factors of repeatability.

The Usefulness of Machine Learning-Based Evaluation of Clinical and Pretreatment [18F]-FDG-PET/CT Radiomic Features for Predicting Prognosis in Hypopharyngeal Cancer

PURPOSE

To examine whether the machine learning (ML) analyses using clinical and pretreatment 2-deoxy-2-[¹⁸F]fluoro-D-glucose positron emission tomography ([¹⁸F]-FDG-PET)-based radiomic features were useful for predicting prognosis in patients with hypopharyngeal cancer.

PROCEDURES

This retrospective study included 100 patients with hypopharyngeal cancer who underwent [¹⁸F]-FDG-PET/X-ray computed tomography (CT) before treatment, and these patients were allocated to the training (n=80) and validation (n=20) cohorts. Eight clinical (age, sex, histology, T stage, N stage, M stage, UICC stage, and treatment) and 40 [¹⁸F]-FDG-PET-based radiomic features were used to predict disease progression. A feature reduction procedure based on the decrease of the Gini impurity was applied. Six ML algorithms (random forest, neural network, k-nearest neighbors, naïve Bayes, logistic regression, and support vector machine) were compared using the area under the receiver operating characteristic curve (AUC). Progression-free survival (PFS) was assessed using Cox regression analysis.

RESULTS

The five most important features for predicting disease progression were UICC stage, N stage, gray level co-occurrence matrix entropy (GLCM_Entropy), gray level run length matrix run length non-uniformity (GLRLM_RLNU), and T stage. Patients who experienced disease progression displayed significantly higher UICC stage, N stage, GLCM_Entropy, GLRLM_RLNU, and T stage than those without progression (each, p<0.001). In both cohorts, the logistic regression model constructed by these 5 features was the best performing classifier (training: AUC=0.860, accuracy=0.800; validation: AUC=0.803, accuracy=0.700). In the logistic regression model, 5-year PFS was significantly higher in patients with predicted non-progression than those with predicted progression (75.8% vs. 8.3%, p<0.001), and this model was only the independent factor for PFS in multivariate analysis (hazard ratio = 3.22; 95% confidence interval = 1.03-10.11; p=0.045).

CONCLUSIONS

The logistic regression model constructed by UICC, T and N stages and pretreatment [¹⁸F]-FDG-PET-based radiomic features, GLCM_Entropy, and GLRLM_RLNU may be the most important predictor of prognosis in patients with hypopharyngeal cancer.

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

OBJECTIVES

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

METHODS

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

RESULTS

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

CONCLUSION

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

KEY POINTS

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

Differentiating high-grade glioma progression from treatment-related changes with dynamic [18F]FDOPA PET: a multicentric study

OBJECTIVES

Diagnostic accuracy of amino-acid PET for distinguishing progression from treatment-related changes (TRC) is currently based on single-center non-homogeneous glioma populations. Our study assesses the diagnostic value of static and dynamic [¹⁸F]FDOPA PET acquisitions to differentiate between high-grade glioma (HGG) recurrence and TRC in a large cohort sourced from two independent nuclear medicine centers.

METHODS

We retrospectively identified 106 patients with suspected glioma recurrences (WHO GIII, n = 38; GIV, n = 68; IDH-mutant, n = 35, IDH-wildtype, n = 71). Patients underwent dynamic [¹⁸F]FDOPA PET/CT (n = 83) or PET/MRI (n = 23), and static tumor-to-background ratios (TBRs), metabolic tumor volumes and dynamic parameters (time to peak and slope) were determined. The final diagnosis was either defined by histopathology or a clinical-radiological follow-up at 6 months. Optimal [¹⁸F]FDOPA PET parameter cut-offs were obtained by receiver operating characteristic analysis. Predictive factors and clinical parameters were assessed using univariate and multivariate Cox regression survival analyses.

RESULTS

Surgery or the clinical-radiological 6-month follow-up identified 71 progressions and 35 treatment-related changes. TBR_mean, with a threshold of 1.8, best-differentiated glioma recurrence/progression from post-treatment changes in the whole population (sensitivity 82%, specificity 71%, p < 0.0001) whereas curve slope was only significantly different in IDH-mutant HGGs (n = 25). In survival analyses, MTV was a clinical independent predictor of progression-free and overall survival on the multivariate analysis (p ≤ 0.01). A curve slope > -0.12/h was an independent predictor for longer PFS in IDH-mutant HGGs CONCLUSION: Our multicentric study confirms the high accuracy of [¹⁸F]FDOPA PET to differentiate recurrent malignant gliomas from TRC and emphasizes the diagnostic and prognostic value of dynamic acquisitions for IDH-mutant HGGs.

KEY POINTS

• The diagnostic accuracy of dynamic amino-acid PET, for distinguishing progression from treatment-related changes, is currently based on single-center non-homogeneous glioma populations. • This multicentric study confirms the high accuracy of static [¹⁸F]FDOPA PET images for differentiating progression from treatment-related changes in a homogeneous population of high-grade gliomas and highlights the diagnostic and prognostic value of dynamic acquisitions for IDH-mutant high-grade gliomas. • Dynamic acquisitions should be performed in IDH-mutant glioma patients to provide valuable information for the differential diagnosis of recurrence and treatment-related changes.

Four-dimensional quantitative analysis using FDG-PET in clinical oncology

Positron emission tomography (PET) with F-18 fluorodeoxyglucose (FDG) has been commonly used in many oncological areas. High-resolution PET permits a three-dimensional analysis of FDG distributions on various lesions in vivo, which can be applied for tissue characterization, risk analysis, and treatment monitoring after chemoradiotherapy and immunotherapy. Metabolic changes can be assessed using the tumor absolute FDG uptake as standardized uptake value (SUV) and metabolic tumor volume (MTV). In addition, tumor heterogeneity assessment can potentially estimate tumor aggressiveness and resistance to chemoradiotherapy. Attempts have been made to quantify intratumoral heterogeneity using radiomics. Recent reports have indicated the clinical feasibility of a dynamic FDG PET-computed tomography (CT) in pilot cohort studies of oncological cases. Dynamic imaging permits the assessment of temporal changes in FDG uptake after administration, which is particularly useful for differentiating pathological from physiological uptakes with high diagnostic accuracy. In addition, several new parameters have been introduced for the in vivo quantitative analysis of FDG metabolic processes. Thus, a four-dimensional FDG PET-CT is available for precise tissue characterization of various lesions. This review introduces various new techniques for the quantitative analysis of FDG distribution and glucose metabolism using a four-dimensional FDG analysis with PET-CT. This elegant study reveals the important role of tissue characterization and treatment strategies in oncology.

