LIFEx: A Freeware for Radiomic Feature Calculation in Multimodality Imaging to Accelerate Advances in the Characterization of Tumor Heterogeneity

Primary tumor heterogeneity on pre-treatment [68Ga]Ga-PSMA PET/CT for the prediction of biochemical recurrence in prostate cancer

PURPOSE

The aim of this study is to research the value of the texture analysis of primary tumors in pre-treatment [68Ga]Ga-PSMA PET in the prediction of the development of biochemical recurrence (BCR) in prostate cancer patients who underwent definitive therapies.

METHODS

51 patients with prostate adenocarcinoma who had a pre-treatment [68Ga]Ga-PSMA-11 PET/CT and underwent definitive radiotherapy (RT) or radical prostatectomy (RP) were included in the study. Demographics, clinicopathologic features, the presence of BCR, and the last follow-up date of patients were recorded. Textural and conventional PET parameters (maximum standardized uptake value (SUVmax), total lesion-PSMA (TL-PSMA), and PSMA-tumor volume (PSMA-TV)) were obtained from PET/CT images using LifeX program. Parameters were grouped using the Youden index in ROC analysis. Factors predicting the BCR were determined using Cox regression analyses.

RESULTS

29 (56.9%) patients have received primary curative RT, while the remaining 22 (43.1%) patients have undergone RP. 5 (22.7%) patients with RP and 3 (10.3%) patients with curative RT have developed BCR during the follow-up. INTENSITY-BASED-minimum grey level (P=.050), GLCM-sum variance (P=.019), and GLCM-cluster prominence (P=.050) were associated with BCR in univariate analysis. INTENSITY-BASED-minimum grey level (P=.009) and GLCM-sum variance (P=.004) were found as independent predictors of BCR in the multivariate analysis.

CONCLUSION

Tumor heterogeneity on pre-treatment [68Ga]Ga-PSMA PET is associated with a high risk of BCR in PCa patients who underwent definitive therapies.

Robust machine learning challenge: An AIFM multicentric competition to spread knowledge, identify common pitfalls and recommend best practice

PURPOSE

A novel and unconventional approach to a machine learning challenge was designed to spread knowledge, identify robust methods and highlight potential pitfalls about machine learning within the Medical Physics community.

METHODS

A public dataset comprising 41 radiomic features and 535 patients was employed to assess the potential of radiomics in distinguishing between primary lung tumors and metastases. Each participant developed two classification models using: (i) all features (base model); (ii) only robust features (robust model). Both models were validated with cross-validation and on unseen data. The population stability index (PSI) was used as diagnostic metric for implementation issues. Performance was compared to reference. Base and robust models were compared in terms of performance and stability (coefficient of variation (CoV) of prediction probabilities).

RESULTS

PSI detected potential implementation errors in 70 % of models. The dataset exhibited strong imbalance. The average Gmean (i.e. an appropriate metric for imbalance) among all participants was 0.67 ± 0.01, significantly higher than reference Gmean of 0.50 ± 0.04. Robust models performances were slightly worse than base models (p < 0.05). Regarding stability, robust models exhibited lower median CoV on training set only.

CONCLUSION

AI4MP-Challenge models overperformed the reference, significantly improving the Gmean. Exclusion of less-robust features did not improve model robustness and it should be avoided when confounding effects are absent. Other methods, like harmonization or data augmentation, should be evaluated. This study demonstrated how the collaborative effort to foster knowledge on machine learning among medical physicists, through interactive sessions and exchange of information among participants, can result in improved models.

Single- and multi-site radiomics may improve overall survival prediction for patients with metastatic lung adenocarcinoma

PURPOSE

The purpose of this study was to assess whether single-site and multi-site radiomics could improve the prediction of overall survival (OS) of patients with metastatic lung adenocarcinoma compared to clinicopathological model.

MATERIALS AND METHODS

Adults with metastatic lung adenocarcinoma, pretreatment whole-body contrast-enhanced computed tomography examinations, and performance status (WHO-PS) ≤ 2 were included in this retrospective single-center study, and randomly assigned to training and testing cohorts. Radiomics features (RFs) were extracted from all measurable lesions with volume ≥ 1 cm3. Radiomics prognostic scores based on the largest tumor (RPSlargest) and the average RF values across all tumors per patient (RPSaverage) were developed in the training cohort using 5-fold cross-validated LASSO-penalized Cox regression. Intra-patient inter-tumor heterogeneity (IPITH) metrics were calculated to quantify the radiophenotypic dissimilarities among all tumors within each patient. A clinicopathological model was built in the training cohort using stepwise Cox regression and enriched with combinations of RPSaverage, RPSlargest and IPITH. Models were compared with the concordance index in the independent testing cohort.

RESULTS

A total of 300 patients (median age: 63.7 years; 40.7% women; median OS, 16.3 months) with 1359 lesions were included (200 and 100 patients in the training and testing cohorts, respectively). The clinicopathological model included WHO-PS = 2 (hazard ratio [HR] = 3.26; P < 0.0001), EGFR, ALK, ROS1 or RET mutations (HR = 0.57; P = 0.0347), IVB stage (HR = 1.65; P = 0.0211), and liver metastases (HR = 1.47; P = 0.0670). In the testing cohort, RPSaverage, RPSlargest and IPITH were associated with OS (HR = 85.50, P = 0.0038; HR = 18.83, P = 0.0082 and HR = 8.00, P = 0.0327, respectively). The highest concordance index was achieved with the combination of clinicopathological variables and RPSaverage, significantly better than that of the clinicopathological model (concordance index = 0.7150 vs. 0.695, respectively; P = 0.0049) CONCLUSION: Single-site and multi-site radiomics-based scores are associated with OS in patients with metastatic lung adenocarcinoma. RPSaverage improves the clinicopathological model.

Prognostic Value of Computed Tomography-Measured Visceral Adipose Tissue in Patients with Pulmonary Infection Caused by Carbapenem-Resistant Klebsiella pneumoniae

Objective

This study aimed to investigate the correlation between computed tomography (CT) derived body composition and 30-day mortality in patients with pulmonary infections caused by carbapenem-resistant Klebsiella pneumoniae (K. pneumoniae).

Methods

A total of 89 eligible participants from a tertiary teaching hospital, enrolled between January 1, 2016, and December 31, 2020, were included in the study. We analyzed the relationship between visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), total adipose tissue (TAT), and skeletal muscle (SM) and 30-day mortality in patients infected with carbapenem-resistant K. pneumoniae (CRKP) in the pulmonary region. Furthermore, we established Cox regression models and a personalized nomogram model to predict the probability of 30-day mortality in these infected patients.

Results

Individuals with high VAT exhibited a higher likelihood of 30-day all-cause mortality (P<0.01) and 30-day mortality due to CRKP infection (P<0.01) compared to those with low VAT. Similar results were observed for TAT. After adjusting for significant comorbidities and other clinical characteristics, Cox regression analysis revealed that male gender (adjusted HR = 4.37; 95% CI = 0.96-19.92, P=0.06), vasopressor use (adjusted HR = 3.65; 95% CI = 1.04-12.85, P=0.04), and VAT (adjusted HR = 1.16; 95% CI = 1.01-1.34, P=0.03) were independent risk factors for 30-day all-cause mortality among these infectious patients.

Conclusion

The study results highlight the significant prognostic value of CT-quantified visceral adipose tissue in patients with CRKP pulmonary infection. Individuals with high VAT are more prone to mortality within 30 days compared to those with low VAT.

Prediction of hepatocellular carcinoma response to radiation segmentectomy using an MRI-based machine learning approach

PURPOSE

To evaluate the value of pre-treatment MRI-based radiomics in patients with hepatocellular carcinoma (HCC) for the prediction of response to Yttrium 90 radiation segmentectomy.

METHODS

This retrospective study included 154 patients (38 female; mean age 66.8 years) who underwent contrast-enhanced MRI prior to radiation segmentectomy. Radiomics features were manually extracted on volumes of interest on post-contrast T1-weighted images at the portal venous phase (PVP). Tumor-based response assessment was evaluated 6 months post-treatment using mRECIST. A logistic regression model was used to predict binary response outcome [complete response at 6 months with no-re-treatment (response group) against the rest (non-response group, including partial response, progressive disease, stable disease and complete response after re-treatment within 6 months after radiation segmentectomy) using baseline clinical parameters and radiomics features. We accessed the value of different sets of predictors using cross-validation technique. AUCs were compared using DeLong tests.

RESULTS

A total 168 HCCs (mean size 2.9 ± 1.7 cm) were analyzed in 154 patients. The response group consisted of 113 HCCs and the non-response group of 55 HCCs. Baseline clinical parameters (AUC 0.531; sensitivity, 0.781; specificity, 0.279; positive predictive value (PPV), 0.345; negative predictive value (NPV), 0.724) and AFP (AUC 0.632; sensitivity, 0.833; specificity, 0.466; PPV, 0.432; NPV, 0.851) showed poor performance for response prediction. The model using a combination of radiomics features and clinical parameters/AFP showed the best performance (AUC 0.736; sensitivity, 0.706; specificity, 0.662; PPV 0.504; NPV, 0.822), significantly better than the clinical model (p < 0.001) or AFP alone (p < 0.001).

CONCLUSION

The combination of radiomics features from pre-treatment MRI with clinical parameters and AFP showed fair performance for predicting HCC response to radiation segmentectomy, better than that of AFP. These results need further validation.

Multicentric study on the reproducibility and robustness of PET-based radiomics features with a realistic activity painting phantom

Previously, we developed an "activity painting" tool for PET image simulation; however, it could simulate heterogeneous patterns only in the air. We aimed to improve this phantom technique to simulate arbitrary lesions in a radioactive background to perform relevant multi-center radiomic analysis. We conducted measurements moving a 22Na point source in a 20-liter background volume filled with 5 kBq/mL activity with an adequately controlled robotic system to prevent the surge of the water. Three different lesion patterns were "activity-painted" in five PET/CT cameras, resulting in 8 different reconstructions. We calculated 46 radiomic indeces (RI) for each lesion and imaging setting, applying absolute and relative discretization. Reproducibility and reliability were determined by the inter-setting coefficient of variation (CV) and the intraclass correlation coefficient (ICC). Hypothesis tests were used to compare RI between lesions. By simulating precisely the same lesions, we confirmed that the reconstructed voxel size and the spatial resolution of different PET cameras were critical for higher order RI. Considering conventional RIs, the SUVpeak and SUVmean proved the most reliable (CV<10%). CVs above 25% are more common for higher order RIs, but we also found that low CVs do not necessarily imply robust parameters but often rather insensitive RIs. Based on the hypothesis test, most RIs could clearly distinguish between the various lesions using absolute resampling. ICC analysis also revealed that most RIs were more reproducible with absolute discretization. The activity painting method in a real radioactive environment proved suitable for precisely detecting the radiomic differences derived from the different camera settings and texture characteristics. We also found that inter-setting CV is not an appropriate metric for analyzing RI parameters' reliability and robustness. Although multicentric cohorts are increasingly common in radiomics analysis, realistic texture phantoms can provide indispensable information on the sensitivity of an RI and how an individual RI parameter measures the texture.

The Predictive Role of Radiomics in Breast Cancer Patients Imaged by [18F]FDG PET: Preliminary Results from a Prospective Cohort

BACKGROUND

Recently, radiomics has emerged as a possible image-derived biomarker, predominantly stemming from retrospective analyses. We aimed to prospectively assess the predictive role of [18F]FDG-PET radiomics in breast cancer (BC).

