Exploring feature-based approaches in PET images for predicting cancer treatment outcomes

Machine learning-based model for predicting outcomes in cerebral hemorrhage patients with leukemia

BACKGROUND AND PURPOSE

Intracranial hemorrhage (ICH) in leukemia patients progresses rapidly with high mortality. Limited data are available on imaging studies in this population. The study aims to develop prediction models for 7-day and short-term mortality risk based on the non-contrast computed tomography (NCCT) image features.

METHODS

The NCCT image features of ICH in 135 leukemia patients between 2007-2023 were retrospectively extracted using manual assessment and radiomics methods. After multiple imputation of missing laboratory data, univariate logistic regression and least absolute shrinkage and selection operator (LASSO) were used for feature selection. Random forest models were built with comprehensive evaluation and ranking of feature importance.

RESULT

135 and 129 patients were included in the studies for 7-day and short-term prognostic models, respectively. The median age of all enrolled patients was 35 years, and there were 86 male patients (63.7 %). Clinical models (validation: AUC [area under the curve] = 0.78, AUPRC [area under the precision-recall curve] = 0.73; AUC = 0.84, AUPRC = 0.86), radiomics models (validation: AUC = 0.82, AUPRC = 0.78; AUC = 0.75, AUPRC = 0.77), and the combined models (validation: AUC = 0.84, AUPRC = 0.83; AUC = 0.87, AUPRC = 0.89) predicted 7-day and short-term mortality with good predictive efficacy. Clinical decision curve analysis showed that the combined models predicted 7-day and 30-day risk of death would be more beneficial than other models. Shape features contributed significantly more than semantic features in both radiomics models and combined models (93.3 %, 52.1 %, as well as 85.2 %,37.4 %, respectively) for 7-day and 30-day mortality.

CONCLUSIONS

Combined models constructed based on NCCT perform well in predicting the risk of 7-day and short-term mortality in ICH patients with leukemia. Shape features extracted by radiomics are important markers for modeling the prognosis.

Assessing Krenning's score on 68 Ga-DOTATATE PET-CT and miPSMA score on 68 Ga-PSMA-11 PET-CT in TENIS: a comparison with FDG PET/CT and examining the feasibility of targeted radionuclide therapy

OBJECTIVES

The objective of this study was to assess receptor expression in metastatic differentiated thyroid carcinoma patients with progressive elevated thyroglobulin and negative iodine scintigraphy, we used 68 Ga-DOTATATE [Gallium-68 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA)-octreotate (DOTATATE)] (Krenning's score) and 68 Ga-PSMA-11 (Gallium-68 prostate-specific membrane antigen-11) PET-computed tomography (CT) [molecular imaging prostate-specific membrane antigen (miPSMA) score]. Patients with Krenning's score 3 and above and miPSMA score 2 and above were considered to determine the incidence of patients, who would qualify for treatment with 177 Lu-DOTATATE/PSMA [Lutetium-177 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA)-octreotate (DOTATATE)/prostate-specific membrane antigen]-based therapy. In addition, we compared 68 Ga-DOTATATE and 68 Ga-PSMA-11 PET-CT with 2-deoxy-2-[F-18]fluoroglucose ( 18 F-FDG) PET-CT (using maximum standardized uptake value).

MATERIALS AND METHODS

A total of 74 patients with histopathologically proven metastatic differentiated thyroid carcinoma with thyroglobulin elevation and negative iodine scintigraphy syndrome were studied retrospectively. They all had 18 F-FDG, 68 Ga-DOTATATE, and 68 Ga-PSMA-11 PET-CT scans available for undertaking this analysis. The lesions detected by 68 Ga-DOTATATE and 68 Ga-PSMA-11 were evaluated using Krenning's and miPSMA scores. In addition, quantitative comparisons of maximum standardized uptake values for 68 Ga-DOTATATE and 68 Ga-PSMA-11, as well as with 18 F-FDG, were conducted.

RESULTS

Patient-wise analysis revealed positivity rates of 40.5% for 68 Ga-DOTATATE, 41.89% for 68 Ga-PSMA-11, and 75.67% for 18 F-FDG. Among the 74 patients, 14 (18.91%) were deemed eligible for 177 Lu-DOTATATE/PSMA-617 therapy based on Krenning's score of 3 and above both/either miPSMA score of 2 and above on 68 Ga-DOTATATE or 68 Ga-PSMA-11 PET-CT. Within this subgroup, seven out of 74 patients (9.45%) were eligible for 177 Lu-DOTATATE therapy, and nine out of 74 patients (12.16%) were eligible for 177 Lu-PSMA-targeted therapy. Four patients were eligible for both therapies.

CONCLUSION

Among thyroglobulin elevation and negative iodine scintigraphy patient's subgroup, 9.45% could qualify for 177 Lu-DOTATATE and 12.16% for 177 Lu-PSMA-617. Four were eligible for both therapies. Given the lack of effective therapies, this subset of patients warrants consideration for radionuclide therapy exploration.

Whole-tumoral metabolic heterogeneity in 18F-FDG PET/CT is a novel prognostic marker for neuroblastoma

BACKGROUND

Neuroblastoma (NB) is a highly heterogeneous tumor, and more than half of newly diagnosed NB are associated with extensive metastases. Accurately characterizing the heterogeneity of whole-body tumor lesions remains clinical challenge. This study aims to quantify whole-tumoral metabolic heterogeneity (WMH) derived from whole-body tumor lesions, and investigate the prognostic value of WMH in NB.

METHODS

We retrospectively enrolled 95 newly diagnosed pediatric NB patients in our department. Traditional semi-quantitative PET/CT parameters including the maximum standardized uptake value (SUVmax), the mean standardized uptake value (SUVmean), the peak standardized uptake value (SUVpeak), metabolic tumor volume (MTV) and total lesion glycolysis (TLG) were measured. These PET/CT parameters were expressed as PSUVmax, PSUVmean, PSUVpeak, PMTV, PTLG for primary tumor, WSUVmax, WSUVmean, WSUVpeak, WMTV, WTLG for whole-body tumor lesions. The metabolic heterogeneity was quantified using the areas under the curve of the cumulative SUV-volume histogram index (AUC-CSH index). Intra-tumoral metabolic heterogeneity (IMH) and WMH were extracted from primary tumor and whole-body tumor lesions, respectively. The outcome endpoints were overall survival (OS) and progression-free survival (PFS). Survival analysis was performed utilizing the univariate and multivariate Cox proportional hazards regression. The optimal cut-off values for metabolic parameters were obtained by receiver operating characteristic curve (ROC).

RESULTS

During follow up, 27 (28.4%) patients died, 21 (22.1%) patients relapsed and 47 (49.5%) patients remained progression-free survival, with a median follow-up of 35.0 months. In survival analysis, WMTV and WTLG were independent indicators of PFS, and WMH was an independent risk factor of PFS and OS. However, IMH only showed association with PFS and OS. In addition to metabolic parameters, the International Neuroblastoma Staging System (INSS) was identified as an independent risk factor for PFS, and neuron-specific enolase (NSE) served as an independent predictor of OS.

CONCLUSION

WMH was an independent risk factor for PFS and OS, suggesting its potential as a novel prognostic marker for newly diagnosed NB patients.