Prognostic analysis of curatively resected pancreatic cancer using harmonized positron emission tomography radiomic features

BACKGROUND

Texture features reflecting tumour heterogeneity enable us to investigate prognostic factors. The R package ComBat can harmonize the quantitative texture features among several positron emission tomography (PET) scanners. We aimed to identify prognostic factors among harmonized PET radiomic features and clinical information from pancreatic cancer patients who underwent curative surgery.

METHODS

Fifty-eight patients underwent preoperative enhanced dynamic computed tomography (CT) scanning and fluorodeoxyglucose PET/CT using four PET scanners. Using LIFEx software, we measured PET radiomic parameters including texture features with higher order and harmonized these PET parameters. For progression-free survival (PFS) and overall survival (OS), we evaluated clinical information, including age, TNM stage, and neural invasion, and the harmonized PET radiomic features based on univariate Cox proportional hazard regression. Next, we analysed the prognostic indices by multivariate Cox proportional hazard regression (1) by using either significant (p < 0.05) or borderline significant (p = 0.05-0.10) indices in the univariate analysis (first multivariate analysis) or (2) by using the selected features with random forest algorithms (second multivariate analysis). Finally, we checked these multivariate results by log-rank test.

RESULTS

Regarding the first multivariate analysis for PFS after univariate analysis, age was the significant prognostic factor (p = 0.020), and MTV and GLCM contrast were borderline significant (p = 0.051 and 0.075, respectively). Regarding the first multivariate analysis of OS, neural invasion, Shape sphericity and GLZLM LZLGE were significant (p = 0.019, 0.042 and 0.0076). In the second multivariate analysis, only MTV was significant (p = 0.046) for PFS, whereas GLZLM LZLGE was significant (p = 0.047), and Shape sphericity was borderline significant (p = 0.088) for OS. In the log-rank test, age, MTV and GLCM contrast were borderline significant for PFS (p = 0.08, 0.06 and 0.07, respectively), whereas neural invasion and Shape sphericity were significant (p = 0.03 and 0.04, respectively), and GLZLM LZLGE was borderline significant for OS (p = 0.08).

CONCLUSIONS

Other than the clinical factors, MTV and GLCM contrast for PFS and Shape sphericity and GLZLM LZLGE for OS may be prognostic PET parameters. A prospective multicentre study with a larger sample size may be warranted.

Federated Transfer Learning for Low-dose PET Denoising: A Pilot Study with Simulated Heterogeneous Data

Positron emission tomography (PET) with a reduced injection dose, i.e., low-dose PET, is an efficient way to reduce radiation dose. However, low-dose PET reconstruction suffers from a low signal-to-noise ratio (SNR), affecting diagnosis and other PET-related applications. Recently, deep learning-based PET denoising methods have demonstrated superior performance in generating high-quality reconstruction. However, these methods require a large amount of representative data for training, which can be difficult to collect and share due to medical data privacy regulations. Moreover, low-dose PET data at different institutions may use different low-dose protocols, leading to non-identical data distribution. While previous federated learning (FL) algorithms enable multi-institution collaborative training without the need of aggregating local data, it is challenging for previous methods to address the large domain shift caused by different low-dose PET settings, and the application of FL to PET is still under-explored. In this work, we propose a federated transfer learning (FTL) framework for low-dose PET denoising using heterogeneous low-dose data. Our experimental results on simulated multi-institutional data demonstrate that our method can efficiently utilize heterogeneous low-dose data without compromising data privacy for achieving superior low-dose PET denoising performance for different institutions with different low-dose settings, as compared to previous FL methods.

Machine learning for normal tissue complication probability prediction: Predictive power with versatility and easy implementation

Background and purpose

A popular Normal tissue Complication (NTCP) model deployed to predict radiotherapy (RT) toxicity is the Lyman-Burman Kutcher (LKB) model of tissue complication. Despite the LKB model's popularity, it can suffer from numerical instability and considers only the generalized mean dose (GMD) to an organ. Machine learning (ML) algorithms can potentially offer superior predictive power of the LKB model, and with fewer drawbacks. Here we examine the numerical characteristics and predictive power of the LKB model and compare these with those of ML.

Materials and methods

Both an LKB model and ML models were used to predict G2 Xerostomia on patients following RT for head and neck cancer, using the dose volume histogram of parotid glands as the input feature. Model speed, convergence characteristics and predictive power was evaluated on an independent training set.

Results

We found that only global optimization algorithms could guarantee a convergent and predictive LKB model. At the same time our results showed that ML models remained unconditionally convergent and predictive, while staying robust to gradient descent optimization. ML models outperform LKB in Brier score and accuracy but compare to LKB in ROC-AUC.

Conclusion

We have demonstrated that ML models can quantify NTCP better than or as well as LKB models, even for a toxicity that the LKB model is particularly well suited to predict. ML models can offer this performance while offering fundamental advantages in model convergence, speed, and flexibility, and so could offer an alternative to the LKB model that could potentially be used in clinical RT planning decisions.

Progress and current utility of radiomics in PET/CT study of non-metastatic breast cancer: A systematic review

AIM

To synthesize the current evidence of the usefulness of radiomics in PET/CT image analysis in local and locally advanced breast cancer. Also, to evaluate the methodological quality of the radiomic studies published.

METHODS

Systematic review of articles in different databases until 2021 using the terms "PET", "radiomics", "texture", "breast". Only articles with human data and that included a PET image were included. Studies with simulated data and with less than 20 patients were excluded. Were extracted sample size, radiotracer used, imaging technique, and radiomics characteristics from each article. The methodological quality of the studies was determined using the QUADAS-2 tool.

RESULTS

18 articles were selected. The retrospective design was the most used. The most studied radiomic characteristic was SUVmax. Several radiomic parameters were correlated with tumor characterization, and tumor heterogeneity proved useful for predicting disease course and response to treatment. Most articles showed a high risk of bias, mainly from the patient selection.

CONCLUSIONS

A high probability of bias was observed in most of the published articles. Radiomics is a developing field and more studies are needed to demonstrate its usefulness in routine clinical practice. The QUADAS-2 tool allows critical assessment of the methodological quality of the available evidence. Despite its limitations, radiomics is shown to be an instrument that can help to achieve personalized oncologic management of breast cancer.

An Automated Method for Artifical Intelligence Assisted Diagnosis of Active Aortitis Using Radiomic Analysis of FDG PET-CT Images

The aim of this study was to develop and validate an automated pipeline that could assist the diagnosis of active aortitis using radiomic imaging biomarkers derived from [18F]-Fluorodeoxyglucose Positron Emission Tomography-Computed Tomography (FDG PET-CT) images. The aorta was automatically segmented by convolutional neural network (CNN) on FDG PET-CT of aortitis and control patients. The FDG PET-CT dataset was split into training (43 aortitis:21 control), test (12 aortitis:5 control) and validation (24 aortitis:14 control) cohorts. Radiomic features (RF), including SUV metrics, were extracted from the segmented data and harmonized. Three radiomic fingerprints were constructed: A-RFs with high diagnostic utility removing highly correlated RFs; B used principal component analysis (PCA); C-Random Forest intrinsic feature selection. The diagnostic utility was evaluated with accuracy and area under the receiver operating characteristic curve (AUC). Several RFs and Fingerprints had high AUC values (AUC > 0.8), confirmed by balanced accuracy, across training, test and external validation datasets. Good diagnostic performance achieved across several multi-centre datasets suggests that a radiomic pipeline can be generalizable. These findings could be used to build an automated clinical decision tool to facilitate objective and standardized assessment regardless of observer experience.