METHODS

Patients affected by stage I-III BC eligible for neoadjuvant chemotherapy (NAC) staged with [18F]FDG-PET/CT were prospectively enrolled. The pathological response to NAC was assessed on surgical specimens. From each primary breast lesion, we extracted radiomic PET features and their predictive role with respect to pCR was assessed. Uni- and multivariate statistics were used for inference; principal component analysis (PCA) was used for dimensionality reduction.

RESULTS

We analysed 93 patients (53 HER2+ and 40 triple-negative (TNBC)). pCR was achieved in 44/93 cases (24/53 HER2+ and 20/40 TNBC). Age, molecular subtype, Ki67 percent, and stage could not predict pCR in multivariate analysis. In univariate analysis, 10 radiomic indices resulted in p < 0.1. We found that 3/22 radiomic principal components were discriminative for pCR. Using a cross-validation approach, radiomic principal components failed to discriminate pCR groups but predicted the stage (mean accuracy = 0.79 ± 0.08).

CONCLUSIONS

This study shows the potential of PET radiomics for staging purposes in BC; the possible role of radiomics in predicting the pCR response to NAC in BC needs to be further investigated.

Feasibility of shortening scan duration of 18F-FDG myocardial metabolism imaging using a total-body PET/CT scanner

PURPOSE

To evaluate 18F-FDG myocardial metabolism imaging (MMI) using a total-body PET/CT scanner and explore the feasible scan duration to guide the clinical practice.

METHODS

A retrospective analysis was conducted on 41 patients who underwent myocardial perfusion-metabolism imaging to assess myocardial viability. The patients underwent 18F-FDG MMI with a total-body PET/CT scanner using a list-mode for 600 s. PET data were trimmed and reconstructed to simulate images of 600-s, 300-s, 120-s, 60-s, and 30-s acquisition time (G600-G30). Images among different groups were subjectively evaluated using a 5-point Likert scale. Semi-quantitative evaluation was performed using standardized uptake value (SUV), myocardial to background activity ratio (M/B), signal to noise ratio (SNR), contrast to noise ratio (CNR), contrast ratio (CR), and coefficient of variation (CV). Myocardial viability analysis included indexes of Mismatch and Scar. G600 served as the reference.

RESULTS

Subjective visual evaluation indicated a decline in the scores of image quality with shortening scan duration. All the G600, G300, and G120 images were clinically acceptable (score ≥ 3), and their image quality scores were 4.9 ± 0.3, 4.8 ± 0.4, and 4.5 ± 0.8, respectively (P > 0.05). Moreover, as the scan duration reduced, the semi-quantitative parameters M/B, SNR, CNR, and CR decreased, while SUV and CV increased, and significant difference was observed in G300-G30 groups when comparing to G600 group (P < 0.05). For myocardial viability analysis of left ventricular and coronary segments, the Mismatch and Scar values of G300-G30 groups were almost identical to G600 group (ICC: 0.968-1.0, P < 0.001).

CONCLUSION

Sufficient image quality for clinical diagnosis could be achieved at G120 for MMI using a total-body PET/CT scanner, while the image quality of G30 was acceptable for myocardial viability analysis.

Clinical value of SUVpeak-to-tumor centroid distance on FDG PET/CT for predicting neoadjuvant chemotherapy response in patients with breast cancer

BACKGROUND

Since it has been found that the maximum metabolic activity of a cancer lesion shifts toward the lesion edge during cancer progression, normalized distances from the hot spot of radiotracer uptake to tumor centroid (NHOC) and tumor perimeter (NHOP) have been suggested as novel F-18 fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) parameters that can reflect cancer aggressiveness. This study aimed to investigate whether NHOC and NHOP parameters could predict pathological response to neoadjuvant chemotherapy (NAC) and progression-free survival (PFS) in breast cancer patients.

METHODS

This study retrospectively enrolled 135 female patients with breast cancer who underwent pretreatment FDG PET/CT and received NAC and subsequent surgical resection. From PET/CT images, normalized distances of maximum SUV and peak SUV-to-tumor centroid (NHOCmax and NHOCpeak) and -to-tumor perimeter (NHOPmax and NHOPpeak) were measured, in addition to conventional PET/CT parameters.

RESULTS

Of 135 patients, 32 (23.7%) achieved pathological complete response (pCR), and 34 (25.2%) had events during follow-up. In the receiver operating characteristic (ROC) curve analysis, NHOCmax showed the highest area under the ROC curve value (0.710) for predicting pCR, followed by NHOCpeak (0.694). In the multivariate logistic regression analysis, NHOCmax, NHOCpeak, and NHOPmax were independent predictors for pCR (p < 0.05). In the multivariate survival analysis, NHOCpeak (p = 0.026) was an independent predictor for PFS along with metabolic tumor volume, with patients having higher NHOCpeak showing worse PFS.

CONCLUSION

NHOCpeak on pretreatment FDG PET/CT could be a potential imaging parameter for predicting NAC response and survival in patients with breast cancer.

Texture analysis can predict response to etoposide-doxorubicin-cisplatin in patients with adrenocortical carcinoma

BACKGROUND

The adrenocortical carcinoma (ACC) is a rare and highly aggressive malignancy originating from the adrenal cortex. These patients usually undergo chemotherapy with etoposide, doxorubicin, cisplatin and mitotane (EDP-M) in case of locally advanced or metastatic ACC. Computed tomography (CT) radiomics showed to be useful in adrenal pathologies. The study aimed to analyze the association between response to EDP-M treatment and CT textural features at diagnosis in patients with locally advanced or metastatic ACCs.

METHODS

We enrolled 17 patients with advanced or metastatic ACC who underwent CT before and after EDP-M therapy. The response to treatment was evaluated according to RECIST 1.1, Choi, and volumetric criteria. Based on the aforementioned criteria, the patients were classified as responders and not responders. Textural features were extracted from the biggest lesion in contrast-enhanced CT images with LifeX software. ROC curves were drawn for the variables that were significantly different (p < 0.05) between the two groups.

RESULTS

Long-run high grey level emphasis (LRHGLE_GLRLM) and histogram kurtosis were significantly different between responder and not responder groups (p = 0.04) and the multivariate ROC curve combining the two features showed a very good AUC (0.900; 95%IC: 0.724-1.000) in discriminating responders from not responders. More heterogeneous tissue texture of initial staging CT in locally advanced or metastatic ACC could predict the positive response to EDP-M treatment.

CONCLUSIONS

Adrenal texture is able to predict the response to EDP-M therapy in patients with advanced ACC.

Pre-treatment [18F]FDG PET/CT biomarkers for the prediction of antibody-drug conjugates efficacy in metastatic breast cancer

PURPOSE

This study aimed to evaluate the association between pretreatment [18F]FDG PET/CT-derived biomarkers and outcomes in metastatic breast cancer (mBC) patients treated with antibody-drug conjugates (ADCs) Sacituzumab Govitecan (SG) and Trastuzumab Deruxtecan (T-DXd).

METHODS

A retrospective bicentric analysis was conducted on triple-negative mBC (mTNBC) patients treated with SG and HER2-low mBC patients treated with T-DXd, who underwent [18F]FDG PET/CT scans before therapy. Key biomarkers, including maximum standardized uptake value (SUVmax), total metabolic tumor volume (TMTV) and maximum tumor dissemination (Dmax), were measured. Their prognostic value for progression-free survival (PFS) and overall survival (OS) was assessed using Cox models and Kaplan-Meier curves.

RESULTS

128 patients were included: 71 mTNBC treated with SG and 57 HR-positive and -negative HER2-low mBC treated with T-DXd. Median follow-up was 12.9 months. In the SG cohort, median PFS and OS were 4.8 and 8.9 months, respectively. High Dmax (HR 2.1, 95% CI 1.1-4.3) and high TMTV (HR 2.9, 95% CI 1.2-6.6) were independently associated with shorter OS. In the T-DXd cohort, median PFS and OS were 5.8 and 9.0 months, respectively. High Dmax (HR 2.1, 95% CI 1.2-3.9) and high TMTV (HR 2.4, 95% CI 1.0-6.5) independently correlated with shorter PFS and shorter OS, respectively.

CONCLUSION

Pretreatment [18F]FDG PET/CT-derived biomarkers, namely TMTV and Dmax, have significant prognostic value in patients with mTNBC and HER2-low mBC treated with SG and T-DXd. These biomarkers improve prognostic prediction and may optimize treatment strategies, warranting their clinical use, but larger studies are needed to validate these findings.

PET radiomics in lung cancer: advances and translational challenges

Radiomics is an emerging field of medical imaging that aims at improving the accuracy of diagnosis, prognosis, treatment planning and monitoring non-invasively through the automated or semi-automated quantitative analysis of high-dimensional image features. Specifically in the field of nuclear medicine, radiomics utilizes imaging methods such as positron emission tomography (PET) and single photon emission computed tomography (SPECT) to evaluate biomarkers related to metabolism, blood flow, cellular activity and some biological pathways. Lung cancer ranks among the leading causes of cancer-related deaths globally, and radiomics analysis has shown great potential in guiding individualized therapy, assessing treatment response, and predicting clinical outcomes. In this review, we summarize the current state-of-the-art radiomics progress in lung cancer, highlighting the potential benefits and existing limitations of this approach. The radiomics workflow was introduced first including image acquisition, segmentation, feature extraction, and model building. Then the published literatures were described about radiomics-based prediction models for lung cancer diagnosis, differentiation, prognosis and efficacy evaluation. Finally, we discuss current challenges and provide insights into future directions and potential opportunities for integrating radiomics into routine clinical practice.

PSMA PET/CT for treatment response evaluation at predefined time points is superior to PSA response for predicting survival in metastatic castration-resistant prostate cancer patients

BACKGROUND

In metastatic castration-resistant prostate cancer (mCRPC), using serum prostate-specific antigen (PSA) levels to evaluate treatment response is not always accurate. This study aimed to assess the efficacy of PSMA PET/CT at specific time points for evaluating treatment response and predicting survival in mCRPC patients, compared to PSA.

METHODS

Sixty mCRPC patients underwent [18F]PSMA-1007 PET/CT at baseline and for treatment response evaluation of either androgen receptor-targeted agents (after 3 months) or chemotherapy (after completion), and were retrospectively analysed. Visual assessment categorised overall response and response of the worst responding lesion as partial response, stable disease, or progressive disease, using the EAU/EANM criteria. Additionally, percentage changes in SUVmax, total tumour volume and total lesion uptake (tumour volume * SUVmean) were calculated. PSA response was defined according to the PCWG3 criteria. Cox regression analysis identified predictors of overall survival.

RESULTS

PSMA PET/CT and PSA response were discordant in 47 % of patients, and PSMA PET/CT response was worse in 89 % of these cases. Overall response on PSMA PET/CT independently predicted overall survival (progression versus non-progression: HR = 4.05, p < 0.001), outperforming PSA response (progression versus non-progression: HR = 2.53, p = 0.010) and other PSMA PET/CT parameters. Among patients with a PSA decline of > 50 %, 31 % showed progressive disease on PSMA PET/CT, correlating with higher mortality risk (progression versus non-progression: HR = 4.38, p = 0.008). No flare in PSMA uptake was observed in this cohort.