PET-based radiomic feature based on the cross-combination method for predicting the mid-term efficacy and prognosis in high-risk diffuse large B-cell lymphoma patients

Objectives

This study aims to develop 7×7 machine-learning cross-combinatorial methods for selecting and classifying radiomic features used to construct Radiomics Score (RadScore) of predicting the mid-term efficacy and prognosis in high-risk patients with diffuse large B-cell lymphoma (DLBCL).

Methods

Retrospectively, we recruited 177 high-risk DLBCL patients from two medical centers between October 2012 and September 2022 and randomly divided them into a training cohort (n=123) and a validation cohort (n=54). We finally extracted 110 radiomic features along with SUVmax, MTV, and TLG from the baseline PET. The 49 features selection-classification pairs were used to obtain the optimal LASSO-LASSO model with 11 key radiomic features for RadScore. Logistic regression was employed to identify independent RadScore, clinical and PET factors. These models were evaluated using receiver operating characteristic (ROC) curves and calibration curves. Decision curve analysis (DCA) was conducted to assess the predictive power of the models. The prognostic power of RadScore was assessed using cox regression (COX) and Kaplan-Meier plots (KM).

Results

177 patients (mean age, 63 ± 13 years,129 men) were evaluated. Multivariate analyses showed that gender (OR,2.760; 95%CI:1.196,6.368); p=0.017), B symptoms (OR,4.065; 95%CI:1.837,8.955; p=0.001), SUVmax (OR,2.619; 95%CI:1.107,6.194; p=0.028), and RadScore (OR,7.167; 95%CI:2.815,18.248; p<0.001) independently contributed to the risk factors for predicting mid-term outcome. The AUC values of the combined models in the training and validation groups were 0.846 and 0.724 respectively, outperformed the clinical model (0.714;0.556), PET based model (0.664; 0.589), NCCN-IPI model (0.523;0.406) and IPI model (0.510;0.412) in predicting mid-term treatment outcome. DCA showed that the combined model incorporating RadScore, clinical risk factors, and PET metabolic metrics has optimal net clinical benefit. COX indicated that the high RadScore group had worse prognosis and survival in progression-free survival (PFS) (HR, 2.1737,95%CI: 1.2983, 3.6392) and overall survival (OS) (HR,2.1356,95%CI: 1.2561, 3.6309) compared to the low RadScore group. KM survival analysis also showed the same prognosis prediction as Cox results.

Conclusion

The combined model incorporating RadScore, sex, B symptoms and SUVmax demonstrates a significant enhancement in predicting medium-term efficacy and prognosis in high-risk DLBCL patients. RadScore using 7×7 machine learning cross-combinatorial methods for selection and classification holds promise as a potential method for evaluating medium-term treatment outcome and prognosis in high-risk DLBCL patients.

Radiomics on spatial-temporal manifolds via Fokker-Planck dynamics

BACKGROUND

Delta radiomics is a high-throughput computational technique used to describe quantitative changes in serial, time-series imaging by considering the relative change in radiomic features of images extracted at two distinct time points. Recent work has demonstrated a lack of prognostic signal of radiomic features extracted using this technique. We hypothesize that this lack of signal is due to the fundamental assumptions made when extracting features via delta radiomics, and that other methods should be investigated.

PURPOSE

The purpose of this work was to show a proof-of-concept of a new radiomics paradigm for sparse, time-series imaging data, where features are extracted from a spatial-temporal manifold modeling the time evolution between images, and to assess the prognostic value on patients with oropharyngeal cancer (OPC).

METHODS

To accomplish this, we developed an algorithm to mathematically describe the relationship between two images acquired at timet = 0 $t = 0$ andt > 0 $t > 0$ . These images serve as boundary conditions of a partial differential equation describing the transition from one image to the other. To solve this equation, we propagate the position and momentum of each voxel according to Fokker-Planck dynamics (i.e., a technique common in statistical mechanics). This transformation is driven by an underlying potential force uniquely determined by the equilibrium image. The solution generates a spatial-temporal manifold (3 spatial dimensions + time) from which we define dynamic radiomic features. First, our approach was numerically verified by stochastically sampling dynamic Gaussian processes of monotonically decreasing noise. The transformation from high to low noise was compared between our Fokker-Planck estimation and simulated ground-truth. To demonstrate feasibility and clinical impact, we applied our approach to 18F-FDG-PET images to estimate early metabolic response of patients (n = 57) undergoing definitive (chemo)radiation for OPC. Images were acquired pre-treatment and 2-weeks intra-treatment (after 20 Gy). Dynamic radiomic features capturing changes in texture and morphology were then extracted. Patients were partitioned into two groups based on similar dynamic radiomic feature expression via k-means clustering and compared by Kaplan-Meier analyses with log-rank tests (p < 0.05). These results were compared to conventional delta radiomics to test the added value of our approach.

RESULTS

Numerical results confirmed our technique can recover image noise characteristics given sparse input data as boundary conditions. Our technique was able to model tumor shrinkage and metabolic response. While no delta radiomics features proved prognostic, Kaplan-Meier analyses identified nine significant dynamic radiomic features. The most significant feature was Gray-Level-Size-Zone-Matrix gray-level variance (p = 0.011), which demonstrated prognostic improvement over its corresponding delta radiomic feature (p = 0.722).

CONCLUSIONS

We developed, verified, and demonstrated the prognostic value of a novel, physics-based radiomics approach over conventional delta radiomics via data assimilation of quantitative imaging and differential equations.

Convoluted Neural Network for Detection of Clinically Significant Prostate Cancer on 68 Ga PSMA PET/CT Delayed Imaging by Analyzing Radiomic Features

Purpose

To assess the utility of convoluted neural network (CNN) in differentiating clinically significant and insignificant prostate cancer in patients with 68 Ga PSMA PET/CT-targeted prostate biopsy-proven prostate cancer.

Methods

In this retrospective study, 142 patients with clinical suspicion of prostate cancer were evaluated who underwent 68 Ga-PSMA PET/CT imaging followed by 68 Ga-PSMA PET/CT-targeted prostate biopsy from the PSMA-avid prostate lesion. Twenty patients with no PSMA-avid lesions were excluded. Local Image Features Extraction (LifeX) software was used to extract radiomic features (RF) from delayed 68 Ga-PSMA PET/CT images of 122 patients. LifeX failed to extract radiomic features in 24 patients, and the remaining 98 were evaluated. RFs were fed to an in-built CNN of the software for computation and results were achieved. Patients with Gleason Score ≥ 7 on histopathology were labeled clinically significant prostate cancer (csPCa). The diagnostic values of radiomic features were evaluated.

Results

The csPCa was revealed in 69/98 (70.4%) patients, and insignificant PCa was noticed in 29/98 (29.6%) patients. The software extracted 124 RF from the delayed 68 Ga-PSMA PET/CT images. The accuracy of the CNN was 80.7% to differentiate clinically significant and clinically insignificant prostate cancer, with an error percentage (E %) of 19.3%. The sensitivity, specificity, positive predictive, and negative predictive values were 90.3%, 57.7%, 83.6%, and 71.4%, respectively, to detect csPCa.

Conclusion

CNN is a feasible pre-biopsy screening tool for identifying clinically significant prostate cancer and can be used as an adjunct in the initial diagnosis and early treatment planning.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13139-023-00832-3.