Criteria for the translation of radiomics into clinically useful tests

Computer-extracted tumour characteristics have been incorporated into medical imaging computer-aided diagnosis (CAD) algorithms for decades. With the advent of radiomics, an extension of CAD involving high-throughput computer-extracted quantitative characterization of healthy or pathological structures and processes as captured by medical imaging, interest in such computer-extracted measurements has increased substantially. However, despite the thousands of radiomic studies, the number of settings in which radiomics has been successfully translated into a clinically useful tool or has obtained FDA clearance is comparatively small. This relative dearth might be attributable to factors such as the varying imaging and radiomic feature extraction protocols used from study to study, the numerous potential pitfalls in the analysis of radiomic data, and the lack of studies showing that acting upon a radiomic-based tool leads to a favourable benefit-risk balance for the patient. Several guidelines on specific aspects of radiomic data acquisition and analysis are already available, although a similar roadmap for the overall process of translating radiomics into tools that can be used in clinical care is needed. Herein, we provide 16 criteria for the effective execution of this process in the hopes that they will guide the development of more clinically useful radiomic tests in the future.

Multiparametric Quantitative Imaging in Risk Prediction: Recommendations for Data Acquisition, Technical Performance Assessment, and Model Development and Validation

Combinations of multiple quantitative imaging biomarkers (QIBs) are often able to predict the likelihood of an event of interest such as death or disease recurrence more effectively than single imaging measurements can alone. The development of such multiparametric quantitative imaging and evaluation of its fitness of use differs from the analogous processes for individual QIBs in several key aspects. A computational procedure to combine the QIB values into a model output must be specified. The output must also be reproducible and be shown to have reasonably strong ability to predict the risk of an event of interest. Attention must be paid to statistical issues not often encountered in the single QIB scenario, including overfitting and bias in the estimates of model performance. This is the fourth in a five-part series on statistical methodology for assessing the technical performance of multiparametric quantitative imaging. Considerations for data acquisition are discussed and recommendations from the literature on methodology to construct and evaluate QIB-based models for risk prediction are summarized. The findings in the literature upon which these recommendations are based are demonstrated through simulation studies. The concepts in this manuscript are applied to a real-life example involving prediction of major adverse cardiac events using automated plaque analysis.

Repeatability and reproducibility of MRI-radiomic features: A phantom experiment on a 1.5 T scanner

PURPOSE

Aim of this study is to assess the repeatability of radiomic features on magnetic resonance images (MRI) and their stability to variations in time of repetition (TR), time of echo (TE), slice thickness (ST), and pixel spacing (PS) using vegetable phantoms.

METHODS

The organic phantom was realized using two cucumbers placed inside a cylindrical container, and the analysis was performed using T1-weighted (T1w), T2-weighted (T2w), and diffusion-weighted images. One dataset was used to test the repeatability of the radiomic features, whereas other four datasets were used to test the sensitivity of the different MRI sequences to image acquisition parameters (TR, TE, ST, and PS). Four regions of interest (ROIs) were segmented: two for the central part of each cucumber and two for the external parts. Radiomic features were extracted from each ROI using Pyradiomics. To assess the effect of preprocessing on the reduction of variability, features were extracted both before and after the preprocessing. The coefficient of variation (CV) and intra-class correlation coefficient (ICC) were used to evaluate variability.

RESULTS

The use of intensity standardization increased the stability for the first-order statistics features. Shape and size features were always stable for all the analyses. Textural features were particularly sensitive to changes in ST and PS, although some increase in stability could be obtained by voxel size resampling. When images underwent image preprocessing, the number of stable features (ICC > 0.75 and mean absolute CV < 0.3) was 33 for apparent diffusion coefficient (ADC), 52 for T1w, and 73 for T2w.

CONCLUSIONS

The most critical source of variability is related to changes in voxel size (either caused by changes in ST or PS). Preprocessing increases features stability to both test-retest and variation of the image acquisition parameters for all the types of analyzed MRI (T1w, T2w, and ADC), except for ST.

Feasibility of radiomic feature harmonization for pooling of [18F]FET or [18F]GE-180 PET images of gliomas

INTRODUCTION

Large datasets are required to ensure reliable non-invasive glioma assessment with radiomics-based machine learning methods. This can often only be achieved by pooling images from different centers. Moreover, trained models should perform with high accuracy when applied to data from different centers. In this study, the impact of reconstruction settings and segmentation methods on radiomic features derived from amino acid and TSPO PET images of glioma patients was examined. Additionally, the ability to model and thus reduce feature differences was investigated.

METHODS

[¹⁸F]FET and [¹⁸F]GE-180 PET data were acquired from 19 glioma patients. For each acquisition, 10 reconstruction settings and 9 segmentation methods were included to emulate multicentric data. Statistical robustness measures were calculated before and after ComBat harmonization. Differences between features due to setting variations were assessed using Friedman test, coefficient of variation (CV) and inter-rater reliability measures, including intraclass and Spearman's rank correlation coefficients and Fleiss' Kappa.

RESULTS

According to Friedman analyses, most features (>60%) showed significant differences. Yet, CV and inter-rater reliability measures indicated higher robustness. ComBat resulted in almost complete harmonization (>87%) according to Friedman test and little to no improvement according to CV and inter-rater reliability measures. [¹⁸F]GE-180 features were more sensitive to reconstruction settings than [¹⁸F]FET features.

CONCLUSIONS

According to Friedman test, feature distributions could be successfully aligned using ComBat. However, depending on settings, changes in patient ranks were observed for some features and could not be eliminated by harmonization. Thus, for clinical utilization it is recommended to exclude affected features.