CONCLUSIONS

PSMA PET/CT for assessing treatment response at predefined time points was superior to PSA-based response for predicting overall survival in mCRPC patients treated with androgen receptor-targeted agents and chemotherapy. PSMA PET/CT showed the ability to detect disease progression earlier than PSA levels, which can affect treatment decisions and has the potential to improve patient outcomes. We recommend further research to validate these findings in larger patient cohorts, to extend the number of treatments, and to evaluate cost-effectiveness and impact on patient outcomes.

Performance and Clinical Impact of Radiomics and 3D-CNN Models for the Diagnosis of Neurodegenerative Parkinsonian Syndromes on 18 F-FDOPA PET

PURPOSE

The aim of this study was to compare the performance and added clinical value of a semiautomated radiomics model and an automated 3-dimensinal convolutional neural network (3D-CNN) model for diagnosing neurodegenerative parkinsonian syndromes on 18 F-FDOPA PET images.

PATIENTS AND METHODS

This 2-center retrospective study included 687 patients with motor symptoms consistent with parkinsonian syndrome. All patients underwent 18 F-FDOPA brain PET scans, acquired on 3 PET systems from 2 different hospitals, and classified as pathological or nonpathological (by an expert nuclear physician). Artificial intelligence models were trained to replicate this medical expert's classification using 2 pipelines. The radiomics pipeline was semiautomated and involved manually segmenting the bilateral caudate and putamen nuclei; 43 radiomic features were extracted and combined using the support vector machine method. The deep learning pipeline was fully automatic and used a 3D-CNN model. Both models were trained on 417 patients and tested on an internal (n = 100) and an external (n = 170) test set. The final models' performance was evaluated using balanced accuracy and compared with that of a junior medical expert and nonexpert nuclear physician.

RESULTS

On the internal test set, the 3D-CNN model outperformed the radiomic model with a balanced accuracy of 99% (vs 96%). It led to diagnostic performance similar to that of a junior medical expert (only 1 in 100 patients misclassified by both). On the external test set from a less experienced hospital, the 3D-CNN model allowed physicians to correctly reclassify the diagnosis of 10 out 170 patients (6%).

CONCLUSIONS

The developed 3D-CNN model can automatically diagnose neurodegenerative parkinsonian syndromes, also reducing diagnostic errors by 6% in less-experienced hospitals.

Radiomics reproducibility in computed tomography through changes of ROI size, resolution, and hounsfield unit: A phantom study

INTRODUCTION

Although radiomics has revealed an intriguing perspective for quantitative radiology, the impact of scanning parameters on its outcomes must be considered. In this study, the effects of changes in the region of interest (ROI) sizes, Hounsfield Unit (HU), and resolution of computed tomography (CT) on feature reproducibility have been investigated.

METHODS

The GAMMEX 464 phantom was used to evaluate the reproducibility of radiomics features across different ROI sizes, HU, and resolution. Data were acquired using a consistent system setup, with the phantom repositioned for each scan. The first acquisition series examined the effects of different ROI sizes and resolutions (1, 3, and 5 mm) on feature reproducibility. The second series assessed the impact of different HU and resolution. Segmentation and feature extraction were performed using LIFEx 7.1.0 software, focusing on textural radiomics features. Statistical analysis involved calculating the coefficient of variation (COV) to categorize feature variability. COV <5 % was considered highly stable.

RESULTS

Out of the 32 textural features studied, the analysis of changes in ROI size with a resolution of 1 mm, 3 mm, and 5 mm revealed that 16, 17, and 18 features had high reproducibility, with a COV<5 %. Polyethylene, acrylic, and water also demonstrated stable textural features across changes in scan parameters and image resolutions, with 4 reproducible features in all resolutions. The grey-level run length matrix (GLRLM) and grey-level zone length matrix (GLZLM) radiomics groups were highly stable in the context of variations in scan parameters and different materials.

CONCLUSION

The results of this study highlight the importance of standardizing radiomics studies to reduce the influence of pre-analysis steps on feature values. This standardization is crucial for guaranteeing the consistency of radiomics features under various imaging conditions. Additional research is required to enhance these results.

IMPLICATIONS FOR PRACTICE

To ensure the reproducibility and reliability of radiomics features, it is imperative to standardize scanning parameters and pre-analysis protocols. This standardization will enhance the consistency of radiomics applications in both clinical and research environments.

High-performance presurgical differentiation of glioblastoma and metastasis by means of multiparametric neurite orientation dispersion and density imaging (NODDI) radiomics

OBJECTIVES

To evaluate the performance of multiparametric neurite orientation dispersion and density imaging (NODDI) radiomics in distinguishing between glioblastoma (Gb) and solitary brain metastasis (SBM).

MATERIALS AND METHODS

In this retrospective study, NODDI images were curated from 109 patients with Gb (n = 57) or SBM (n = 52). Automatically segmented multiple volumes of interest (VOIs) encompassed the main tumor regions, including necrosis, solid tumor, and peritumoral edema. Radiomics features were extracted for each main tumor region, using three NODDI parameter maps. Radiomics models were developed based on these three NODDI parameter maps and their amalgamation to differentiate between Gb and SBM. Additionally, radiomics models were constructed based on morphological magnetic resonance imaging (MRI) and diffusion imaging (diffusion-weighted imaging [DWI]; diffusion tensor imaging [DTI]) for performance comparison.

RESULTS

The validation dataset results revealed that the performance of a single NODDI parameter map model was inferior to that of the combined NODDI model. In the necrotic regions, the combined NODDI radiomics model exhibited less than ideal discriminative capabilities (area under the receiver operating characteristic curve [AUC] = 0.701). For peritumoral edema regions, the combined NODDI radiomics model achieved a moderate level of discrimination (AUC = 0.820). Within the solid tumor regions, the combined NODDI radiomics model demonstrated superior performance (AUC = 0.904), surpassing the models of other VOIs. The comparison results demonstrated that the NODDI model was better than the DWI and DTI models, while those of the morphological MRI and NODDI models were similar.

CONCLUSION

The NODDI radiomics model showed promising performance for preoperative discrimination between Gb and SBM.

CLINICAL RELEVANCE STATEMENT

The NODDI radiomics model showed promising performance for preoperative discrimination between Gb and SBM, and radiomics features can be incorporated into the multidimensional phenotypic features that describe tumor heterogeneity.

KEY POINTS

• The neurite orientation dispersion and density imaging (NODDI) radiomics model showed promising performance for preoperative discrimination between glioblastoma and solitary brain metastasis. • Compared with other tumor volumes of interest, the NODDI radiomics model based on solid tumor regions performed best in distinguishing the two types of tumors. • The performance of the single-parameter NODDI model was inferior to that of the combined-parameter NODDI model.

Value of Presurgical 18F-FDG PET/CT Radiomics for Predicting Mediastinal Lymph Node Metastasis in Patients with Lung Adenocarcinoma

Objective: The aim of this study was to develop a 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) radiomic model for predicting mediastinal lymph node metastasis (LNM) in presurgical patients with lung adenocarcinoma. Methods: The study enrolled 320 patients with lung adenocarcinoma (288 internal and 32 external cases) and extracted 190 radiomic features using the LIFEx package. Optimal radiomic features to build a radiomic model were selected using the least absolute shrinkage and selection operator algorithm. Logistic regression was used to build the clinical and complex (combined radiomic and clinical variables) models. Results: Ten radiomic features were selected. In the training group, the area under the receiver operating characteristic curve of the complex model was significantly higher than that of the radiomic and clinical models [0.924 (95% CI: 0.887-0.961) vs. 0.863 (95% CI: 0.814-0.912; p = 0.001) and 0.838 (95% CI: 0.783-0.894; p = 0.000), respectively]. The sensitivity, specificity, accuracy, and positive and negative predictive values of the radiomic model were 0.857, 0.790, 0.811, and 0.651 and 0.924, respectively, which were better than that of visual evaluation (0.539, 0.724, 0.667, and 0.472 and 0.775, respectively) and PET semiquantitative analyses (0.619, 0.732, 0.697, and 0.513 and 0.808, respectively). Conclusions: 18F-FDG PET/CT radiomics showed good predictive performance for LNM and improved the N-stage accuracy of lung adenocarcinoma.

Dual Somatostatin Receptor/18F-FDG PET/CT Imaging in Patients with Well-Differentiated, Grade 2 and 3 Gastroenteropancreatic Neuroendocrine Tumors

Our purpose was to prospectively assess the distribution of NETPET scores in well-differentiated (WD) grade 2 and 3 gastroenteropancreatic (GEP) neuroendocrine tumors (NETs) and to determine the impact of the NETPET score on clinical management. Methods: This single-arm, institutional ethics review board-approved prospective study included 40 patients with histologically proven WD GEP NETs. 68Ga-DOTATATE PET and 18F-FDG PET were performed within 21 d of each other. NETPET scores were evaluated qualitatively by 2 reviewers, with up to 10 marker lesions selected for each patient. The quantitative parameters that were evaluated included marker lesion SUVmax for each tracer; 18F-FDG/68Ga-DOTATATE SUVmax ratios; functional tumor volume (FTV) and metabolic tumor volume (MTV) on 68Ga-DOTATATE and 18F-FDG PET, respectively; and FTV/MTV ratios. The treatment plan before and after 18F-FDG PET was recorded. Results: There were 22 men and 18 women (mean age, 60.8 y) with grade 2 (n = 24) or grade 3 (n = 16) tumors and a mean Ki-67 index of 16.1%. NETPET scores of P0, P1, P2A, P2B, P3B, P4B, and P5 were documented in 2 (5%), 5 (12.5%), 5 (12.5%) 20 (50%), 2 (5%), 4 (10%), and 2 (5%) patients, respectively. No association was found between the SUVmax of target lesions on 68Ga-DOTATATE and the SUVmax of target lesions on 18F-FDG PET (P = 0.505). 18F-FDG/68Ga-DOTATATE SUVmax ratios were significantly lower for patients with low (P1-P2) primary NETPET scores than for those with high (P3-P5) primary NETPET scores (mean ± SD, 0.20 ± 0.13 and 1.68 ± 1.44, respectively; P < 0.001). MTV on 18F-FDG PET was significantly lower for low primary NETPET scores than for high ones (mean ± SD, 464 ± 601 cm3 and 66 ± 114 cm3, respectively; P = 0.005). A change in the type of management was observed in 42.5% of patients after 18F-FDG PET, with the most common being a change from systemic therapy to peptide receptor radionuclide therapy and from debulking surgery to systemic therapy. Conclusion: There was a heterogeneous distribution of NETPET scores in patients with WD grade 2 and 3 GEP NETs, with more than 1 in 5 patients having a high NETPET score and a frequent change in management after 18F-FDG PET. Quantitative parameters including 18F-FDG/68Ga-DOTATATE SUVmax ratios in target lesions and FTV/MTV ratios can discriminate between patients with high and low NETPET scores.

Prediction of Lesion-Based Treatment Response after Two Cycles of Lu-177 Prostate Specific Membrane Antigen Treatment in Metastatic Castration-Resistant Prostate Cancer Using Machine Learning

INTRODUCTION

Lutetium-177 (Lu-177) prostate-specific membrane antigen (PSMA) therapy is a radionuclide treatment that prolongs overall survival in metastatic castration-resistant prostate cancer (MCRPC). We aimed to predict lesion-based treatment response after Lu-177 PSMA treatment using machine learning with texture analysis data obtained from pretreatment Gallium-68 (Ga-68) PSMA positron emission tomography/computed tomography (PET/CT).