A common [18F]-FDG PET radiomic signature to predict survival in patients with HPV-induced cancers

Locally advanced cervical cancer (LACC) and anal and oropharyngeal squamous cell carcinoma (ASCC and OPSCC) are mostly caused by oncogenic human papillomaviruses (HPV). In this paper, we developed machine learning (ML) models based on clinical, biological, and radiomic features extracted from pre-treatment fluorine-18-fluorodeoxyglucose positron emission tomography ([18F]-FDG PET) images to predict the survival of patients with HPV-induced cancers. For this purpose, cohorts from five institutions were used: two cohorts of patients treated for LACC including 104 patients from Gustave Roussy Campus Cancer (Center 1) and 90 patients from Leeds Teaching Hospitals NHS Trust (Center 2), two datasets of patients treated for ASCC composed of 66 patients from Institut du Cancer de Montpellier (Center 3) and 67 patients from Oslo University Hospital (Center 4), and one dataset of 45 OPSCC patients from the University Hospital of Zurich (Center 5). Radiomic features were extracted from baseline [18F]-FDG PET images. The ComBat technique was applied to mitigate intra-scanner variability. A modified consensus nested cross-validation for feature selection and hyperparameter tuning was applied on four ML models to predict progression-free survival (PFS) and overall survival (OS) using harmonized imaging features and/or clinical and biological variables as inputs. Each model was trained and optimized on Center 1 and Center 3 cohorts and tested on Center 2, Center 4, and Center 5 cohorts. The radiomic-based CoxNet model achieved C-index values of 0.75 and 0.78 for PFS and 0.76, 0.74, and 0.75 for OS on the test sets. Radiomic feature-based models had superior performance compared to the bioclinical ones, and combining radiomic and bioclinical variables did not improve the performances. Metabolic tumor volume (MTV)-based models obtained lower C-index values for a majority of the tested configurations but quite equivalent performance in terms of time-dependent AUCs (td-AUC). The results demonstrate the possibility of identifying common PET-based image signatures for predicting the response of patients with induced HPV pathology, validated on multi-center multiconstructor data.

From coarse to fine: a deep 3D probability volume contours framework for tumour segmentation and dose painting in PET images

With the increasing integration of functional imaging techniques like Positron Emission Tomography (PET) into radiotherapy (RT) practices, a paradigm shift in cancer treatment methodologies is underway. A fundamental step in RT planning is the accurate segmentation of tumours based on clinical diagnosis. Furthermore, novel tumour control methods, such as intensity modulated radiation therapy (IMRT) dose painting, demand the precise delineation of multiple intensity value contours to ensure optimal tumour dose distribution. Recently, convolutional neural networks (CNNs) have made significant strides in 3D image segmentation tasks, most of which present the output map at a voxel-wise level. However, because of information loss in subsequent downsampling layers, they frequently fail to precisely identify precise object boundaries. Moreover, in the context of dose painting strategies, there is an imperative need for reliable and precise image segmentation techniques to delineate high recurrence-risk contours. To address these challenges, we introduce a 3D coarse-to-fine framework, integrating a CNN with a kernel smoothing-based probability volume contour approach (KsPC). This integrated approach generates contour-based segmentation volumes, mimicking expert-level precision and providing accurate probability contours crucial for optimizing dose painting/IMRT strategies. Our final model, named KsPC-Net, leverages a CNN backbone to automatically learn parameters in the kernel smoothing process, thereby obviating the need for user-supplied tuning parameters. The 3D KsPC-Net exploits the strength of KsPC to simultaneously identify object boundaries and generate corresponding probability volume contours, which can be trained within an end-to-end framework. The proposed model has demonstrated promising performance, surpassing state-of-the-art models when tested against the MICCAI 2021 challenge dataset (HECKTOR).

A Machine Learning Approach Using FDG PET-Based Radiomics for Prediction of Tumor Mutational Burden and Prognosis in Stage IV Colorectal Cancer

INTRODUCTION

We assessed the performance of F-18 fluorodeoxyglucose positron emission tomography (FDG PET)-based radiomics for the prediction of tumor mutational burden (TMB) and prognosis using a machine learning (ML) approach in patients with stage IV colorectal cancer (CRC).

METHODS

Ninety-one CRC patients who underwent pretreatment FDG PET/computed tomography (CT) and palliative chemotherapy were retrospectively included. PET-based radiomics were extracted from the primary tumor on PET imaging using the software LIFEx. For feature selection, PET-based radiomics associated with TMB were selected by logistic regression analysis. The performances of seven ML algorithms to predict high TMB were compared by the area under the receiver's operating characteristic curves (AUCs) and validated by five-fold cross-validation. A PET radiomic score was calculated by averaging the z-score of each radiomic feature. The prognostic power of the PET radiomic score was assessed using Cox proportional hazards regression analysis.

RESULTS

Ten significant radiomic features associated with TMB were selected: surface-to-volume ratio, total lesion glycolysis, tumor volume, area, compacity, complexity, entropy, correlation, coarseness, and zone size non-uniformity. The k-nearest neighbors model obtained the good performance for prediction of high TMB (AUC: 0.791, accuracy: 0.814, sensitivity: 0.619, specificity: 0.871). On multivariable Cox regression analysis, the PET radiomic score (Hazard ratio = 4.498, 95% confidential interval = 1.024-19.759; p = 0.046) was a significant independent prognostic factor for OS.

CONCLUSIONS

This study demonstrates that PET-based radiomics are useful image biomarkers for the prediction of TMB status in stage IV CRC. PET radiomic score, which integrates significant radiomic features, has the potential to predict survival in stage IV CRC patients.

Role of Textural Analysis of Pretreatment 18F Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography in Response Prediction in Esophageal Carcinoma Patients

Introduction

Positron emission tomography/computed tomography (PET/CT) is routinely used for staging, response assessment, and surveillance in esophageal carcinoma patients. The aim of this study was to investigate whether textural features of pretreatment 18F-fluorodeoxyglucose (18F-FDG) PET/CT images can contribute to prognosis prediction in carcinoma oesophagus patients.

Materials and Methods

This is a retrospective study of 30 diagnosed carcinoma esophagus patients. These patients underwent pretreatment 18F-FDG PET/CT for staging. The images were processed in a commercially available textural analysis software. Region of interest was drawn over primary tumor with a 40% threshold and was processed further to derive 92 textural and radiomic parameters. These parameters were then compared between progression group and nonprogression group. The original dataset was subject separately to receiver operating curve analysis. Receiver operating characteristic (ROC) curves were used to identify the cutoff values for textural features with a P < 0.05 for statistical significance. Feature selection was done with principal component analysis. The selected features of each evaluator were subject to 4 machine-learning algorithms. The highest area under the curve (AUC) values was selected for 10 features.