Radiomics-Based Inter-Lesion Relation Network to Describe [18F]FMCH PET/CT Imaging Phenotypes in Prostate Cancer

Advanced image analysis, including radiomics, has recently acquired recognition as a source of biomarkers, although there are some technical and methodological challenges to face for its application in the clinic. Among others, proper phenotyping of metastatic or systemic disease where multiple lesions coexist is an issue, since each lesion contributes to characterization of the disease. Therefore, the radiomic profile of each lesion should be modeled into a more complex architecture able to reproduce each "unit" (lesion) as a part of the "entire" (patient). This work aimed to characterize intra-tumor heterogeneity underpinning metastatic prostate cancer using an exhaustive innovative approach which consist of a i) feature transformation method to build an agnostic (i.e., irrespective of pre-existence knowledge, experience, and expertise) radiomic profile of lesions extracted from [¹⁸F]FMCH PET/CT, ii) qualitative assessment of intra-tumor heterogeneity of patients, iii) quantitative representation of the intra-tumor heterogeneity of patients in terms of the relationship between their lesions' profiles, to be associated with prognostic factors. We confirmed that metastatic prostate cancer patients encompassed lesions with different radiomic profiles that exhibited intra-tumor radiomic heterogeneity and that the presence of many radiomic profiles within the same patient impacted the outcome.

Radiomics prognostic analysis of PET/CT images in a multicenter head and neck cancer cohort: investigating ComBat strategies, sub-volume characterization, and automatic segmentation

PURPOSE

This study aimed to investigate the impact of several ComBat harmonization strategies, intra-tumoral sub-volume characterization, and automatic segmentations for progression-free survival (PFS) prediction through radiomics modeling for patients with head and neck cancer (HNC) in PET/CT images.

METHODS

The HECKTOR MICCAI 2021 challenge set containing PET/CT images and clinical data of 325 oropharynx HNC patients was exploited. A total of 346 IBSI-compliant radiomic features were extracted for each patient's primary tumor volume defined by the reference manual contours. Modeling relied on least absolute shrinkage Cox regression (Lasso-Cox) for feature selection (FS) and Cox proportional-hazards (CoxPH) models were built to predict PFS. Within this methodological framework, 8 different strategies for ComBat harmonization were compared, including before or after FS, in feature groups separately or all features directly, and with center or clustering-determined labels. Features extracted from tumor sub-volume clustering were also investigated for their prognostic additional value. Finally, 3 automatic segmentations (2 threshold-based and a 3D U-Net) were also compared. All results were evaluated with the concordance index (C-index).

RESULTS

Radiomics features without harmonization, combined with clinical factors, led to models with C-index values of 0.69 in the testing set. The best version of ComBat harmonization, i.e., after FS, for feature groups separately and relying on clustering-determined labels, achieved a C-index of 0.71. The use of features extracted from tumor sub-volumes further improved the C-index to 0.72. Models that relied on the automatic segmentations yielded close but slightly lower prognostic performance (0.67-0.70) compared to reference contours.

CONCLUSION

A standard radiomics pipeline allowed for prediction of PFS in a multicenter HNC cohort. Applying a specific strategy of ComBat harmonization improved the performance. The extraction of intra-tumoral sub-volume features and automatic segmentation could contribute to the improvement and automation of prognosis modeling, respectively.

Current Role of Delta Radiomics in Head and Neck Oncology

The latest developments in the management of head and neck cancer show an increasing trend in the implementation of novel approaches using artificial intelligence for better patient stratification and treatment-related risk evaluation. Radiomics, or the extraction of data from various imaging modalities, is a tool often used to evaluate specific features related to the tumour or normal tissue that are not identifiable by the naked eye and which can add value to existing clinical data. Furthermore, the assessment of feature variations from one time point to another based on subsequent images, known as delta radiomics, was shown to have even higher value for treatment-outcome prediction or patient stratification into risk categories. The information gathered from delta radiomics can, further, be used for decision making regarding treatment adaptation or other interventions found to be beneficial to the patient. The aim of this work is to collate the existing studies on delta radiomics in head and neck cancer and evaluate its role in tumour response and normal-tissue toxicity predictions alike. Moreover, this work also highlights the role of holomics, which brings under the same umbrella clinical and radiomic features, for a more complex patient characterization and treatment optimisation.

Assessing the generalisability of radiomics features previously identified as predictive of radiation-induced sticky saliva and xerostomia

Background and purpose

While core to the scientific approach, reproducibility of experimental results is challenging in radiomics studies. A recent publication identified radiomics features that are predictive of late irradiation-induced toxicity in head and neck cancer (HNC) patients. In this study, we assessed the generalisability of these findings.

Materials and Methods

The procedure described in the publication in question was applied to a cohort of 109 HNC patients treated with 50-70 Gy in 20-35 fractions using helical radiotherapy although there were inherent differences between the two patient populations and methodologies. On each slice of the planning CT with delineated parotid and submandibular glands, the imaging features that were previously identified as predictive of moderate-to-severe xerostomia and sticky saliva 12 months post radiotherapy (Xer12m and SS12m) were calculated. Specifically, Short Run Emphasis (SRE) and maximum CT intensity (maxHU) were evaluated for improvement in prediction of Xer12m and SS12m respectively, compared to models solely using baseline toxicity and mean dose to the salivary glands.

Results

None of the associations previously identified as statistically significant and involving radiomics features in univariate or multivariate models could be reproduced on our cohort.

Conclusion

The discrepancies observed between the results of the two studies delineate limits to the generalisability of the previously reported findings. This may be explained by the differences in the approaches, in particular the imaging characteristics and subsequent methodological implementation. This highlights the importance of external validation, high quality reporting guidelines and standardisation protocols to ensure generalisability, replication and ultimately clinical implementation.

A review of harmonization strategies for quantitative PET

PET can reveal in vivo biological processes at the molecular level. PET-derived quantitative values have been used as a surrogate marker for clinical decision-making in numerous clinical studies and trials. However, quantitative values in PET are variable depending on technical, biological, and physical factors. The variability may have a significant impact on a study outcome. Appropriate scanner calibration and quality control, standardization of imaging protocols, and any necessary harmonization strategies are essential to make use of PET as a biomarker with low bias and variability. This review summarizes benefits, limitations, and remaining challenges for harmonization of quantitative PET, including whole-body PET in oncology, brain PET in neurology, PET/MR, and non-¹⁸F PET imaging. This review is expected to facilitate harmonization of quantitative PET and to promote the contribution of PET-derived biomarkers to research and development in medicine.

The role of artificial intelligence based on PET/CT radiomics in NSCLC: Disease management, opportunities, and challenges

Objectives

Lung cancer has been widely characterized through radiomics and artificial intelligence (AI). This review aims to summarize the published studies of AI based on positron emission tomography/computed tomography (PET/CT) radiomics in non-small-cell lung cancer (NSCLC).

Materials and methods

A comprehensive search of literature published between 2012 and 2022 was conducted on the PubMed database. There were no language or publication status restrictions on the search. About 127 articles in the search results were screened and gradually excluded according to the exclusion criteria. Finally, this review included 39 articles for analysis.