METHODS

Eighty-three progressed, and 91 nonprogressed malignant foci on pretreatment Ga-68 PSMA PET/CT of 9 patients were used for analysis. Malignant foci with at least a 30% increase in Ga-68 PSMA uptake after two cycles of treatment were considered progressed lesions. All other changes in Ga-68 PSMA uptake of the lesions were considered nonprogressed lesions. The classifiers tried to predict progressed lesions.

RESULTS

Logistic regression, Naive Bayes, and k-nearest neighbors' area under the ROC curve (AUC) values in detecting progressed lesions in the training group were 0.956, 0.942, and 0.950, respectively, and their accuracy was 87%, 85%, and 89%, respectively. The AUC values of the classifiers in the testing group were 0.937, 0.954, and 0.867, respectively, and their accuracy was 85%, 88%, and 79%, respectively.

CONCLUSION

Using machine learning with texture analysis data obtained from pretreatment Ga-68 PSMA PET/CT in MCRPC predicted lesion-based treatment response after two cycles of Lu-177 PSMA treatment.

Value of 18F-FDG PET/CT-based radiomics features for differentiating primary lung cancer and solitary lung metastasis in patients with colorectal adenocarcinoma

OBJECTIVE

To investigate the value and applicability of 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) radiomics in differentiating primary lung cancer (PLC) from solitary lung metastasis (SLM) in patients with colorectal cancer (CRC).

MATERIALS AND METHODS

This retrospective study included 103 patients with CRC and solitary pulmonary nodules (SPNs). The least absolute shrinkage and selection operator (LASSO) was used to screen for optimal radiomics features and establish a PET/CT radiomics model. PET/CT Visual and complex models (combining radiomics with PET/CT visual features) were developed. The area under the receiver operating characteristic (ROC) curve (AUC) was used to determine the predictive value and diagnostic efficiency of the models.

RESULTS

The AUC of the PET/CT radiomics model for differentiating PLC from SLM was 0.872 (95% CI: 0.806-0.939), which was not different from that of the visual (0.829 [95% CI: 0.749-0.908; p = .352]). However, the AUC of the complex model (0.936 [95% CI:0.892-0.981]) was significantly higher than that of the PET/CT radiomics (p = .005) and visual model (p = .001). The sensitivity (SEN), specificity (SPE), accuracy (ACC), positive predictive value (PPV), and negative predictive value (NPV) of PET/CT radiomics for differentiating PLC from SLM were 0.720, 0.887, 0.806, 0.857, and 0.770, respectively.

CONCLUSION

PET/CT radiomics can effectively distinguish PLC and SLM in patients with CRC and SPNs and guide the implementation of personalized treatment.

Applying deep learning-based ensemble model to [18F]-FDG-PET-radiomic features for differentiating benign from malignant parotid gland diseases

OBJECTIVES

To develop and identify machine learning (ML) models using pretreatment 2-deoxy-2-[18F]fluoro-D-glucose ([18F]-FDG)-positron emission tomography (PET)-based radiomic features to differentiate benign from malignant parotid gland diseases (PGDs).

MATERIALS AND METHODS

This retrospective study included 62 patients with 63 PGDs who underwent pretreatment [18F]-FDG-PET/computed tomography (CT). The lesions were assigned to the training (n = 44) and testing (n = 19) cohorts. In total, 49 [18F]-FDG-PET-based radiomic features were utilized to differentiate benign from malignant PGDs using five different conventional ML algorithmic models (random forest, neural network, k-nearest neighbors, logistic regression, and support vector machine) and the deep learning (DL)-based ensemble ML model. In the training cohort, each conventional ML model was constructed using the five most important features selected by the recursive feature elimination method with the tenfold cross-validation and synthetic minority oversampling technique. The DL-based ensemble ML model was constructed using the five most important features of the bagging and multilayer stacking methods. The area under the receiver operating characteristic curves (AUCs) and accuracies were used to compare predictive performances.

RESULTS

In total, 24 benign and 39 malignant PGDs were identified. Metabolic tumor volume and four GLSZM features (GLSZM_ZSE, GLSZM_SZE, GLSZM_GLNU, and GLSZM_ZSNU) were the five most important radiomic features. All five features except GLSZM_SZE were significantly higher in malignant PGDs than in benign ones (each p < 0.05). The DL-based ensemble ML model had the best performing classifier in the training and testing cohorts (AUC = 1.000, accuracy = 1.000 vs AUC = 0.976, accuracy = 0.947).

CONCLUSIONS

The DL-based ensemble ML model using [18F]-FDG-PET-based radiomic features can be useful for differentiating benign from malignant PGDs. The DL-based ensemble ML model using [18F]-FDG-PET-based radiomic features can overcome the previously reported limitation of [18F]-FDG-PET/CT scan for differentiating benign from malignant PGDs. The DL-based ensemble ML approach using [18F]-FDG-PET-based radiomic features can provide useful information for managing PGD.

Peritumoral Adipose Tissue Features Derived from [18F]fluoro-2-deoxy-2-d-glucose Positron Emission Tomography/Computed Tomography as Predictors for Response to Neoadjuvant Chemotherapy in Breast Cancer Patients

This study investigated whether the textural features of peritumoral adipose tissue (AT) on [18F]fluoro-2-deoxy-2-d-glucose (FDG) positron emission tomography/computed tomography (PET/CT) can predict the pathological response to neoadjuvant chemotherapy (NAC) and progression-free survival (PFS) in breast cancer patients. We retrospectively enrolled 147 female breast cancer patients who underwent staging FDG PET/CT and completed NAC and underwent curative surgery. We extracted 10 first-order features, 6 gray-level co-occurrence matrix (GLCM) features, and 3 neighborhood gray-level difference matrix (NGLDM) features of peritumoral AT and evaluated the predictive value of those imaging features for pathological complete response (pCR) and PFS. The results of our study demonstrated that GLCM homogeneity showed the highest predictability for pCR among the peritumoral AT imaging features in the receiver operating characteristic curve analysis. In multivariate logistic regression analysis, the mean standardized uptake value (SUV), 50th percentile SUV, 75th percentile SUV, SUV histogram entropy, GLCM entropy, and GLCM homogeneity of the peritumoral AT were independent predictors for pCR. In multivariate survival analysis, SUV histogram entropy and GLCM correlation of peritumoral AT were independent predictors of PFS. Textural features of peritumoral AT on FDG PET/CT could be potential imaging biomarkers for predicting the response to NAC and disease progression in breast cancer patients.

Baseline 18F-FDG PET Radiomics Predicting Therapeutic Efficacy of Diffuse Large B-Cell Lymphoma after R-CHOP (-Like) Therapy

Objective: This study aimed to predict therapeutic efficacy among diffuse large B-cell lymphoma (DLBCL) after R-CHOP (-like) therapy using baseline 18F-fluorodeoxyglucose positron emission tomography (18F-FDG PET) radiomics. Methods: A total of 239 patients with DLBCL were enrolled in this study, with 82 patients having refractory/relapsed disease. The radiomics signatures were developed using a stacking ensemble approach. The efficacy of the radiomics signatures, the National Comprehensive Cancer Network-International Prognostic Index (NCCN-IPI), conventional PET parameters model, and their combinations in assessing refractory/relapse risk were evaluated using receiver operating characteristic (ROC) curves, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy and decision curve analysis. Results: The stacking model, along with the integrated model that combines stacking with the NCCN-IPI and SDmax (the distance between the two lesions farthest apart, normalized to the patient's body surface area), showed remarkable predictive capabilities with a high area under the curve (AUC), sensitivity, specificity, PPV, NPV, accuracy, and significant net benefit of the AUC (NB-AUC). Although no significant differences were observed between the combined and stacking models in terms of the AUC in either the training cohort (AUC: 0.992 vs. 0.985, p = 0.139) or the testing cohort (AUC: 0.768 vs. 0.781, p = 0.668), the integrated model exhibited higher values for sensitivity, PPV, NPV, accuracy, and NB-AUC than the stacking model. Conclusion: Baseline PET radiomics could predict therapeutic efficacy in DLBCL after R-CHOP (-like) therapy, with improved predictive performance when incorporating clinical features and SDmax.

Prognostic Significance of Baseline Clinical and [68Ga]Ga-PSMA PET Derived Parameters on Biochemical Response, Overall Survival, and PSA Progression-Free Survival in Metastatic Castration-Resistant Prostate Cancer (mCRPC) Patients Undergoing [177Lu]Lu-PSMA Therapy

BACKGROUND

In this study, we sought to identify the clinical baseline characteristics and pre-therapy 68Ga-PSMA PET derived parameters that can have impact on PSA (biochemical) response, OS and PSA PFS in patients with metastatic castration-resistant prostate cancer (mCRPC) who undergo RLT with [177Lu]Lu-PSMA-617.

METHODS

Various pre-treatment clinical and PSMA PET derived parameters were gathered and computed. We used PSA response as the criteria for more than a 50% decrease in PSA level, and OS and PSA PFS as endpoints. We assessed the collected parameters in relation to PSA response. Additionally, we employed univariable Cox regression and Kaplan-Meier analysis with log rank to evaluate the influence of the parameters on OS and PFS.

RESULTS

A total of 125 mCRPC patients were included in this study. The median age was 68 years (range: 49-89). Among the cases, 77 patients (62%) showed PSARS, while 48 patients (38%) did not show PSA response. The median OS was 14 months (range: 1-60), and the median PSA-PFS was 10 months (range: 1-56). Age, prior history of chemotherapy, and SUVmax had a significant impact on PSA response (p<0.05). PSA response, RBC count, hemoglobin, hematocrit, neutrophil to lymphocyte ratio (NLR), alkaline phosphatase (ALP), number of metastases, wbPSMA-TV, and wbTL-PSMA significantly affected OS. GS, platelet count, NLR, and number of metastases were found to have a significant impact on PSA PFS.

CONCLUSION

We have identified several baseline clinical and PSMA PET derived parameters that can serve as prognostic factors for predicting PSA response, OS, and PSA PFS after RLT. Based on the findings, we believe that these clinical baseline characteristics can assist nuclear medicine specialists in identifying RLT responders who have long-term survival and PFS.

CT-based body composition parameters predict the loss of response to infliximab in patients with Crohn's disease

OBJECTIVE

Infliximab is a first-line biologic agent for the treatment of Crohn's disease (CD), in which loss of response (LOR) remains a challenge in the treatment of patients with CD. The study aimed to explore the association between body composition parameters and LOR to infliximab in CD patients.

METHODS

118 patients with CD admitted to the First Affiliated Hospital of Wenzhou Medical University and treated with infliximab from June 2015 to December 2021 were retrospectively enrolled. The body composition of patients was analyzed by computed tomography (CT). The primary outcome measure was the one-year LOR. Patients were divided into the Remission group and the LOR group to analyze the association between body composition parameters and the LOR to infliximab.

RESULTS

The rate of sarcopenia in the LOR group was higher than in the Remission group (83.7% vs. 60.0%, P=0.008). Multivariate analysis showed that females had a lower risk of sarcopenia than males (OR=0.30, 95% CI 0.11-0.81, P =0.017); BMI was significantly associated with sarcopenia (OR=0.68, 95% CI 0.56-0.83, P <0.001); L1 CD and L2 CD had a lower risk of sarcopenia than L3 CD (OR=0.29, 95% CI 0.10-0.83, P =0.021; OR=0.25, 95% CI 0.07-0.87, P=0.028).

CONCLUSIONS

Sarcopenia was identified as a risk factor for developing LOR in infliximab-treated patients.