Results

A retrospective study of 30 primary carcinoma esophagus patients was done. Patients were followed up after chemo-radiotherapy and they underwent follow-up PET/CT. On the basis of their response, patients were divided into progression group and nonprogression group. Among them, 15 patients showed disease progression and 15 patients were in the nonprogression group. Ten textural analysis parameters turned out to be significant in the prediction of disease progression. Cutoff values were calculated for these parameters according to the ROC curves, GLZLM_long zone emphasis (Gray Level Zone Length Matrix)_long zone emphasis (44.9), GLZLM_low gray level zone emphasis (0.006), GLZLM_short zone low gray level emphasis (0.0032), GLZLM_long zone low gray level emphasis (0.185), GLRLM_long run emphasis (Gray Level Run Length Matrix) (1.31), GLRLM_low gray level run emphasis (0.0058), GLRLM_short run low gray level emphasis (0.005496), GLRLM_long run low gray level emphasis (0.00727), NGLDM_Busyness (Neighborhood Gray Level Difference Matrix) (0.75), and gray level co-occurrence matrix_homogeneity (0.37). Feature selection by principal components analysis and feature classification by the K-nearest neighbor machine-learning model using independent training and test samples yielded the overall highest AUC.

Conclusions

Textural analysis parameters could provide prognostic information in carcinoma esophagus patients. Larger multicenter studies are needed for better clinical prognostication of these parameters.

Utility of pre-treatment 18 F-fluorodeoxyglucose PET radiomic analysis in assessing nodal involvement in cervical cancer

OBJECTIVE

Intratumor heterogeneity has prognostic value in cervical cancer, which can be depicted on 18 F-fluorodeoxyglucose ( 18 F-FDG) PET/computed tomography (PET/CT) and then quantitatively characterized by texture features. This study aimed to evaluate the discriminative performance and predictive ability of the texture features in determining lymph node involvement in cervical cancer.

METHODS

A total of 101 patients with newly diagnosed cervical cancer, who underwent pre-treatment whole-body 18 F-FDG PET/CT imaging were retrospectively recruited. Patients were categorized based on their nodal status. Thirty-five radiomic features together with the maximum standardized uptake value (SUVmax), metabolic tumor volume (MTV) and total lesion glycolysis (TLG) of the primary cervical tumors were extracted. Conventional indices were used to build logistic regression model and texture features were used to build random forest model. The performances for differentiating nodal status were assessed by receiver operating characteristic analysis.

RESULTS

Conventional PET indices were significantly higher in patients with nodal involvement compared to those without: SUVmax = 14.22 vs. 10.05; MTV = 57.02 vs. 28.73; TLG = 492.8 vs. 188.8 ( P  < 0.05). Nineteen radiomic features describing regional heterogeneity were significantly different between nodal involvements. Area under the curves of the models with conventional indices and PET texture features for discriminating nodal status were 0.72 and 0.76, respectively.

CONCLUSION

PET-derived radiomic features had moderate performance in discriminating nodal involvement in cervical cancer; and they did not outperform model based on conventional indices.

Assessment of the Association between Entropy in PET/CT and Response to Anti-PD-1/PD-L1 Monotherapy in Stage III or IV NSCLC

Anti-PD-1/PD-L1 therapy indications are broadened in non-small cell lung cancer (NSCLC) although immune checkpoint inhibitors (ICI) do not provide benefits for the entire population. Texture features based on positron emission tomography/computed tomography (PET/CT), especially entropy (based on a gray-level co-occurrence matrix (GLCM)), could be interesting as predictors in NSCLC. The aim of our retrospective study was to evaluate the association between GLCM-entropy and response to anti-PD-1/PD-L1 monotherapy at the first evaluation in stage III or IV NSCLC, comparing patients with progressive disease (PD) and non-progressive disease (non-PD). In total, 47 patients were included. Response Evaluation Criteria in Solid Tumors (RECIST 1.1) were used to evaluate the response to ICI treatment (nivolumab, pembrolizumab, or atezolizumab). At the first evaluation, 25 patients were PD and 22 were non-PD. GLCM-entropy was not predictive of response at the first evaluation. Furthermore, GLCM-entropy was not associated with progression-free survival (PFS) (p = 0.393) or overall survival (OS) (p = 0.220). Finally, GLCM-entropy measured in PET/CT performed before ICI initiation in stage III or IV NSCLC was not predictive of response at the first evaluation. However, this study demonstrates the feasibility of using texture parameters in routine clinical practice. The interest of measuring PET/CT texture parameters in NSCLC remains to be evaluated in larger prospective studies.

Radiomics Applications in Head and Neck Tumor Imaging: A Narrative Review

Recent advances in machine learning and artificial intelligence technology have ensured automated evaluation of medical images. As a result, quantifiable diagnostic and prognostic biomarkers have been created. We discuss radiomics applications for the head and neck region in this paper. Molecular characterization, categorization, prognosis and therapy recommendation are given special consideration. In a narrative manner, we outline the fundamental technological principles, the overall idea and usual workflow of radiomic analysis and what seem to be the present and potential challenges in normal clinical practice. Clinical oncology intends for all of this to ensure informed decision support for personalized and useful cancer treatment. Head and neck cancers present a unique set of diagnostic and therapeutic challenges. These challenges are brought on by the complicated anatomy and heterogeneity of the area under investigation. Radiomics has the potential to address these barriers. Future research must be interdisciplinary and focus on the study of certain oncologic functions and outcomes, with external validation and multi-institutional cooperation in order to achieve this.

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

PURPOSE

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

METHODS

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

CONCLUSION

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

A radiomic model to classify response to neoadjuvant chemotherapy in breast cancer

BACKGROUND

Medical image analysis has evolved to facilitate the development of methods for high-throughput extraction of quantitative features that can potentially contribute to the diagnostic and treatment paradigm of cancer. There is a need for further improvement in the accuracy of predictive markers of response to neo-adjuvant chemotherapy (NAC). The aim of this study was to develop a radiomic classifier to enhance current approaches to predicting the response to NAC breast cancer.

METHODS

Data on patients treated for breast cancer with NAC prior to surgery who had a pre-NAC dynamic contrast enhanced breast MRI were included. Response to NAC was assessed using the Miller-Payne system on the excised tumor. Tumor segmentation was carried out manually under the supervision of a consultant breast radiologist. Features were selected using least absolute shrinkage selection operator regression. A support vector machine learning model was used to classify response to NAC.

RESULTS

74 patients were included. Patients were classified as having a poor response to NAC (reduction in cellularity < 90%, n = 44) and an excellent response (> 90% reduction in cellularity, n = 30). 4 radiomics features (discretized kurtosis, NGDLM contrast, GLZLM_SZE and GLZLM_ZP) were identified as pertinent predictors of response to NAC. A SVM model using these features stratified patients into poor and excellent response groups producing an AUC of 0.75. Addition of estrogen receptor status improved the accuracy of the model with an AUC of 0.811.

CONCLUSION

This study identified a radiomic classifier incorporating 4 radiomics features to augment subtype based classification of response to NAC in breast cancer.

Can Persistent Homology Features Capture More Intrinsic Information about Tumors from 18F-Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography Images of Head and Neck Cancer Patients?