Results

Classification is conducted according to purposes and several studies were identified at each stage of disease:1) Cancer detection (n=8), 2) histology and stage of cancer (n=11), 3) metastases (n=6), 4) genotype (n=6), 5) treatment outcome and survival (n=8). There is a wide range of heterogeneity among studies due to differences in patient sources, evaluation criteria and workflow of radiomics. On the whole, most models show diagnostic performance comparable to or even better than experts, and the common problems are repeatability and clinical transformability.

Conclusion

AI-based PET/CT Radiomics play potential roles in NSCLC clinical management. However, there is still a long way to go before being translated into clinical application. Large-scale, multi-center, prospective research is the direction of future efforts, while we need to face the risk of repeatability of radiomics features and the limitation of access to large databases.

Joint EANM/SNMMI guideline on radiomics in nuclear medicine : Jointly supported by the EANM Physics Committee and the SNMMI Physics, Instrumentation and Data Sciences Council

PURPOSE

The purpose of this guideline is to provide comprehensive information on best practices for robust radiomics analyses for both hand-crafted and deep learning-based approaches.

METHODS

In a cooperative effort between the EANM and SNMMI, we agreed upon current best practices and recommendations for relevant aspects of radiomics analyses, including study design, quality assurance, data collection, impact of acquisition and reconstruction, detection and segmentation, feature standardization and implementation, as well as appropriate modelling schemes, model evaluation, and interpretation. We also offer an outlook for future perspectives.

CONCLUSION

Radiomics is a very quickly evolving field of research. The present guideline focused on established findings as well as recommendations based on the state of the art. Though this guideline recognizes both hand-crafted and deep learning-based radiomics approaches, it primarily focuses on the former as this field is more mature. This guideline will be updated once more studies and results have contributed to improved consensus regarding the application of deep learning methods for radiomics. Although methodological recommendations in the present document are valid for most medical image modalities, we focus here on nuclear medicine, and specific recommendations when necessary are made for PET/CT, PET/MR, and quantitative SPECT.

Predictive and prognostic potential of pretreatment 68Ga-PSMA PET tumor heterogeneity index in patients with metastatic castration-resistant prostate cancer treated with 177Lu-PSMA

INTRODUCTION

This study was conducted to evaluate the predictive values of volumetric parameters and radiomic features (RFs) extracted from pretreatment 68Ga-PSMA PET and baseline clinical parameters in response to 177Lu-PSMA therapy.

MATERIALS AND METHODS

In this retrospective multicenter study, mCRPC patients undergoing 177Lu-PSMA therapy were enrolled. According to the outcome of therapy, the patients were classified into two groups including positive biochemical response (BCR) (≥ 50% reduction in the serum PSA value) and negative BCR (< 50%). Sixty-five RFs, eight volumetric parameters, and also seventeen clinical parameters were evaluated for the prediction of BCR. In addition, the impact of such parameters on overall survival (OS) was evaluated.

RESULTS

33 prostate cancer patients with a median age of 69 years (range: 49-89) were enrolled. BCR was observed in 22 cases (66%), and 16 cases (48.5%) died during the follow-up time. The results of Spearman correlation test indicated a significant relationship between BCR and treatment cycle, administered dose, HISTO energy, GLCM entropy, and GLZLM LZLGE (p<0.05). In addition, according to the Mann-Whitney U test, age, cycle, dose, GLCM entropy, and GLZLM LZLGE were significantly different between BCR and non BCR patients (p<0.05). According to the ROC curve analysis for feature selection for prediction of BCR, GLCM entropy, age, treatment cycle, and administered dose showed acceptable results (p<0.05). According to SVM for assessing the best model for prediction of response to therapy, GLCM entropy alone showed the highest predictive performance in treatment planning. For the entire cohort, the Kaplan-Meier test revealed a median OS of 21 months (95% CI: 12.12-29.88). The median OS was estimated at 26 months (95% CI: 17.43-34.56) for BCR patients and 13 months (95% CI: 9.18-16.81) for non BCR patients. Among all variables included in the Kaplan Meier, the only response to therapy was statistically significant (p=0.01).

CONCLUSION

This exploratory study showed that the heterogeneity parameter of pretreatment 68Ga-PSMA PET images might be a potential predictive value for response to 177Lu-PSMA therapy in mCRPC; however, further prospective studies need to be carried out to verify these findings.

Validation of cross-sectional and longitudinal ComBat harmonization methods for magnetic resonance imaging data on a travelling subject cohort

Background

The growth in multi-center neuroimaging studies generated a need for methods that mitigate the differences in hardware and acquisition protocols across sites i.e., scanner effects. ComBat harmonization methods have shown promise but have not yet been tested on all the data types commonly studied with magnetic resonance imaging (MRI). This study aimed to validate neuroCombat, longCombat and gamCombat on both structural and diffusion metrics in both cross-sectional and longitudinal data.

Methods

We used a travelling subject design whereby 73 healthy volunteers contributed 161 scans across two sites and four machines using one T1 and five diffusion MRI protocols. Scanner was defined as a composite of site, machine and protocol. A common pipeline extracted two structural metrics (volumes and cortical thickness) and two diffusion tensor imaging metrics (mean diffusivity and fractional anisotropy) for seven regions of interest including gray and (except for cortical thickness) white matter regions.

Results

Structural data exhibited no significant scanner effect and therefore did not benefit from harmonization in our particular cohort. Indeed, attempting harmonization obscured the true biological effect for some regions of interest. Diffusion data contained marked scanner effects and was successfully harmonized by all methods, resulting in smaller scanner effects and better detection of true biological effects. LongCombat less effectively reduced the scanner effect for cross-sectional white matter data but had a slightly lower probability of incorrectly finding group differences in simulations, compared to neuroCombat and gamCombat. False positive rates for all methods and all metrics did not significantly exceed 5%.

Conclusions

Statistical harmonization of structural data is not always necessary and harmonization in the absence of a scanner effect may be harmful. Harmonization of diffusion MRI data is highly recommended with neuroCombat, longCombat and gamCombat performing well in cross-sectional and longitudinal settings.

A methodological framework for AI-assisted diagnosis of active aortitis using radiomic analysis of FDG PET-CT images: Initial analysis

BACKGROUND

The aim of this study was to explore the feasibility of assisted diagnosis of active (peri-)aortitis using radiomic imaging biomarkers derived from [18F]-Fluorodeoxyglucose Positron Emission Tomography-Computed Tomography (FDG PET-CT) images.

METHODS

The aorta was manually segmented on FDG PET-CT in 50 patients with aortitis and 25 controls. Radiomic features (RF) (n = 107), including SUV (Standardized Uptake Value) metrics, were extracted from the segmented data and harmonized using the ComBat technique. Individual RFs and groups of RFs (i.e., signatures) were used as input in Machine Learning classifiers. The diagnostic utility of these classifiers was evaluated with area under the receiver operating characteristic curve (AUC) and accuracy using the clinical diagnosis as the ground truth.