International Benchmark for Total Metabolic Tumor Volume Measurement in Baseline 18F-FDG PET/CT of Lymphoma Patients: A Milestone Toward Clinical Implementation

Total metabolic tumor volume (TMTV) is prognostic in lymphoma. However, cutoff values for risk stratification vary markedly, according to the tumor delineation method used. We aimed to create a standardized TMTV benchmark dataset allowing TMTV to be tested and applied as a reproducible biomarker. Methods: Sixty baseline 18F-FDG PET/CT scans were identified with a range of disease distributions (20 follicular, 20 Hodgkin, and 20 diffuse large B-cell lymphoma). TMTV was measured by 12 nuclear medicine experts, each analyzing 20 cases split across subtypes, with each case processed by 3-4 readers. LIFEx or ACCURATE software was chosen according to reader preference. Analysis was performed stepwise: TMTV1 with automated preselection of lesions using an SUV of at least 4 and a volume of at least 3 cm3 with single-click removal of physiologic uptake; TMTV2 with additional removal of reactive bone marrow and spleen with single clicks; TMTV3 with manual editing to remove other physiologic uptake, if required; and TMTV4 with optional addition of lesions using mouse clicks with an SUV of at least 4 (no volume threshold). Results: The final TMTV (TMTV4) ranged from 8 to 2,288 cm3, showing excellent agreement among all readers in 87% of cases (52/60) with a difference of less than 10% or less than 10 cm3 In 70% of the cases, TMTV4 equaled TMTV1, requiring no additional reader interaction. Differences in the TMTV4 were exclusively related to reader interpretation of lesion inclusion or physiologic high-uptake region removal, not to the choice of software. For 5 cases, large TMTV differences (>25%) were due to disagreement about inclusion of diffuse splenic uptake. Conclusion: The proposed segmentation method enabled highly reproducible TMTV measurements, with minimal reader interaction in 70% of the patients. The inclusion or exclusion of diffuse splenic uptake requires definition of specific criteria according to lymphoma subtype. The publicly available proposed benchmark allows comparison of study results and could serve as a reference to test improvements using other segmentation approaches.

Coefficient of variation and texture analysis of 18F-FDG PET/CT images for the prediction of outcome in patients with multiple myeloma

In multiple myeloma (MM) bone marrow infiltration by monoclonal plasma cells can occur in both focal and diffuse manner, making staging and prognosis rather difficult. The aim of our study was to test whether texture analysis of 18 F-2-deoxy-d-glucose positron emission tomography/computed tomography (18F-FDG PET/CT) images can predict survival in MM patients. Forty-six patients underwent 18 F-FDG-PET/CT before treatment. We used an automated contouring program for segmenting the hottest focal lesion (FL) and a lumbar vertebra for assessing diffuse bone marrow involvement (DI). Maximum standardized uptake value (SUVmax), mean standardized uptake value (SUVmean) and texture features such as Coefficient of variation (CoV), were obtained from 46 FL and 46 DI. After a mean follow-up of 51 months, 24 patients died of myeloma and were compared to the 22 survivors. At univariate analysis, FL SUVmax (p = 0.0453), FL SUVmean (p = 0.0463), FL CoV (p = 0.0211) and DI SUVmax (p = 0.0538) predicted overall survival (OS). At multivariate analysis only FL CoV and DI SUVmax were retained in the model (p = 0.0154). By Kaplan-Meier method and log-rank testing, patients with FL CoV below the cut-off had significantly better OS than those with FL CoV above the cut-off (p = 0.0003), as well as patients with DI SUVmax below the threshold versus those with DI SUVmax above the threshold (p = 0.0006). Combining FL CoV and DI SUVmax by using their respective cut-off values, a statistically significant difference was found between the resulting four survival curves (p = 0.0001). Indeed, patients with both FL CoV and DI SUVmax below their respective cut-off values showed the best prognosis. Conventional and texture parameters derived from 18F-FDG PET/CT analysis can predict survival in MM patients by assessing the heterogeneity and aggressiveness of both focal and diffuse infiltration.

Challenges and limitations in applying radiomics to PET imaging: Possible opportunities and avenues for research

Radiomics, the high-throughput extraction of quantitative imaging features from medical images, holds immense potential for advancing precision medicine in oncology and beyond. While radiomics applied to positron emission tomography (PET) imaging offers unique insights into tumor biology and treatment response, it is imperative to elucidate the challenges and constraints inherent in this domain to facilitate their translation into clinical practice. This review examines the challenges and limitations of applying radiomics to PET imaging, synthesizing findings from the last five years (2019-2023) and highlights the significance of addressing these challenges to realize the full clinical potential of radiomics in oncology and molecular imaging. A comprehensive search was conducted across multiple electronic databases, including PubMed, Scopus, and Web of Science, using keywords relevant to radiomics issues in PET imaging. Only studies published in peer-reviewed journals were eligible for inclusion in this review. Although many studies have highlighted the potential of radiomics in predicting treatment response, assessing tumor heterogeneity, enabling risk stratification, and personalized therapy selection, various challenges regarding the practical implementation of the proposed models still need to be addressed. This review illustrates the challenges and limitations of radiomics in PET imaging across various cancer types, encompassing both phantom and clinical investigations. The analyzed studies highlight the importance of reproducible segmentation methods, standardized pre-processing and post-processing methodologies, and the need to create large multicenter studies registered in a centralized database to promote the continuous validation and clinical integration of radiomics into PET imaging.

Amino-acid PET as a prognostic tool after post Stupp protocol temozolomide therapy in high-grade glioma patients

PURPOSE

This study aimed to evaluate the prognostic performance of amino-acid PET in high-grade gliomas (HGG) patients at the time of temozolomide (TMZ) treatment discontinuation, after the Stupp protocol.

METHODS

The analysis included consecutive HGG patients with dynamic [18F]FDOPA PET imaging within 3 months of the end of TMZ therapy, post-Stupp protocol. Static and dynamic PET parameters, responses to RANO criteria for MRI and clinical and histo-molecular factors were correlated to progression-free (PFS).

RESULTS

Thirty-two patients (59.4 [54.0;67.6] years old, 13 (41%) women) were included. Static PET parameters peak tumor-to-background ratio and metabolic tumor volume (respective thresholds of 1.9 and 1.5 mL) showed the best 84% accuracies for predicting PFS at 6 months (p = 0.02). These static PET parameters were also independent predictor of PFS in multivariate analysis (p ≤ 0.05).

CONCLUSION

In HGG patients having undergone a Stupp protocol, the absence of significant PET uptake after TMZ constitutes a favorable prognostic factor.

The value of PET/CT radiomic texture analysis of primary mass and mediastinal lymph node on survival in patients with non-small cell lung cancer

OBJECTIVE

This study was designed to determine the potential prognostic value of radiomic texture analysis and metabolic-volumetric parameters obtained from positron emission tomography (PET) in primary mass and metastatic hilar/mediastinal lymph nodes in stage 2-3 non-small cell lung cancer (NSCLC).

METHODS

Images of patients diagnosed with stage 2-3 NSCLC who underwent 18F-FDG PET/CT imaging for staging up to 4 weeks before the start of treatment were evaluated using LIFEx software. Volume of interest (VOI) was generated from the primary tumor and metastatic lymph node separately, and volumetric and textural features were obtained from these VOIs. The relationship between the parameters obtained from PET of primary mass and the metastatic hilar/mediastinal lymph nodes with overall survival (OS) and progression-free survival (PFS) was analyzed.

RESULTS

When radiomic features, gender and stage obtained from lymph nodes were evaluated by Cox regression analysis; GLCM_correlation (p: 0.033, HR: 4,559, 1.660-12.521, 95% CI), gender and stage were determined as prognostic factors predicting OS. In predicting PFS; stage, smoking and lymph node MTV (p: 0.033, HR: 1.008, 1.001-1.016, 95% CI) were determined as prognostic factors. However, the radiomic feature of the primary tumor could not show a significant relationship with either OS or PFS.

CONCLUSIONS

In a retrospective cohort of NSCLC patients with Stage 2 and 3 disease, volumetric and radiomic texture characteristics obtained from metastatic lymph nodes were associated with PFS and OS. Tumor heterogeneity, defined by radiomic texture features of 18 F-FDG PET/CT images, may provide complementary prognostic value in NSCLC.

Radiomics of Intrahepatic Cholangiocarcinoma and Peritumoral Tissue Predicts Postoperative Survival: Development of a CT-Based Clinical-Radiomic Model

BACKGROUND

For many tumors, radiomics provided a relevant prognostic contribution. This study tested whether the computed tomography (CT)-based textural features of intrahepatic cholangiocarcinoma (ICC) and peritumoral tissue improve the prediction of survival after resection compared with the standard clinical indices.

METHODS

All consecutive patients affected by ICC who underwent hepatectomy at six high-volume centers (2009-2019) were considered for the study. The arterial and portal phases of CT performed fewer than 60 days before surgery were analyzed. A manual segmentation of the tumor was performed (Tumor-VOI). A 5-mm volume expansion then was applied to identify the peritumoral tissue (Margin-VOI).

RESULTS

The study enrolled 215 patients. After a median follow-up period of 28 months, the overall survival (OS) rate was 57.0%, and the progression-free survival (PFS) rate was 34.9% at 3 years. The clinical predictive model of OS had a C-index of 0.681. The addition of radiomic features led to a progressive improvement of performances (C-index of 0.71, including the portal Tumor-VOI, C-index of 0.752 including the portal Tumor- and Margin-VOI, C-index of 0.764, including all VOIs of the portal and arterial phases). The latter model combined clinical variables (CA19-9 and tumor pattern), tumor indices (density, homogeneity), margin data (kurtosis, compacity, shape), and GLRLM indices. The model had performance equivalent to that of the postoperative clinical model including the pathology data (C-index of 0.765). The same results were observed for PFS.

CONCLUSIONS

The radiomics of ICC and peritumoral tissue extracted from preoperative CT improves the prediction of survival. Both the portal and arterial phases should be considered. Radiomic and clinical data are complementary and achieve a preoperative estimation of prognosis equivalent to that achieved in the postoperative setting.

Combining PSMA-PET and PROMISE to re-define disease stage and risk in patients with prostate cancer: a multicentre retrospective study

BACKGROUND

Prostate-specific membrane antigen (PSMA)-PET was introduced into clinical practice in 2012 and has since transformed the staging of prostate cancer. Prostate Cancer Molecular Imaging Standardized Evaluation (PROMISE) criteria were proposed to standardise PSMA-PET reporting. We aimed to compare the prognostic value of PSMA-PET by PROMISE (PPP) stage with established clinical nomograms in a large prostate cancer dataset with follow-up data for overall survival.

METHODS

In this multicentre retrospective study, we used data from patients of any age with histologically proven prostate cancer who underwent PSMA-PET at the University Hospitals in Essen, Münster, Freiburg, and Dresden, Germany, between Oct 30, 2014, and Dec 27, 2021. We linked a subset of patient hospital records with patient data, including mortality data, from the Cancer Registry North-Rhine Westphalia, Germany. Patients from Essen University Hospital were randomly assigned to the development or internal validation cohorts (2:1). Patients from Münster, Freiburg, and Dresden University Hospitals were included in an external validation cohort. Using the development cohort, we created quantitative and visual PPP nomograms based on Cox regression models, assessing potential PPP predictors for overall survival, with least absolute shrinkage and selection operator penalty for overall survival as the primary endpoint. Performance was measured using Harrell's C-index in the internal and external validation cohorts and compared with established clinical risk scores (International Staging Collaboration for Cancer of the Prostate [STARCAP], European Association of Urology [EAU], and National Comprehensive Cancer Network [NCCN] risk scores) and a previous nomogram defined by Gafita et al (hereafter referred to as GAFITA) using receiver operating characteristic (ROC) curves and area under the ROC curve (AUC) estimates.