This study hypothesized that persistent homology (PH) features could capture more intrinsic information about the metabolism and morphology of tumors from ¹⁸F-fluorodeoxyglucose positron emission tomography (PET)/computed tomography (CT) images of patients with head and neck (HN) cancer than other conventional features. PET/CT images and clinical variables of 207 patients were selected from the publicly available dataset of the Cancer Imaging Archive. PH images were generated from persistent diagrams obtained from PET/CT images. The PH features were derived from the PH PET/CT images. The signatures were constructed in a training cohort from features from CT, PET, PH-CT, and PH-PET images; clinical variables; and the combination of features and clinical variables. Signatures were evaluated using statistically significant differences (p-value, log-rank test) between survival curves for low- and high-risk groups and the C-index. In an independent test cohort, the signature consisting of PH-PET features and clinical variables exhibited the lowest log-rank p-value of 3.30 × 10⁻⁵ and C-index of 0.80, compared with log-rank p-values from 3.52 × 10⁻² to 1.15 × 10⁻⁴ and C-indices from 0.34 to 0.79 for other signatures. This result suggests that PH features can capture the intrinsic information of tumors and predict prognosis in patients with HN cancer.

The potential of biology-guided radiation therapy in thoracic cancer: A preliminary treatment planning study

Purpose

We investigated the feasibility of biology-guided radiotherapy (BgRT), a technique that utilizes real-time positron emission imaging to minimize tumor motion uncertainties, to spare nearby organs at risk.

Methods

Volumetric modulated arc therapy (VMAT), intensity-modulated proton (IMPT) therapy, and BgRT plans were created for a paratracheal node recurrence (case 1; 60 Gy in 10 fractions) and a primary peripheral left upper lobe adenocarcinoma (case 2; 50 Gy in four fractions).

Results

For case 1, BgRT produced lower bronchus V40 values compared to VMAT and IMPT. For case 2, total lung V20 was lower in the BgRT case compared to VMAT and IMPT.

Conclusions

BgRT has the potential to reduce the radiation dose to proximal critical structures but requires further detailed investigation.

Can the combination of DWI and T2WI radiomics improve the diagnostic efficiency of cervical squamous cell carcinoma?

BACKGROUND

To investigate the value of MRI multi-sequence imaging model in differentiation of cervical squamous cell carcinoma (CSCC).

METHODS

A total of 104 CSCC patients confirmed with pathology were retrospectively enrolled. All patients underwent conventional MRI examination before treatment. The lesions were segmented using ITK-SNAP software manually and radiomics features were extracted by Artificial Intelligence Kit (AK) software. 396 tumor texture features were obtained and then the mRMR and Lasso algorithms were used to reduce the feature dimension. Three models including T2WI model, DWI model and Joint model (combined TWI and DWI) were constructed in training group and evaluated in validation group. and the receiver operator characteristics and calibration curve were used to evaluate the model performance.

RESULTS

The Joint model and T2WI model both showed a better diagnostic efficacy than single DWI model in differentiation of CSCC in training group (Joint model: AUC = 0.841; T2WI model: AUC = 0.804; DWI model: AUC = 0.732) and validation group (Joint model: AUC = 0.822; T2WI model: AUC = 0.791; DWI model: AUC = 0.724). But there was no statistical difference between Joint model and T2WI model by Delong test(P > 0.05).

CONCLUSIONS

The study suggested that the conventional T2WI sequence may be more suitable for prognosis evaluation of CSCC, which can provide a potential tool to facilitate the differential diagnosis of low-differentiation and high-differentiation CSCC.

A Machine Learning-Based Predictive Model of Epidermal Growth Factor Mutations in Lung Adenocarcinomas

Data from 758 patients with lung adenocarcinoma were retrospectively collected. All patients had undergone computed tomography imaging and EGFR gene testing. Radiomic features were extracted using the medical imaging tool 3D-Slicer and were combined with the clinical features to build a machine learning prediction model. The high-dimensional feature set was screened for optimal feature subsets using principal component analysis (PCA) and the least absolute shrinkage and selection operator (LASSO). Model prediction of EGFR mutation status in the validation group was evaluated using multiple classifiers. We showed that six clinical features and 622 radiomic features were initially collected. Thirty-one radiomic features with non-zero correlation coefficients were obtained by LASSO regression, and 24 features correlated with label values were obtained by PCA. The shared radiomic features determined by these two methods were selected and combined with the clinical features of the respective patient to form a subset of features related to EGFR mutations. The full dataset was partitioned into training and test sets at a ratio of 7:3 using 10-fold cross-validation. The area under the curve (AUC) of the four classifiers with cross-validations was: (1) K-nearest neighbor (AUCmean = 0.83, Acc = 81%); (2) random forest (AUCmean = 0.91, Acc = 83%); (3) LGBM (AUCmean = 0.94, Acc = 88%); and (4) support vector machine (AUCmean = 0.79, Acc = 83%). In summary, the subset of radiographic and clinical features selected by feature engineering effectively predicted the EGFR mutation status of this NSCLC patient cohort.

State of the Art: Lung Cancer Staging Using Updated Imaging Modalities

Lung cancer is among the most common mortality causes worldwide. This scientific article is a comprehensive review of current knowledge regarding screening, subtyping, imaging, staging, and management of treatment response for lung cancer. The traditional imaging modality for screening and initial lung cancer diagnosis is computed tomography (CT). Recently, a dual-energy CT was proven to enhance the categorization of variable pulmonary lesions. The National Comprehensive Cancer Network (NCCN) recommends usage of fluorodeoxyglucose positron emission tomography (FDG PET) in concert with CT to properly stage lung cancer and to prevent fruitless thoracotomies. Diffusion MR is an alternative to FDG PET/CT that is radiation-free and has a comparable diagnostic performance. For response evaluation after treatment, FDG PET/CT is a potent modality which predicts survival better than CT. Updated knowledge of lung cancer genomic abnormalities and treatment regimens helps to improve the radiologists’ skills. Incorporating the radiologic experience is crucial for precise diagnosis, therapy planning, and surveillance of lung cancer.

18F-FDG PET/CT-Based Prognostic Survival Model After Surgery for Head and Neck Cancer

The aims of this multicenter study were to identify clinical and preoperative PET/CT parameters predicting overall survival (OS) and distant metastasis-free survival (DMFS) in a cohort of head and neck squamous cell carcinoma patients treated with surgery, to generate a prognostic model of OS and DMFS, and to validate this prognostic model with an independent cohort. Methods: A total of 382 consecutive patients with head and neck squamous cell carcinoma, divided into training (n = 318) and validation (n = 64) cohorts, were retrospectively included. The following PET/CT parameters were analyzed: clinical parameters, SUV_max, SUV_mean, metabolic tumor volume (MTV), total lesion glycolysis, and distance parameters for the primary tumor and lymph nodes defined by 2 segmentation methods (relative SUV_max threshold and absolute SUV threshold). Cox analyses were performed for OS and DMFS in the training cohort. The concordance index (c-index) was used to identify highly prognostic parameters. These prognostic parameters were externally tested in the validation cohort. Results: In multivariable analysis, the significant parameters for OS were T stage and nodal MTV, with a c-index of 0.64 (P < 0.001). For DMFS, the significant parameters were T stage, nodal MTV, and maximal tumor-node distance, with a c-index of 0.76 (P < 0.001). These combinations of parameters were externally validated, with c-indices of 0.63 (P < 0.001) and 0.71 (P < 0.001) for OS and DMFS, respectively. Conclusion: The nodal MTV associated with the maximal tumor-node distance was significantly correlated with the risk of DMFS. Moreover, this parameter, in addition to clinical parameters, was associated with a higher risk of death. These prognostic factors may be used to tailor individualized treatment.