RESULTS

Several RFs had high accuracy, 84% to 86%, and AUC scores 0.83 to 0.97 when used individually. Radiomic signatures performed similarly, AUC 0.80 to 1.00.

CONCLUSION

A methodological framework for a radiomic-based approach to support diagnosis of aortitis was outlined. Selected RFs, individually or in combination, showed similar performance to the current standard of qualitative assessment in terms of AUC for identifying active aortitis. This framework could support development of a clinical decision-making tool for a more objective and standardized assessment of aortitis.

PET-Derived Radiomics and Artificial Intelligence in Breast Cancer: A Systematic Review

Breast cancer (BC) is a heterogeneous malignancy that still represents the second cause of cancer-related death among women worldwide. Due to the heterogeneity of BC, the correct identification of valuable biomarkers able to predict tumor biology and the best treatment approaches are still far from clear. Although molecular imaging with positron emission tomography/computed tomography (PET/CT) has improved the characterization of BC, these methods are not free from drawbacks. In recent years, radiomics and artificial intelligence (AI) have been playing an important role in the detection of several features normally unseen by the human eye in medical images. The present review provides a summary of the current status of radiomics and AI in different clinical settings of BC. A systematic search of PubMed, Web of Science and Scopus was conducted, including all articles published in English that explored radiomics and AI analyses of PET/CT images in BC. Several studies have demonstrated the potential role of such new features for the staging and prognosis as well as the assessment of biological characteristics. Radiomics and AI features appear to be promising in different clinical settings of BC, although larger prospective trials are needed to confirm and to standardize this evidence.

A Framework for Harmonization of Radiomics Data for Multicenter Studies and Clinical Trials

PURPOSE

Variability in computed tomography images intrinsic to individual scanners limits the application of radiomics in clinical and research settings. The development of reproducible and generalizable radiomics-based models to assess lesions requires harmonization of data. The purpose of this study was to develop, test, and analyze the efficacy of a radiomics data harmonization model.

MATERIALS AND METHODS

Radiomic features from biopsy-proven untreated hepatic metastasis (N = 380) acquired from 167 unique patients with pancreatic, colon, and breast cancers were analyzed. Radiomic features from volume-match 551 samples of normal liver tissue and 188 hepatic cysts were included as references. A novel linear mixed effect model was used to identify effects associated with lesion size, tissue type, and scanner model. Six separate machine learning models were then used to test the effectiveness of radiomic feature harmonization using multivariate analysis.

RESULTS

Proposed model identifies and removes scanner-associated effects while preserving cancer-specific functional dependence of radiomic features on the tumor size. Data harmonization improves the performance of classification models by reducing the scanner-associated variability. For example, the multiclass logistic regression model, LogitBoost, demonstrated the improvement in sensitivity in the range from 15% to 40% for each type of liver metastasis, whereas the overall model accuracy and the kappa coefficient increased by 5% and 8% accordingly.

CONCLUSION

The model removed scanner-associated effects while preserving cancer-specific functional dependence of radiomic features.

How Machine Learning is Powering Neuroimaging to Improve Brain Health

This report presents an overview of how machine learning is rapidly advancing clinical translational imaging in ways that will aid in the early detection, prediction, and treatment of diseases that threaten brain health. Towards this goal, we aresharing the information presented at a symposium, "Neuroimaging Indicators of Brain Structure and Function - Closing the Gap Between Research and Clinical Application", co-hosted by the McCance Center for Brain Health at Mass General Hospital and the MIT HST Neuroimaging Training Program on February 12, 2021. The symposium focused on the potential for machine learning approaches, applied to increasingly large-scale neuroimaging datasets, to transform healthcare delivery and change the trajectory of brain health by addressing brain care earlier in the lifespan. While not exhaustive, this overview uniquely addresses many of the technical challenges from image formation, to analysis and visualization, to synthesis and incorporation into the clinical workflow. Some of the ethical challenges inherent to this work are also explored, as are some of the regulatory requirements for implementation. We seek to educate, motivate, and inspire graduate students, postdoctoral fellows, and early career investigators to contribute to a future where neuroimaging meaningfully contributes to the maintenance of brain health.

FDG PET and CT radiomics in diagnosis and prognosis of non-small-cell lung cancer

BACKGROUND

18F-FDG PET and CT radiomics has been the object of a wide research for over 20 years but its contribution to clinical practice remains not yet well established. We have investigated its impact versus that of only histo-clinical data, for the routine management of non-small-cell lung cancer (NSCLC).

METHODS

Our patients were retrospectively considered. They all had a FDG PET-CT and immuno-histo-chemistry (IHC) to assess PD-L1 expression at the beginning of the disease. A prognosis univariate and multivariate Cox survival analyses was performed for overall survival (OS) and progression free survival (PFS) prediction, including a training/testing procedure. Two sets of 47 PET and 47 CT radiomics features (RFs) were extracted. Difference between RFs according to PD-L1 expression, the histology status and the stage level were tested using suited non parametric statistical tests and the receiver operating characteristics (ROC) curve and the area under curve (AUC).

RESULTS

From 2017 to 2019, 212 NSCLC patients treated in our institution were included. The main conventional prognostic variables were stage and gender with a low added prognostic value in the models including PET and CT RFs. Neither PET nor CT RFs were significant to separate the different levels of PD-L1 expression. Several RFs differ between adenocarcinoma (ADC) and squamous cell carcinoma (SCC) tumours and a large number of PET and CT RFs are significantly linked to patient stage.

CONCLUSIONS

In our population, PET and CT RFs show their intrinsic power to predict survival but do not significantly improve OS and PFS prediction in the different multivariate models, in comparison to conventional data. It would seem necessary to carry out one's own survival analysis before determining a radiomics signature.

Frontiers and hotspots of 18F-FDG PET/CT radiomics: A bibliometric analysis of the published literature

Objective

To illustrate the knowledge hotspots and cutting-edge research trends of ¹⁸F-FDG PET/CT radiomics, the knowledge structure of was systematically explored and the visualization map was analyzed.

Methods

Studies related to ¹⁸F-FDG PET/CT radiomics from 2013 to 2021 were identified and selected from the Web of Science Core Collection (WoSCC) using retrieval formula based on an interview. Bibliometric methods are mainly performed by CiteSpace 5.8.R3, which we use to build knowledge structures including publications, collaborative and co-cited studies, burst analysis, and so on. The performance and relevance of countries, institutions, authors, and journals were measured by knowledge maps. The research foci were analyzed through research of keywords, as well as literature co-citation analysis. Predicting trends of ¹⁸F-FDG PET/CT radiomics in this field utilizes a citation burst detection method.