FINDINGS

We analysed 2414 male patients (1110 included in the development cohort, 502 in the internal cohort, and 802 in the external validation cohort), among whom 901 (37%) had died as of data cutoff (June 30, 2023; median follow-up of 52·9 months [IQR 33·9-79·0]). Predictors in the quantitative PPP nomogram were locoregional lymph node metastases (molecular imaging N2), distant metastases (extrapelvic nodal metastases, bone metastases [disseminated or diffuse marrow involvement], and organ metastases), tumour volume (in L), and tumour mean standardised uptake value. Predictors in the visual PPP nomogram were distant metastases (extrapelvic nodal metastases, bone metastases [disseminated or diffuse marrow involvement], and organ metastases) and total tumour lesion count. In the internal and external validation cohorts, C-indices were 0·80 (95% CI 0·77-0·84) and 0·77 (0·75-0·78) for the quantitative nomogram, respectively, and 0·78 (0·75-0·82) and 0·77 (0·75-0·78) for the visual nomogram, respectively. In the combined development and internal validation cohort, the quantitative PPP nomogram was superior to STARCAP risk score for patients at initial staging (n=139 with available staging data; AUC 0·73 vs 0·54; p=0·018), EAU risk score at biochemical recurrence (n=412; 0·69 vs 0·52; p<0·0001), and NCCN pan-stage risk score (n=1534; 0·81 vs 0·74; p<0·0001) for the prediction of overall survival, but was similar to GAFITA nomogram for metastatic hormone-sensitive prostate cancer (mHSPC; n=122; 0·76 vs 0·72; p=0·49) and metastatic castration-resistant prostate cancer (mCRPC; n=270; 0·67 vs 0·75; p=0·20). The visual PPP nomogram was superior to EAU at biochemical recurrence (n=414; 0·64 vs 0·52; p=0·0004) and NCCN across all stages (n=1544; 0·79 vs 0·73; p<0·0001), but similar to STARCAP for initial staging (n=140; 0·56 vs 0·53; p=0·74) and GAFITA for mHSPC (n=122; 0·74 vs 0·72; p=0·66) and mCRPC (n=270; 0·71 vs 0·75; p=0·23).

INTERPRETATION

Our PPP nomograms accurately stratify high-risk and low-risk groups for overall survival in early and late stages of prostate cancer and yield equal or superior prediction accuracy compared with established clinical risk tools. Validation and improvement of the nomograms with long-term follow-up is ongoing (NCT06320223).

FUNDING

Cancer Registry North-Rhine Westphalia.

Semi-supervised learning towards automated segmentation of PET images with limited annotations: application to lymphoma patients

Manual segmentation poses a time-consuming challenge for disease quantification, therapy evaluation, treatment planning, and outcome prediction. Convolutional neural networks (CNNs) hold promise in accurately identifying tumor locations and boundaries in PET scans. However, a major hurdle is the extensive amount of supervised and annotated data necessary for training. To overcome this limitation, this study explores semi-supervised approaches utilizing unlabeled data, specifically focusing on PET images of diffuse large B-cell lymphoma (DLBCL) and primary mediastinal large B-cell lymphoma (PMBCL) obtained from two centers. We considered 2-[18F]FDG PET images of 292 patients PMBCL (n = 104) and DLBCL (n = 188) (n = 232 for training and validation, and n = 60 for external testing). We harnessed classical wisdom embedded in traditional segmentation methods, such as the fuzzy clustering loss function (FCM), to tailor the training strategy for a 3D U-Net model, incorporating both supervised and unsupervised learning approaches. Various supervision levels were explored, including fully supervised methods with labeled FCM and unified focal/Dice loss, unsupervised methods with robust FCM (RFCM) and Mumford-Shah (MS) loss, and semi-supervised methods combining FCM with supervised Dice loss (MS + Dice) or labeled FCM (RFCM + FCM). The unified loss function yielded higher Dice scores (0.73 ± 0.11; 95% CI 0.67-0.8) than Dice loss (p value < 0.01). Among the semi-supervised approaches, RFCM + αFCM (α = 0.3) showed the best performance, with Dice score of 0.68 ± 0.10 (95% CI 0.45-0.77), outperforming MS + αDice for any supervision level (any α) (p < 0.01). Another semi-supervised approach with MS + αDice (α = 0.2) achieved Dice score of 0.59 ± 0.09 (95% CI 0.44-0.76) surpassing other supervision levels (p < 0.01). Given the time-consuming nature of manual delineations and the inconsistencies they may introduce, semi-supervised approaches hold promise for automating medical imaging segmentation workflows.

Towards improved diagnosis: radiomics and quantitative biomarkers in 18 F-PSMA-1007 and 18 F-fluorocholine PET/CT for prostate cancer recurrence

OBJECTIVE

This study compared the radiomic features and quantitative biomarkers of 18 F-PSMA-1007 [prostate-specific membrane antigen (PSMA)] and 18 F-fluorocholine (FCH) PET/computed tomography (CT) in prostate cancer patients with biochemical recurrence (BCR) enrolled in the phase 3, prospective, multicenter BIO-CT-001 trial.

METHODS

A total of 106 patients with BCR, who had undergone primary definitive treatment for prostate cancer, were recruited to this prospective study. All patients underwent one PSMA and one FCH PET/CT examination in randomized order within 10 days. They were followed up for a minimum of 6 months. Pathology, prostate-specific antigen (PSA), PSA doubling time, PSA velocity, and previous or ongoing treatment were analyzed. Using LifeX software, standardized uptake value (SUV) maximum, SUV mean , PSMA and choline total volume (PSMA-TV/FCH-TV), and total lesion PSMA and choline (TL-PSMA/TL-FCH) of all identified metastatic lesions in both tracers were calculated.

RESULTS

Of the 286 lesions identified, the majority 140 (49%) were lymph node metastases, 118 (41.2%) were bone metastases and 28 lesions (9.8%) were locoregional recurrences of prostate cancer. The median SUV max value was significantly higher for 18 F-PSMA compared with FCH for all 286 lesions (8.26 vs. 4.99, respectively, P  < 0.001). There were statistically significant differences in median SUV mean , TL-PSMA/FCH, and PSMA/FCH-TV between the two radiotracers (4.29 vs. 2.92, 1.97 vs. 1.53, and 7.31 vs. 4.37, respectively, P  < 0.001). The correlation between SUV mean /SUV max and PSA level was moderate, both for 18 F-PSMA ( r  = 0.44, P  < 0.001; r  = 0.44, P  < 0.001) and FCH ( r  = 0.35, P  < 0.001; r  = 0.41, P  < 0.001). TL-PSMA/FCH demonstrated statistically significant positive correlations with both PSA level and PSA velocity for both 18 F-PSMA ( r  = 0.56, P  < 0.001; r  = 0.57, P  < 0.001) and FCH ( r  = 0.49, P  < 0.001; r  = 0.51, P  < 0.001). While patients who received hormone therapy showed higher median SUV max values for both radiotracers compared with those who did not, the difference was statistically significant only for 18 F-PSMA ( P  < 0.05).

CONCLUSION

Our analysis using both radiomic features and quantitative biomarkers demonstrated the improved performance of 18 F-PSMA-1007 compared with FCH in identifying metastatic lesions in prostate cancer patients with BCR.

A Novel Approach to Identifying Hibernating Myocardium Using Radiomics-Based Machine Learning

Background To assess the feasibility of a machine learning (ML) approach using radiomics features of perfusion defects on rest myocardial perfusion imaging (MPI) to detect the presence of hibernating myocardium. Methodology Data of patients who underwent 99mTc-sestamibi MPI and 18F-FDG PET/CT for myocardial viability assessment were retrieved. Rest MPI data were processed on ECToolbox, and polar maps were saved using the NFile PMap tool. The reference standard for defining hibernating myocardium was the presence of mismatched perfusion-metabolism defect with impaired myocardial contractility at rest. Perfusion defects on the polar maps were delineated with regions of interest (ROIs) after spatial resampling and intensity discretization. Replicable random sampling allocated 80% (257) of the perfusion defects of the patients from January 2017 to September 2022 to the training set and the remaining 20% (64) to the validation set. An independent dataset of perfusion defects from 29 consecutive patients from October 2022 to January 2023 was used as the testing set for model evaluation. One hundred ten first and second-order texture features were extracted for each ROI. After feature normalization and imputation, 14 best-ranked features were selected using a multistep feature selection process including the Logistic Regression and Fast Correlation-Based Filter. Thirteen supervised ML algorithms were trained with stratified five-fold cross-validation on the training set and validated on the validation set. The ML algorithms with a Log Loss of <0.688 and <0.672 in the cross-validation and validation steps were evaluated on the testing set. Performance matrices of the algorithms assessed included area under the curve (AUC), classification accuracy (CA), F1 score, precision, recall, and specificity. To provide transparency and interpretability, SHapley Additive exPlanations (SHAP) values were assessed and depicted as beeswarm plots. Results Two hundred thirty-nine patients (214 males; mean age 56 ± 11 years) were enrolled in the study. There were 371 perfusion defects (321 in the training and validation sets; 50 in the testing set). Based on the reference standard, 168 perfusion defects had hibernating myocardium (139 in the training and validation sets; 29 in the testing set). On cross-validation, six ML algorithms with Log Loss <0.688 had AUC >0.800. On validation, 10 ML algorithms had a Log Loss value <0.672, among which six had AUC >0.800. On model evaluation of the selected models on the unseen testing set, nine ML models had AUC >0.800 with Gradient Boosting Random Forest (xgboost) [GB RF (xgboost)] achieving the highest AUC of 0.860 and could detect the presence of hibernating myocardium in 21/29 (72.4%) perfusion defects with a precision of 87.5% (21/24), specificity 85.7% (18/21), CA 78.0% (39/50) and F1 Score 0.792. Four models depicted a clear pattern of model interpretability based on the beeswarm SHAP plots. These were GB RF (xgboost), GB (scikit-learn), GB (xgboost), and Random Forest. Conclusion Our study demonstrates the potential of ML in detecting hibernating myocardium using radiomics features extracted from perfusion defects on rest MPI images. This proof-of-concept underscores the notion that radiomics features capture nuanced information beyond what is perceptible to the human eye, offering promising avenues for improved myocardial viability assessment.

From code sharing to sharing of implementations: Advancing reproducible AI development for medical imaging through federated testing

BACKGROUND

The reproducibility crisis in AI research remains a significant concern. While code sharing has been acknowledged as a step toward addressing this issue, our focus extends beyond this paradigm. In this work, we explore "federated testing" as an avenue for advancing reproducible AI research and development especially in medical imaging. Unlike federated learning, where a model is developed and refined on data from different centers, federated testing involves models developed by one team being deployed and evaluated by others, addressing reproducibility across various implementations.