Temporal Heterogeneity of HER2 Expression and Spatial Heterogeneity of 18F-FDG Uptake Predicts Treatment Outcome of Pyrotinib in Patients with HER2-Positive Metastatic Breast Cancer

Background: This study aimed to evaluate tumor heterogeneity of metastatic breast cancer (MBC) and investigate its impact on the efficacy of pyrotinib in patients with HER2-positive MBC. Methods: MBC patients who underwent 18F-FDG PET/CT before pyrotinib treatment were included. Temporal and spatial tumor heterogeneity was evaluated by the discordance between primary and metastatic immunohistochemistry (IHC) results and baseline 18F-FDG uptake heterogeneity (intertumoral and intratumoral heterogeneity indexes: HI-inter and HI-intra), respectively. Progression-free survival (PFS) was estimated by the Kaplan−Meier method and compared by a log-rank test. Results: A total of 572 patients were screened and 51 patients were included. In 36 patients with matched IHC results, 25% of them had HER2 status conversion. Patients with homogenous HER2 positivity had the longest PFS, followed by patients with gained HER2 positivity, while patients with HER2 negative conversion could not benefit from pyrotinib (16.8 vs. 13.7 vs. 3.6 months, p < 0.0001). In terms of spatial heterogeneity, patients with high HI-intra and HI-inter had significantly worse PFS compared to those with low heterogeneity (10.6 vs. 25.3 months, p = 0.023; 11.2 vs. 25.3 months, p = 0.040). Conclusions: Temporal heterogeneity of HER2 status and spatial heterogeneity of 18F-FDG uptake could predict the treatment outcome of pyrotinib in patients with HER2-positive MBC, which provide practically applicable methods to assess tumor heterogeneity and guidance for treatment decisions.

The latest application progress of radiomics in prediction and diagnosis of liver diseases

INTRODUCTION

Early detection and individualized treatment of patients with liver disease is the key to survival. Radiomics can extract high-throughput quantitative features by multimode imaging, which has good application prospects for the diagnosis, staging and prognosis of benign and malignant liver diseases. Therefore, this paper summarizes the current research status in the field of liver disease, in order to help these patients achieve personalized and precision medical care.

AREAS COVERED

This paper uses several keywords on the PubMed database to search the references, and reviews the workflow of traditional radiomics, as well as the characteristics and influencing factors of different imaging modes. At the same time, the references on the application of imaging in different benign and malignant liver diseases were also summarized.

EXPERT OPINION

For patients with liver disease, the traditional imaging evaluation can only provide limited information. Radiomics exploits the characteristics of high-throughput and high-dimensional extraction, enabling liver imaging capabilities far beyond the scope of traditional visual image analysis. Recent studies have demonstrated the prospect of this technology in personalized diagnosis and treatment decision in various fields of the liver. However, further clinical validation is needed in its application and practice.

The prognostic value of 18F-FDG PET/CT intra-tumoural metabolic heterogeneity in pretreatment neuroblastoma patients

Background

Neuroblastoma (NB) is the most common tumour in children younger than 5 years old and notable for highly heterogeneous. Our aim was to quantify the intra-tumoural metabolic heterogeneity of primary tumour lesions by using ¹⁸F-FDG PET/CT and evaluate the prognostic value of intra-tumoural metabolic heterogeneity in NB patients.

Methods

We retrospectively enrolled 38 pretreatment NB patients in our study. ¹⁸F-FDG PET/CT images were reviewed and analyzed using 3D slicer software. The semi-quantitative metabolic parameters of primary tumour were measured, including the maximum standard uptake value (SUVmax), metabolic tumour volume (MTV), and total lesion glycolysis (TLG). The areas under the curve of cumulative SUV-volume histogram index (AUC-CSH index) was used to quantify intra-tumoural metabolic heterogeneity. The median follow-up was 21.3 months (range 3.6 - 33.4 months). The outcome endpoint was event-free survival (EFS), including progression-free survival and overall survival. Survival analysis was performed using Cox regression models and Kaplan Meier survival plots.

Results

In all 38 newly diagnosed NB patients, 2 patients died, and 17 patients experienced a relapse. The AUC-CSH_total (r=0.630, P<0.001) showed moderate correlation with the AUC-CSH_40%. In univariate analysis, chromosome 11q deletion (P=0.033), Children's Oncology Group (COG) risk grouping (P=0.009), bone marrow involvement (BMI, P=0.015), and AUC-CSH_total (P=0.007) were associated with EFS. The AUC-CSH_total (P=0.036) and BMI (P=0.045) remained significant in multivariate analysis. The Kaplan Meier survival analyses demonstrated that patients with higher intra-tumoural metabolic heterogeneity and BMI had worse outcomes (log-rank P=0.002).

Conclusion

The intra-tumoural metabolic heterogeneity of primary lesions in NB was an independent prognostic factor for EFS. The combined predictive effect of intra-tumoural metabolic heterogeneity and BMI provided prognostic survival information in NB patients.

Nasopharyngeal Carcinoma Radiomic Evaluation with Serial PET/CT: Exploring Features Predictive of Survival in Patients with Long-Term Follow-Up

PURPOSE

We aim determine the value of PET and CT radiomic parameters on survival with serial follow-up PET/CT in patients with nasopharyngeal carcinoma (NPC) for which curative intent therapy is undertaken.

METHODS

Patients with NPC and available pre-treatment as well as follow up PET/CT were included from 2005 to 2006 and were followed to 2021. Baseline demographic, radiological and outcome data were collected. Univariable Cox proportional hazard models were used to evaluate features from baseline and follow-up time points, and landmark analyses were performed for each time point.

RESULTS

Sixty patients were enrolled, and two-hundred and seventy-eight (278) PET/CT were at baseline and during follow-up. Thirty-eight percent (38%) were female, and sixty-two patients were male. All patients underwent curative radiation or chemoradiation therapy. The median follow-up was 11.72 years (1.26-14.86). Five-year and ten-year overall survivals (OSs) were 80.0% and 66.2%, and progression-free survival (PFS) was 90.0% and 74.4%. Time-dependent modelling suggested that, among others, PET gray-level zone length matrix (GLZLM) gray-level non-uniformity (GLNU) (HR 2.74 95% CI 1.06, 7.05) was significantly associated with OS. Landmark analyses suggested that CT parameters were most predictive at 15 month, whereas PET parameters were most predictive at time points 3, 6, 9 and 15 month.

CONCLUSIONS

This study with long-term follow up data on NPC suggests that mainly PET-derived radiomic features are predictive for OS but not PFS in a time-dependent evaluation. Furthermore, CT radiomic measures may predict OS and PFS best at initial and long-term follow-up time points and PET measures may be more predictive in the interval. These modalities are commonly used in NPC surveillance, and prospective validation should be considered.

Synergistic impact of motion and acquisition/reconstruction parameters on 18 F-FDG PET radiomic features in non-small cell lung cancer: phantom and clinical studies

Objectives

This study is aimed at examining the synergistic impact of motion and acquisition/reconstruction parameters on ¹⁸F‐FDG PET image radiomic features in non‐small cell lung cancer (NSCLC) patients, and investigating the robustness of features performance in differentiating NSCLC histopathology subtypes.