Results

Through a systematic literature search, 457 articles, which were mainly published in the United States (120 articles) and China (83 articles), were finally included in this study for analysis. Memorial Sloan-Kettering Cancer Center and Southern Medical University are the most productive institutions, both with a frequency of 17. ¹⁸F-FDG PET/CT radiomics–related literature was frequently published with high citation in European Journal of Nuclear Medicine and Molecular Imaging (IF9.236, 2020), Frontiers in Oncology (IF6.244, 2020), and Cancers (IF6.639, 2020). Further cluster profile of keywords and literature revealed that the research hotspots were primarily concentrated in the fields of image, textural feature, and positron emission tomography, and the hot research disease is a malignant tumor. Document co-citation analysis suggested that many scholars have a co-citation relationship in studies related to imaging biomarkers, texture analysis, and immunotherapy simultaneously. Burst detection suggests that adenocarcinoma studies are frontiers in ¹⁸F-FDG PET/CT radiomics, and the landmark literature put emphasis on the reproducibility of ¹⁸F-FDG PET/CT radiomics features.

Conclusion

First, this bibliometric study provides a new perspective on ¹⁸F-FDG PET/CT radiomics research, especially for clinicians and researchers providing scientific quantitative analysis to measure the performance and correlation of countries, institutions, authors, and journals. Above all, there will be a continuing growth in the number of publications and citations in the field of ¹⁸F-FDG PET/CT. Second, the international research frontiers lie in applying ¹⁸F-FDG PET/CT radiomics to oncology research. Furthermore, new insights for researchers in future studies will be adenocarcinoma-related analyses. Moreover, our findings also offer suggestions for scholars to give attention to maintaining the reproducibility of ¹⁸F-FDG PET/CT radiomics features.

Feasibility of using CT radiomic signatures for predicting CD8-T cell infiltration and PD-L1 expression in renal cell carcinoma

Objectives

To identify computed tomography (CT)-based radiomic signatures of cluster of differentiation 8 (CD8)-T cell infiltration and programmed cell death ligand 1 (PD-L1) expression levels in patients with clear-cell renal cell carcinoma (ccRCC).

Methods

Seventy-eight patients with pathologically confirmed localized ccRCC, preoperative multiphase CT and tumor resection specimens were enrolled in this retrospective study. Regions of interest (ROI) of the ccRCC volume were manually segmented from the CT images and processed using a radiomics panel comprising of 1708 metrics. The extracted metrics were used as inputs to three machine learning classifiers: Random Forest, AdaBoost, and ElasticNet to create radiomic signatures for CD8-T cell infiltration and PD-L1 expression, respectively.

Results

Using a cut-off of 80 lymphocytes per high power field, 59 % were classified to CD8 highly infiltrated tumors and 41 % were CD8 non highly infiltrated tumors, respectively. An ElasticNet classifier discriminated between these two groups of CD8-T cells with an AUC of 0.68 (95 % CI, 0.55-0.80). In addition, based on tumor proportion score with a cut-off of > 1 % tumor cells expressing PD-L1, 76 % were PD-L1 positive and 24 % were PD-L1 negative. An Adaboost classifier discriminated between PD-L1 positive and PD-L1 negative tumors with an AUC of 0.8 95 % CI: (0.66, 0.95). 3D radiomics metrics of graylevel co-occurrence matrix (GLCM) and graylevel run-length matrix (GLRLM) metrics drove the performance for CD8-Tcell and PD-L1 classification, respectively.

Conclusions

CT-radiomic signatures can differentiate tumors with high CD8-T cell infiltration with moderate accuracy and positive PD-L1 expression with good accuracy in ccRCC.

Harmonization of Multi-site Cortical Data Across the Human Lifespan

Neuroimaging data harmonization has become a prerequisite in integrative data analytics for standardizing a wide variety of data collected from multiple studies and enabling interdisciplinary research. The lack of standardized image acquisition and computational procedures introduces non-biological variability and inconsistency in multi-site data, complicating downstream statistical analyses. Here, we propose a novel statistical technique to retrospectively harmonize multi-site cortical data collected longitudinally and cross-sectionally between birth and 100 years. We demonstrate that our method can effectively eliminate non-biological disparities from cortical thickness and myelination measurements, while preserving biological variation across the entire lifespan. Our harmonization method will foster large-scale population studies by providing comparable data required for investigating developmental and aging processes.

A role for artificial intelligence in molecular imaging of infection and inflammation

The detection of occult infections and low-grade inflammation in clinical practice remains challenging and much depending on readers’ expertise. Although molecular imaging, like [¹⁸F]FDG PET or radiolabeled leukocyte scintigraphy, offers quantitative and reproducible whole body data on inflammatory responses its interpretation is limited to visual analysis. This often leads to delayed diagnosis and treatment, as well as untapped areas of potential application. Artificial intelligence (AI) offers innovative approaches to mine the wealth of imaging data and has led to disruptive breakthroughs in other medical domains already. Here, we discuss how AI-based tools can improve the detection sensitivity of molecular imaging in infection and inflammation but also how AI might push the data analysis beyond current application toward predicting outcome and long-term risk assessment.

Utility of pre-treatment FDG PET/CT-derived machine learning models for outcome prediction in classical Hodgkin lymphoma

Objectives

Relapse occurs in ~20% of patients with classical Hodgkin lymphoma (cHL) despite treatment adaption based on 2-deoxy-2-[¹⁸F]fluoro-d-glucose positron emission tomography/computed tomography response. The objective was to evaluate pre-treatment FDG PET/CT–derived machine learning (ML) models for predicting outcome in patients with cHL.

Methods

All cHL patients undergoing pre-treatment PET/CT at our institution between 2008 and 2018 were retrospectively identified. A 1.5 × mean liver standardised uptake value (SUV) and a fixed 4.0 SUV threshold were used to segment PET/CT data. Feature extraction was performed using PyRadiomics with ComBat harmonisation. Training (80%) and test (20%) cohorts stratified around 2-year event-free survival (EFS), age, sex, ethnicity and disease stage were defined. Seven ML models were trained and hyperparameters tuned using stratified 5-fold cross-validation. Area under the curve (AUC) from receiver operator characteristic analysis was used to assess performance.

Results

A total of 289 patients (153 males), median age 36 (range 16–88 years), were included. There was no significant difference between training (n = 231) and test cohorts (n = 58) (p value > 0.05). A ridge regression model using a 1.5 × mean liver SUV segmentation had the highest performance, with mean training, validation and test AUCs of 0.82 ± 0.002, 0.79 ± 0.01 and 0.81 ± 0.12. However, there was no significant difference between a logistic model derived from metabolic tumour volume and clinical features or the highest performing radiomic model.

Conclusions

Outcome prediction using pre-treatment FDG PET/CT–derived ML models is feasible in cHL patients. Further work is needed to determine optimum predictive thresholds for clinical use.

Key points

• A fixed threshold segmentation method led to more robust radiomic features.