METHODS

Our study follows an exploratory design aimed at systematically evaluating the sources of discrepancies in shared model execution for medical imaging and outputs on the same input data, independent of generalizability analysis. We distributed the same model code to multiple independent centers, monitoring execution in different runtime environments while considering various real-world scenarios for pre- and post-processing steps. We analyzed deployment infrastructure by comparing the impact of different computational resources (GPU vs. CPU) on model performance. To assess federated testing in AI models for medical imaging, we performed a comparative evaluation across different centers, each with distinct pre- and post-processing steps and deployment environments, specifically targeting AI-driven positron emission tomography (PET) imaging segmentation. More specifically, we studied federated testing for an AI-based model for surrogate total metabolic tumor volume (sTMTV) segmentation in PET imaging: the AI algorithm, trained on maximum intensity projection (MIP) data, segments lymphoma regions and estimates sTMTV.

RESULTS

Our study reveals that relying solely on open-source code sharing does not guarantee reproducible results due to variations in code execution, runtime environments, and incomplete input specifications. Deploying the segmentation model on local and virtual GPUs compared to using Docker containers showed no effect on reproducibility. However, significant sources of variability were found in data preparation and pre-/post- processing techniques for PET imaging. These findings underscore the limitations of code sharing alone in achieving consistent and accurate results in federated testing.

CONCLUSION

Achieving consistently precise results in federated testing requires more than just sharing models through open-source code. Comprehensive pipeline sharing, including pre- and post-processing steps, is essential. Cloud-based platforms that automate these processes can streamline AI model testing across diverse locations. Standardizing protocols and sharing complete pipelines can significantly enhance the robustness and reproducibility of AI models.

Could CT Radiomic Analysis of Benign Adrenal Incidentalomas Suggest the Need for Further Endocrinological Evaluation?

We studied the application of CT texture analysis in adrenal incidentalomas with baseline characteristics of benignity that are highly suggestive of adenoma to find whether there is a correlation between the extracted features and clinical data. Patients with hormonal hypersecretion may require medical attention, even if it does not cause any symptoms. A total of 206 patients affected by adrenal incidentaloma were retrospectively enrolled and divided into non-functioning adrenal adenomas (NFAIs, n = 115) and mild autonomous cortisol secretion (MACS, n = 91). A total of 136 texture parameters were extracted in the unenhanced phase for each volume of interest (VOI). Random Forest was used in the training and validation cohorts to test the accuracy of CT textural features and cortisol-related comorbidities in identifying MACS patients. Twelve parameters were retained in the Random Forest radiomic model, and in the validation cohort, a high specificity (81%) and positive predictive value (74%) were achieved. Notably, if the clinical data were added to the model, the results did not differ. Radiomic analysis of adrenal incidentalomas, in unenhanced CT scans, could screen with a good specificity those patients who will need a further endocrinological evaluation for mild autonomous cortisol secretion, regardless of the clinical information about the cortisol-related comorbidities.

The role of brown adipose tissue in branched-chain amino acid clearance in people

Brown adipose tissue (BAT) in rodents appears to be an important tissue for the clearance of plasma branched-chain amino acids (BCAAs) contributing to improved metabolic health. However, the role of human BAT in plasma BCAA clearance is poorly understood. Here, we evaluate patients with prostate cancer who underwent positron emission tomography-computed tomography imaging after an injection of 18F-fluciclovine (L-leucine analog). Supraclavicular adipose tissue (AT; primary location of human BAT) has a higher net uptake rate for 18F-fluciclovine compared to subcutaneous abdominal and upper chest AT. Supraclavicular AT 18F-fluciclovine net uptake rate is lower in patients with obesity and type 2 diabetes. Finally, the expression of genes involved in BCAA catabolism is higher in the supraclavicular AT of healthy people with high BAT volume compared to those with low BAT volume. These findings support the notion that BAT can potentially function as a metabolic sink for plasma BCAA clearance in people.

Value of Dynamic Contrast-Enhanced MRI for Grade Group Prediction in Prostate Cancer: A Radiomics Pilot Study

RATIONALE AND OBJECTIVES

To determine the role of dynamic contrast-enhanced (DCE) MRI-radiomics in predicting the International Society of Urological Pathology Grade Group (ISUP-GG) in therapy-naïve prostate cancer (PCa) patients.

MATERIALS AND METHODS

In this ethics review board-approved retrospective study on two prospective clinical trials between 2017 and 2020, 73 men with suspected/confirmed PCa were included. All participants underwent multiparametric MRI. On MRI, dominant lesions (per PI-RADS) were identified. DCE-MRI radiomic features were extracted from the segmented volumes following the image biomarker standardisation initiative (IBSI) guidelines through 14 time points. Histopathology evaluation on the cognitive-fusion targeted biopsies was set as the reference standard. Univariate regression was done to evaluate potential predictors across all calculated features. Random forest imputation was used for multivariate modelling.

RESULTS

73 index lesions were reviewed. Histopathology revealed 28, 16, 13 and 16 lesions with ISUP-GG-Negative/1/2, ISUP-GG-3, ISUP-GG-4 and ISUP-GG-5, respectively. From the extracted features, total lesion enhancement (TLE), minimum enhancement intensity and Grey-Level Run Length Matrix (GLRLM) were the most significantly different parameters among ISUP-GGs (Neg/1/2 vs 3/4 vs 5). 16 features with significant cross-sectional associations with ISUP-GGs entered the multivariate analysis. The final DCE partitioning model used only four features (lesion sphericity, TLE, GLRLM and Grey-Level Zone Length Matrix). For the binarized diagnosis (ISUP-GG≤2 vs ISUP-GG>2), the accuracy reached 81%.

CONCLUSION

DCE-MRI radiomics might be used as a non-invasive tool for aiding pathological grade group prediction in therapy-naïve PCa patients, potentially adding complementary information to PI-RADS for supporting tailored diagnostic pathways and treatment planning.

RadShap: An Explanation Tool for Highlighting the Contributions of Multiple Regions of Interest to the Prediction of Radiomic Models

Explaining the decisions made by a radiomic model is of significant interest, as it can provide valuable insights into the information learned by complex models and foster trust in well-performing ones, thereby facilitating their clinical adoption. Promising radiomic approaches that aggregate information from multiple regions within an image currently lack suitable explanation tools that could identify the regions that most significantly influence their decisions. Here we present a model- and modality-agnostic tool (RadShap, https://github.com/ncaptier/radshap), based on Shapley values, that explains the predictions of multiregion radiomic models by highlighting the contribution of each individual region. Methods: The explanation tool leverages Shapley values to distribute the aggregative radiomic model's output among all the regions of interest of an image, highlighting their individual contribution. RadShap was validated using a retrospective cohort of 130 patients with advanced non-small cell lung cancer undergoing first-line immunotherapy. Their baseline PET scans were used to build 1,000 synthetic tasks to evaluate the degree of alignment between the tool's explanations and our data generation process. RadShap's potential was then illustrated through 2 real case studies by aggregating information from all segmented tumors: the prediction of the progression-free survival of the non-small cell lung cancer patients and the classification of the histologic tumor subtype. Results: RadShap demonstrated strong alignment with the ground truth, with a median frequency of 94% for consistently explained predictions in the synthetic tasks. In both real-case studies, the aggregative models yielded superior performance to the single-lesion models (average [±SD] time-dependent area under the receiver operating characteristic curve was 0.66 ± 0.02 for the aggregative survival model vs. 0.55 ± 0.04 for the primary tumor survival model). The tool's explanations provided relevant insights into the behavior of the aggregative models, highlighting that for the classification of the histologic subtype, the aggregative model used information beyond the biopsy site to correctly classify patients who were initially misclassified by a model focusing only on the biopsied tumor. Conclusion: RadShap aligned with ground truth explanations and provided valuable insights into radiomic models' behaviors. It is implemented as a user-friendly Python package with documentation and tutorials, facilitating its smooth integration into radiomic pipelines.

Radiomics for clinical decision support in radiation oncology

Radiomics is a promising tool for the development of quantitative biomarkers to support clinical decision-making. It has been shown to improve the prediction of response to treatment and outcome in different settings, particularly in the field of radiation oncology by optimising the dose delivery solutions and reducing the rate of radiation-induced side effects, leading to a fully personalised approach. Despite the promising results offered by radiomics at each of these stages, standardised methodologies, reproducibility and interpretability of results are still lacking, limiting the potential clinical impact of these tools. In this review, we briefly describe the principles of radiomics and the most relevant applications of radiomics at each stage of cancer management in the framework of radiation oncology. Furthermore, the integration of radiomics into clinical decision support systems is analysed, defining the challenges and offering possible solutions for translating radiomics into a clinically applicable tool.

Inter-equipment validation of PET-based radiomics for predicting EGFR mutation statuses in patients with non-small cell lung cancer

AIM

To validate the inter-equipment generality of the radiomics based on PET images to predict the EGFR mutation status of patients with non-small cell lung cancer.

MATERIALS AND METHODS

Patients were retrospectively collected in the departments of nuclear medicine of Heyi branch (Siemens equipment) and East branch (General Electric (GE) equipment) of the first affiliated hospital of Zhengzhou university. 5 predicting logistic regression models were established. The 1st one was trained and tested by the GE dataset; The 2nd one was trained and tested by the Siemens dataset; The 3rd one was trained and tested by the mixed dataset consisting of GE and Siemens. The 4th one was trained by GE and tested by Siemens; The 5th one was trained by Siemens and tested by GE.

RESULTS

For the 1st ∼ 5th models, the mean values of AUCs for training/testing datasets were 0.78/0.73, 0.74/0.72, 0.75/0.70, 0.74/0.65 and 0.68/0.63, respectively.

CONCLUSION

The AUCs of the models trained and tested on the datasets from the same equipment were higher than those for different equipment. The inter-equipment generality of the radiomics was not good enough in clinical practice.

MRI radiomics nomogram integrating postoperative adjuvant treatments in recurrence risk prediction for patients with early-stage cervical cancer

BACKGROUND AND PURPOSE

Adjuvant treatments are valuable to decrease the recurrence rate and improve survival for early-stage cervical cancer patients (ESCC), Therefore, recurrence risk evaluation is critical for the choice of postoperative treatment. A magnetic resonance imaging (MRI) based radiomics nomogram integrating postoperative adjuvant treatments was constructed and validated externally to improve the recurrence risk prediction for ESCC.

MATERIAL AND METHODS

212 ESCC patients underwent surgery and adjuvant treatments from three centers were enrolled and divided into the training, internal validation, and external validation cohorts. Their clinical data, pretreatment T2-weighted images (T2WI) were retrieved and analyzed. Radiomics models were constructed using machine learning methods with features extracted and screen from sagittal and axial T2WI. A nomogram for recurrence prediction was build and evaluated using multivariable logistic regression analysis integrating radiomic signature and adjuvant treatments.

RESULTS

A total of 8 radiomic features were screened out of 1020 extracted features. The extreme gradient boosting (XGboost) model based on MRI radiomic features performed best in recurrence prediction with an area under curve (AUC) of 0.833, 0.822 in the internal and external validation cohorts, respectively. The nomogram integrating radiomic signature and clinical factors achieved an AUC of 0.806, 0.718 in the internal and external validation cohorts, respectively, for recurrence risk prediction for ESCC.

CONCLUSION

In this study, the nomogram integrating T2WI radiomic signature and clinical factors is valuable to predict the recurrence risk, thereby allowing timely planning for effective treatments for ESCC with high risk of recurrence.

Radiomics-based machine learning in the differentiation of benign and malignant bowel wall thickening radiomics in bowel wall thickening

PURPOSE

To distinguish malignant and benign bowel wall thickening (BWT) by using computed tomography (CT) texture features based on machine learning (ML) models and to compare its success with the clinical model and combined model.