Methods

An in‐house developed thoracic phantom incorporating lesions with different sizes was used with different reconstruction settings, including various reconstruction algorithms, number of subsets and iterations, full‐width at half‐maximum of post‐reconstruction smoothing filter and acquisition parameters, including injected activity and test–retest with and without motion simulation. To simulate motion, a special motor was manufactured to simulate respiratory motion based on a normal patient in two directions. The lesions were delineated semi‐automatically to extract 174 radiomic features. All radiomic features were categorized according to the coefficient of variation (COV) to select robust features. A cohort consisting of 40 NSCLC patients with adenocarcinoma (n = 20) and squamous cell carcinoma (n = 20) was retrospectively analyzed. Statistical analysis was performed to discriminate robust features in differentiating histopathology subtypes of NSCLC lesions.

Results

Overall, 29% of radiomic features showed a COV ≤5% against motion. Forty‐five percent and 76% of the features showed a COV ≤ 5% against the test–retest with and without motion in large lesions, respectively. Thirty‐three percent and 45% of the features showed a COV ≤ 5% against different reconstruction parameters with and without motion, respectively. For NSCLC histopathological subtype differentiation, statistical analysis showed that 31 features were significant (p‐value < 0.05). Two out of the 31 significant features, namely, the joint entropy of GLCM (AUC = 0.71, COV = 0.019) and median absolute deviation of intensity histogram (AUC = 0.7, COV = 0.046), were robust against the motion (same reconstruction setting).

Conclusions

Motion, acquisition, and reconstruction parameters significantly impact radiomic features, just as their synergies. Radiomic features with high predictive performance (statistically significant) in differentiating histopathological subtype of NSCLC may be eliminated due to non‐reproducibility.

Robustness and reproducibility of radiomics in T2 weighted images from magnetic resonance image guided linear accelerator in a phantom study

PURPOSE

Quantitative radiomics features extracted from medical images have been shown to provide value in predicting clinical outcomes. The study for robustness and reproducibility of radiomics features obtained with magnetic resonance image guided linear accelerator (MR-Linac) is insufficient. The objective of this work was to investigate the stability of radiomics features extracted from T2-weighted images of MR-Linac for five common effect factors.

MATERIALS AND METHOD

In this work, ten jellies, five fruits/vegetables, and a dynamic phantom were used to evaluate the impact of test-retest, intraobserver, varied thicknesses, radiation, and motion. These phantoms were scanned on a 1.5 T MRI system of MR-Linac. For test-retest data, the phantoms were scanned twice with repositioning within 15 min. To assess for intraobserver comparison, the segmentation of MR images was repeated by one observer in a double-blind manner. Three slice thicknesses (1.2 mm, 2.4 mm, and 4.8 mm) were used to select robust features that were insensitive to different thicknesses. The effect of radiation on features was studied by acquiring images when the beam was on. Common movement images of patients during radiotherapy were simulated by a dynamic phantom with five motion states to study the motion effect. A total of 1409 radiomics features, including shape features, first-order features, and texture features, were extracted from the original, wavelet, square, logarithmic, exponential and gradient images. The robustness and reproducibility features were evaluated using the concordance correlation coefficient (CCC).

RESULT

The intraobserver group had the most robust features (936/1079, 86.7%), while the group of motion effects had the lowest robustness (56/936, 6.0%), followed by the group of different thickness cohorts (374/936, 40.0%). The stability of features in the test-retest and radiation groups was 1072 of 1312 (81.7%) and 810 of 936 (86.5%), respectively. Overall, 25 of 1409 (2.4%) radiomics features remained robust in all five tests, mostly focusing on the image type of the wavelet. The number of stable features extracted from when the beam was on was less than that extracted when the beam was off. Shape features were the most robust of all of the features in all of the groups, excluding the motion group.

CONCLUSION

Compared with other factors fewer features remained robust to the effect of motion. This result emphasizes the need to consider the effect of respiration motion. The study for T2-weighted images from MR-Linac under different conditions will help us to build a robust predictive model applicable for radiotherapy.

Radiomics-Based Predictive Model of Radiation-Induced Liver Disease in Hepatocellular Carcinoma Patients Receiving Stereo-Tactic Body Radiotherapy

(1) Background: The application of stereotactic body radiation therapy (SBRT) in hepatocellular carcinoma (HCC) limited the risk of the radiation-induced liver disease (RILD) and we aimed to predict the occurrence of RILD more accurately. (2) Methods: 86 HCC patients were enrolled. We identified key predictive factors from clinical, radiomic, and dose-volumetric parameters using a multivariate analysis, sequential forward selection (SFS), and a K-nearest neighbor (KNN) algorithm. We developed a predictive model for RILD based on these factors, using the random forest or logistic regression algorithms. (3) Results: Five key predictive factors in the training set were identified, including the albumin–bilirubin grade, difference average, strength, V5, and V30. After model training, the F1 score, sensitivity, specificity, and accuracy of the final random forest model were 0.857, 100, 93.3, and 94.4% in the test set, respectively. Meanwhile, the logistic regression model yielded an F1 score, sensitivity, specificity, and accuracy of 0.8, 66.7, 100, and 94.4% in the test set, respectively. (4) Conclusions: Based on clinical, radiomic, and dose-volumetric factors, our models achieved satisfactory performance on the prediction of the occurrence of SBRT-related RILD in HCC patients. Before undergoing SBRT, the proposed models may detect patients at high risk of RILD, allowing to assist in treatment strategies accordingly.

Assessment of different quantification metrics of [18F]-NaF PET/CT images of patients with abdominal aortic aneurysm

BACKGROUND

We aim to assess the spill-in effect and the benefit in quantitative accuracy for [18F]-NaF PET/CT imaging of abdominal aortic aneurysms (AAA) using the background correction (BC) technique.

METHODS

Seventy-two datasets of patients diagnosed with AAA were reconstructed with ordered subset expectation maximization algorithm incorporating point spread function (PSF). Spill-in effect was investigated for the entire aneurysm (AAA), and part of the aneurysm excluding the region close to the bone (AAAexc). Quantifications of PSF and PSF+BC images using different thresholds (% of max. SUV in target regions-of-interest) to derive target-to-background (TBR) values (TBRmax, TBR90, TBR70 and TBR50) were compared at 3 and 10 iterations.

RESULTS

TBR differences were observed between AAA and AAAexc due to spill-in effect from the bone into the aneurysm. TBRmax showed the highest sensitivity to the spill-in effect while TBR50 showed the least. The spill-in effect was reduced at 10 iterations compared to 3 iterations, but at the expense of reduced contrast-to-noise ratio (CNR). TBR50 yielded the best trade-off between increased CNR and reduced spill-in effect. PSF+BC method reduced TBR sensitivity to spill-in effect, especially at 3 iterations, compared to PSF (P-value ≤ 0.05).

CONCLUSION

TBR50 is robust metric for reduced spill-in and increased CNR.

Development and Validation of a CT-Based Signature for the Prediction of Distant Metastasis Before Treatment of Non-Small Cell Lung Cancer

RATIONALE AND OBJECTIVES

To develop and validate a radiomics model, a clinical-semantic model and a combined model by using standard methods for the pretreatment prediction of distant metastasis (DM) in patients with non-small-cell lung cancer (NSCLC) and to explore whether the combined model provides added value compared to the individual models.