• A radiomic-based model for predicting 2-year event-free survival in classical Hodgkin lymphoma patients is feasible.

• A predictive model based on ridge regression was the best performing model on our dataset.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00330-022-09039-0.

T2-weighted MRI-based radiomics for discriminating between benign and borderline epithelial ovarian tumors: a multicenter study

Background

Preoperative differentiation between benign and borderline epithelial ovarian tumors (EOTs) is challenging and can significantly impact clinical decision making. The purpose was to investigate whether radiomics based on T2-weighted MRI can discriminate between benign and borderline EOTs preoperatively.

Methods

A total of 417 patients (309, 78, and 30 samples in the training and internal and external validation sets) with pathologically proven benign and borderline EOTs were included in this multicenter study. In total, 1130 radiomics features were extracted from manually delineated tumor volumes of interest on images. The following three different models were constructed and evaluated: radiomics features only (radiomics model); clinical and radiological characteristics only (clinic-radiological model); and a combination of them all (combined model). The diagnostic performances of models were assessed using receiver operating characteristic (ROC) analysis, and area under the ROC curves (AUCs) were compared using the DeLong test.

Results

The best machine learning algorithm to distinguish borderline from benign EOTs was the logistic regression. The combined model achieved the best performance in discriminating between benign and borderline EOTs, with an AUC of 0.86 ± 0.07. The radiomics model showed a moderate AUC of 0.82 ± 0.07, outperforming the clinic-radiological model (AUC of 0.79 ± 0.06). In the external validation set, the combined model performed significantly better than the clinic-radiological model (AUCs of 0.86 vs. 0.63, p = 0.021 [DeLong test]).

Conclusions

Radiomics, based on T2-weighted MRI, can provide critical diagnostic information for discriminating between benign and borderline EOTs, thus having the potential to aid personalized treatment options.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13244-022-01264-x.

Impact of image quality on radiomics applications

Radiomics features extracted from medical images have been widely reported to be useful in the patient specific outcome modeling for variety of assessment and prediction purposes. Successful application of radiomics features as imaging biomarkers, however, is dependent on the robustness of the approach to the variation in each step of the modeling workflow. Variation in the input image quality is one of the main sources that impacts the reproducibility of radiomics analysis when a model is applied to broader range of medical imaging data. The quality of medical image is generally affected by both the scanner related factors such as image acquisition/reconstruction settings and the patient related factors such as patient motion. This article aimed to review the published literatures in this field that reported the impact of various imaging factors on the radiomics features through the change in image quality. The literatures were categorized by different imaging modalities and also tabulated based on the imaging parameters and the class of radiomics features included in the study. Strategies for image quality standardization were discussed based on the relevant literatures and recommendations for reducing the impact of image quality variation on the radiomics in multi-institutional clinical trial were summarized at the end of this article.

Radiomics-Based Deep Learning Prediction of Overall Survival in Non-Small-Cell Lung Cancer Using Contrast-Enhanced Computed Tomography

Simple Summary

The five-year survival rate of non-small-cell lung cancer (NSCLC), which accounts for 85% of all lung cancer cases, is only 10–20%. A reliable prediction model of overall survival (OS) that integrates imaging and clinical data is required. Overall, 492 patients with NSCLC from two hospitals were enrolled in this study. The compensation method was applied to reduce the variation of imaging features among different hospitals. We constructed a deep learning prediction model, DeepSurv, based on computed tomography radiomics and key clinical features to generate a personalized survival curve for each patient. The results of DeepSurv showed a good performance in discriminating high and low risk of survival. Furthermore, the generated personalized survival curves could be intuitively applied for individual OS prediction in clinical practice. We concluded that the proposed prediction model could benefit physicians, patients, and caregivers in managing NSCLC and facilitate personalized medicine.

Abstract

Patient outcomes of non-small-cell lung cancer (NSCLC) vary because of tumor heterogeneity and treatment strategies. This study aimed to construct a deep learning model combining both radiomic and clinical features to predict the overall survival of patients with NSCLC. To improve the reliability of the proposed model, radiomic analysis complying with the Image Biomarker Standardization Initiative and the compensation approach to integrate multicenter datasets were performed on contrast-enhanced computed tomography (CECT) images. Pretreatment CECT images and the clinical data of 492 patients with NSCLC from two hospitals were collected. The deep neural network architecture, DeepSurv, with the input of radiomic and clinical features was employed. The performance of survival prediction model was assessed using the C-index and area under the curve (AUC) 8, 12, and 24 months after diagnosis. The performance of survival prediction that combined eight radiomic features and five clinical features outperformed that solely based on radiomic or clinical features. The C-index values of the combined model achieved 0.74, 0.75, and 0.75, respectively, and AUC values of 0.76, 0.74, and 0.73, respectively, 8, 12, and 24 months after diagnosis. In conclusion, combining the traits of pretreatment CECT images, lesion characteristics, and treatment strategies could effectively predict the survival of patients with NSCLC using a deep learning model.

Prediction of lung malignancy progression and survival with machine learning based on pre-treatment FDG-PET/CT

BACKGROUND

Pre-treatment FDG-PET/CT scans were analyzed with machine learning to predict progression of lung malignancies and overall survival (OS).

METHODS

A retrospective review across three institutions identified patients with a pre-procedure FDG-PET/CT and an associated malignancy diagnosis. Lesions were manually and automatically segmented, and convolutional neural networks (CNNs) were trained using FDG-PET/CT inputs to predict malignancy progression. Performance was evaluated using area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity. Image features were extracted from CNNs and by radiomics feature extraction, and random survival forests (RSF) were constructed to predict OS. Concordance index (C-index) and integrated brier score (IBS) were used to evaluate OS prediction.

FINDINGS

1168 nodules (n=965 patients) were identified. 792 nodules had progression and 376 were progression-free. The most common malignancies were adenocarcinoma (n=740) and squamous cell carcinoma (n=179). For progression risk, the PET+CT ensemble model with manual segmentation (accuracy=0.790, AUC=0.876) performed similarly to the CT only (accuracy=0.723, AUC=0.888) and better compared to the PET only (accuracy=0.664, AUC=0.669) models. For OS prediction with deep learning features, the PET+CT+clinical RSF ensemble model (C-index=0.737) performed similarly to the CT only (C-index=0.730) and better than the PET only (C-index=0.595), and clinical only (C-index=0.595) models. RSF models constructed with radiomics features had comparable performance to those with CNN features.

INTERPRETATION

CNNs trained using pre-treatment FDG-PET/CT and extracted performed well in predicting lung malignancy progression and OS. OS prediction performance with CNN features was comparable to a radiomics approach. The prognostic models could inform treatment options and improve patient care.

FUNDING

NIH NHLBI training grant (5T35HL094308-12, John Sollee).