METHODS

One hundred twenty-two patients with BWT identified on contrast-enhanced abdominal CT and underwent colonoscopy were included in this retrospective study. Texture features were extracted from CT images using LifeX software. Feature selection and reduction were performed using the Least Absolute Shrinkage and Selection Operator (LASSO). Six radiomic features were selected with LASSO. In the clinical model, six features (age, gender, thickness, fat stranding, symmetry, and lymph node) were included. Six radiomic and six clinical features were used in the combined model. Classification was done using two machine learning algorithms: Support Vector Machine (SVM) and Logistic Regression (LR). The data sets were divided into 80% training set and 20% test set. Then, training took place with all three datasets. The model's success was tested with the test set consisting of features not used during training.

RESULTS

In the training set, the combined model had the best performance with the area under the curve (AUC) value of 0.99 for SVM and 0.95 for LR. In the radiomic-derived model, the AUC value is 0.87 in SVM and 0.79 in LR. In the clinical model, SVM made this distinction with 0.95 AUC and LR with 0.92 AUC value. In the test set, the classifier with the highest success distinguishing malignant wall thickening is SVM in the radiomic-derived model with an AUC value of 0.90. In other models, the AUC value is in the range of 0.75-0.86, and the accuracy values are in the range of 0.72-0.84.

CONCLUSION

In conclusion, radiomic-based machine learning has shown high success in distinguishing malignant and benign BWT and may improve diagnostic accuracy compared to clinical features only. The results of our study may help ensure early diagnosis and treatment of colorectal cancers by facilitating the recognition of malignant BWT.

Defining the optimal segmentation method for measuring somatostatin receptor expressing tumor volume on 68 Ga-DOTATATE positron emission tomography/computed tomography to predict prognosis in patients with gastroenteropancreatic neuroendocrine tumors

OBJECTIVE

We aimed to compare different segmentation methods used to calculate prognostically valuable volumetric parameters, somatostatin receptor expressing tumor volume (SRETV), and total lesion somatostatin receptor expression (TLSRE), measured by 68 Ga-DOTATATE PET/CT and to find the optimal segmentation method to predict prognosis.

PATIENTS AND METHODS

Images of 34 patients diagnosed with gastroenteropancreatic neuroendocrine tumor (GEPNET) who underwent 68 Ga-DOTATATE PET/CT imaging were reanalyzed. Four different threshold-based methods (fixed relative threshold method, normal liver background threshold method, fixed absolute standardized uptake value (SUV) threshold method, and adaptive threshold method) were used to calculate SRETV and TLSRE values. SRETV of all lesions of a patient was summarized as whole body SRETV (WB-SRETV) and TLSRE of all lesions of a patient was computed as whole body TLSRE (WB-TLSRE).

RESULTS

WB-SRETVs calculated with all segmentation methods were statistically significantly associated with progression-free survival except WB-SRETV at which was calculated using adaptive threshold method. The fixed relative threshold methods calculated by using 45% (WB-SRETV 45% ) and 60% (WB-SRETV 60% ) of the SUV value as threshold respectively, were found to have statistically significant highest prognostic value (C-index = 0.704, CI = 0.622-0.786, P  = 0.007). Among WB-TLSRE parameters, WB-TLSRE 35% , WB-TLSRE 40% , and WB-TLSRE 50% had the highest prognostic value (C-index = 0.689, CI = 0.604-0.774, P  = 0.008).

CONCLUSION

The fixed relative threshold method was found to be the most effective and easily applicable method to measure SRETV on pretreatment 68 Ga-DOTATATE PET/CT to predict prognosis in GEPNET patients. WB-SRETV 45% (cutoff value of 11.8 cm 3 ) and WB-SRETV 60% (cutoff value of 6.3 cm 3 ) were found to be the strongest predictors of prognosis in GEPNET patients.

Development and validation of a machine learning-based 18F-fluorodeoxyglucose PET/CT radiomics signature for predicting gastric cancer survival

BACKGROUND

Survival prognosis of patients with gastric cancer (GC) often influences physicians' choice of their follow-up treatment. This study aimed to develop a positron emission tomography (PET)-based radiomics model combined with clinical tumor-node-metastasis (TNM) staging to predict overall survival (OS) in patients with GC.

METHODS

We reviewed the clinical information of a total of 327 patients with pathological confirmation of GC undergoing 18 F-fluorodeoxyglucose (18 F-FDG) PET scans. The patients were randomly classified into training (n = 229) and validation (n = 98) cohorts. We extracted 171 PET radiomics features from the PET images and determined the PET radiomics scores (RS) using the least absolute shrinkage and selection operator (LASSO) and random survival forest (RSF). A radiomics model, including PET RS and clinical TNM staging, was constructed to predict the OS of patients with GC. This model was evaluated for discrimination, calibration, and clinical usefulness.

RESULTS

On multivariate COX regression analysis, the difference between age, carcinoembryonic antigen (CEA), clinical TNM, and PET RS in GC patients was statistically significant (p < 0.05). A radiomics model was developed based on the results of COX regression. The model had the Harrell's concordance index (C-index) of 0.817 in the training cohort and 0.707 in the validation cohort and performed better than a single clinical model and a model with clinical features combined with clinical TNM staging. Further analyses showed higher PET RS in patients who were older (p < 0.001) and those who had elevated CEA (p < 0.001) and higher clinical TNM (p < 0.001). At different clinical TNM stages, a higher PET RS was associated with a worse survival prognosis.

CONCLUSIONS

Radiomics models based on PET RS, clinical TNM, and clinical features may provide new tools for predicting OS in patients with GC.

Metabolic tumour area: a novel prognostic indicator based on 18F-FDG PET/CT in patients with diffuse large B-cell lymphoma in the R-CHOP era

BACKGROUND

The metabolic tumour area (MTA) was found to be a promising predictor of prostate cancer. However, the role of MTA based on 18F-FDG PET/CT in diffuse large B-cell lymphoma (DLBCL) prognosis remains unclear. This study aimed to elucidate the prognostic significance of MTA and evaluate its incremental value to the National Comprehensive Cancer Network International Prognostic Index (NCCN-IPI) for DLBCL patients treated with first-line R-CHOP regimens.

METHODS

A total of 280 consecutive patients with newly diagnosed DLBCL and baseline 18F-FDG PET/CT data were retrospectively evaluated. Lesions were delineated via a semiautomated segmentation method based on a 41% SUVmax threshold to estimate semiquantitative metabolic parameters such as total metabolic tumour volume (TMTV) and MTA. Receiver operating characteristic (ROC) curve analysis was used to determine the optimal cut-off values. Progression-free survival (PFS) and overall survival (OS) were the endpoints that were used to evaluate the prognosis. PFS and OS were estimated via Kaplan‒Meier curves and compared via the log-rank test.

RESULTS

Univariate analysis revealed that patients with high MTA, high TMTV and NCCN-IPI ≥ 4 were associated with inferior PFS and OS (P < 0.0001 for all). Multivariate analysis indicated that MTA remained an independent predictor of PFS and OS [hazard ratio (HR), 2.506; 95% confidence interval (CI), 1.337-4.696; P = 0.004; and HR, 1.823; 95% CI, 1.005-3.310; P = 0.048], whereas TMTV was not. Further analysis using the NCCN-IPI model as a covariate revealed that MTA and NCCN-IPI were still independent predictors of PFS (HR, 2.617; 95% CI, 1.494-4.586; P = 0.001; and HR, 2.633; 95% CI, 1.650-4.203; P < 0.0001) and OS (HR, 2.021; 95% CI, 1.201-3.401; P = 0.008; and HR, 3.869; 95% CI, 1.959-7.640; P < 0.0001; respectively). Furthermore, MTA was used to separate patients with high NCCN-IPI risk scores into two groups with significantly different outcomes.

CONCLUSIONS

Pre-treatment MTA based on 18F-FDG PET/CT and NCCN-IPI were independent predictor of PFS and OS in DLBCL patients treated with R-CHOP. MTA has additional predictive value for the prognosis of patients with DLBCL, especially in high-risk patients with NCCN-IPI ≥ 4. In addition, the combination of MTA and NCCN-IPI may be helpful in further improving risk stratification and guiding individualised treatment options.

TRIAL REGISTRATION

This research was retrospectively registered with the Ethics Committee of the Third Affiliated Hospital of Soochow University, and the registration number was approval No. 155 (approved date: 31 May 2022).

Radiomic Gradient in Peritumoural Tissue of Liver Metastases: A Biomarker for Clinical Practice? Analysing Density, Entropy, and Uniformity Variations with Distance from the Tumour

The radiomic analysis of the tissue surrounding colorectal liver metastases (CRLM) enhances the prediction accuracy of pathology data and survival. We explored the variation of the textural features in the peritumoural tissue as the distance from CRLM increases. We considered patients with hypodense CRLMs >10 mm and high-quality computed tomography (CT). In the portal phase, we segmented (1) the tumour, (2) a series of concentric rims at a progressively increasing distance from CRLM (from one to ten millimetres), and (3) a cylinder of normal parenchyma (Liver-VOI). Sixty-three CRLMs in 51 patients were analysed. Median peritumoural HU values were similar to Liver-VOI, except for the first millimetre around the CRLM. Entropy progressively decreased (from 3.11 of CRLM to 2.54 of Liver-VOI), while uniformity increased (from 0.135 to 0.199, p < 0.001). At 10 mm from CRLM, entropy was similar to the Liver-VOI in 62% of cases and uniformity in 46%. In small CRLMs (≤30 mm) and responders to chemotherapy, normalisation of entropy and uniformity values occurred in a higher proportion of cases and at a shorter distance. The radiomic analysis of the parenchyma surrounding CRLMs unveiled a wide halo of progressively decreasing entropy and increasing uniformity despite a normal radiological aspect. Underlying pathology data should be investigated.

Amount of ascites impacts survival in patients with hepatocellular carcinoma undergoing transarterial chemoembolization advocating for volumetric assessment

Preliminary work has shown that portal hypertension plays a key role for the prognosis in patients with hepatocellular carcinoma (HCC) undergoing transarterial chemoembolization (TACE). Specifically, the presence of ascites appears to be a strong negative predictor for these patients. However, it remains unclear whether different ascites volumes influence prognosis. Therefore, the aim of this work was to investigate the influence of different ascites volumes on survival for patients with HCC undergoing TACE. A total of 327 treatment-naïve patients with HCC undergoing initial TACE at our tertiary care center between 2010 and 2020 were included. In patients with ascites, the fluid was segmented, and the volume quantified by slice-wise addition using contrast-enhanced CT imaging. Median overall survival (OS) was calculated and univariate and multivariate Cox regression analysis has been performed. Ascites was present in 102 (31.9%) patients. Ascites volume as continuous variable was significantly associated with an increased hazard ratio in univariate analysis (p < 0.001) and remained an independent predictor of impaired median OS in multivariate analysis (p < 0.001). Median OS without ascites was 17.1 months, and therefore significantly longer than in patients with ascites (6.4 months, p < 0.001). When subdivided into groups of low and high ascites volume in relation to the median ascites volume, patients with low ascites volume had a significantly longer median OS (8.6 vs 3.6 months, p < 0.001). Ascites in patients with HCC undergoing TACE is strongly associated with a poor prognosis. Our results show that not only the presence but also the amount of ascites is highly relevant. Therefore, true ascites volume as opportunistic quantitative biomarker is likely to impact clinical decision-making once automated solutions become available.