MATERIALS AND METHODS

This retrospective study involved 356 patients with NSCLC. According to the image biomarker standardization initiative reference manual, we standardized the image processing and feature extraction using in-house software. Finally, 6692 radiomics features were extracted from each lesion based on contrast-enhanced chest CT images. The least absolute shrinkage selection operator and the recursive feature elimination algorithm were used to select features. The logistic regression classifier was used to build the model. Three models (radiomics model, clinical-semantic model and combined model) were constructed to predict DM in NSCLC. Area under the receiver operating characteristic curves were used to validate the ability of the three models to predict DM. A visual nomogram based on the combined model was developed for DM risk assessment in each patient.

RESULTS

The receiver operating characteristic curve showed predictive performance for DM of the radiomics model (area under the curve [AUC] values for training and validation were 0.76 [95% CI, 0.704 - 0.820] and 0.76 [95% CI, 0.653 - 0.858], respectively). The combined model had AUCs of 0.78 (95% CI, 0.723 - 0.835) and 0.77 (95% CI, 0.673 - 0.870) in the training and validation cohorts, respectively. Both the radiomics model and combined model performed better than the clinical-semantic model (0.70 [95% CI, 0.634 - 0.760] and 0.67 [95% CI, 0.554 - 0.787] in the training and validation cohorts, respectively).

CONCLUSION

The radiomics model and combined model may be useful for the prediction of DM in patients with NSCLC.

Radiomics in Triple Negative Breast Cancer: New Horizons in an Aggressive Subtype of the Disease

In the last decade, the analysis of the medical images has evolved significantly, applications and tools capable to extract quantitative characteristics of the images beyond the discrimination capacity of the investigator's eye being developed. The applications of this new research field, called radiomics, presented an exponential growth with direct implications in the diagnosis and prediction of response to therapy. Triple negative breast cancer (TNBC) is an aggressive breast cancer subtype with a severe prognosis, despite the aggressive multimodal treatments applied according to the guidelines. Radiomics has already proven the ability to differentiate TNBC from fibroadenoma. Radiomics features extracted from digital mammography may also distinguish between TNBC and non-TNBC. Recent research has identified three distinct subtypes of TNBC using IRM breast images voxel-level radiomics features (size/shape related features, texture features, sharpness). The correlation of these TNBC subtypes with the clinical response to neoadjuvant therapy may lead to the identification of biomarkers in order to guide the clinical decision. Furthermore, the variation of some radiomics features in the neoadjuvant settings provides a tool for the rapid evaluation of treatment efficacy. The association of radiomics features with already identified biomarkers can generate complex predictive and prognostic models. Standardization of image acquisition and also of radiomics feature extraction is required to validate this method in clinical practice.

Extraction parameter optimized radiomics for neoadjuvant chemotherapy response prognosis in advanced nasopharyngeal carcinoma

•

MRI radiomics is promising for NAC early response prediction in NPC patients.

•

Predictive performance could be improved by the optimized strategy.

•

The model could help with NPC individualized treatment.

Background and purpose

Neoadjuvant Chemotherapy (NAC) followed by concurrent chemoradiotherapy (CCRT) is promising in improving the survival rate for advanced nasopharyngeal carcinoma (NPC) patients relative to CCRT alone. However, not all patients respond well to NAC. Therefore, we aimed to develop and evaluate a modified radiomics model for the NAC response prognosis in NPC patients.

Methods

A total of 165 patients with biopsy-proven locally advanced NPC were retrospectively selected from the database of our hospital. 85 out of them were for training and cross-validation, while the other 80 patients were for independent testing. All patients were treated with NAC and underwent MRI inspection, including T1-weighted (T1), T2-weighted (T2), and contrast-enhanced T1-weighted (T1-cs) sequences before and after two cycles of NAC. We classified the patients into the response or non-response groups by the Response Evaluation Criteria in Solid Tumors 1.1 (RECIST 1.1). Radiomics features were extracted from the primary and lymph node gross tumor volume in each sequence. To further improve the predictive performance, the permutation of multiple combinations of extraction parameters has first ever been investigated in the NAC prognosis for NPC patients. The model was constructed by logistic regression and cross-validated by bootstrapping with a resampling number of 1000. Independent testing was also implemented. In addition, we also applied an imbalance-adjusted bootstrap strategy to decrease the bias of small samples.

Results

For the cross-validation cohort, the resultant AUC, sensitivity, and specificity in terms of 95% confidence interval were 0.948 ± 0.004, 0.849 ± 0.005, and 0.840 ± 0.010. For the independent testing cohort, the model reached an AUC of 0.925, a sensitivity of 0.821, and a specificity of 0.792. There was a significant difference in the estimated radiomics score between the response and non-response groups (P < 0.005).

Conclusions

An MRI-based radiomics model was developed and demonstrated promising capability for the individual prediction of NAC response in NPC patients. In particular, we have optimized the multiple combinations of texture extraction parameters with the permutation test and observed an encouraging improvement of the prediction performance compared to the previously published studies. The proposed model might provide chances for individualized treatment in NPC patients while retrenching the cost of clinical resources.

Quantification of tumor heterogeneity: from data acquisition to metric generation

Tumors are unique and complex ecosystems, in which heterogeneous cell subpopulations with variable molecular profiles, aggressiveness, and proliferation potential coexist and interact. Understanding how heterogeneity influences tumor progression has important clinical implications for improving diagnosis, prognosis, and treatment response prediction. Several recent innovations in data acquisition methods and computational metrics have enabled the quantification of spatiotemporal heterogeneity across different scales of tumor organization. Here, we summarize the most promising efforts from a common experimental and computational perspective, discussing their advantages, shortcomings, and challenges. With personalized medicine entering a new era of unprecedented opportunities, our vision is that of future workflows integrating across modalities, scales, and dimensions to capture intricate aspects of the tumor ecosystem and to open new avenues for improved patient care.

Feasibility on the Use of Radiomics Features of 11[C]-MET PET/CT in Central Nervous System Tumours: Preliminary Results on Potential Grading Discrimination Using a Machine Learning Model

Background/Aim: Nowadays, Machine Learning (ML) algorithms have demonstrated remarkable progress in image-recognition tasks and could be useful for the new concept of precision medicine in order to help physicians in the choice of therapeutic strategies for brain tumours. Previous data suggest that, in the central nervous system (CNS) tumours, amino acid PET may more accurately demarcate the active disease than paramagnetic enhanced MRI, which is currently the standard method of evaluation in brain tumours and helps in the assessment of disease grading, as a fundamental basis for proper clinical patient management. The aim of this study is to evaluate the feasibility of ML on 11[C]-MET PET/CT scan images and to propose a radiomics workflow using a machine-learning method to create a predictive model capable of discriminating between low-grade and high-grade CNS tumours. Materials and Methods: In this retrospective study, fifty-six patients affected by a primary brain tumour who underwent 11[C]-MET PET/CT were selected from January 2016 to December 2019. Pathological examination was available in all patients to confirm the diagnosis and grading of disease. PET/CT acquisition was performed after 10 min from the administration of 11C-Methionine (401–610 MBq) for a time acquisition of 15 min. 11[C]-MET PET/CT images were acquired using two scanners (24 patients on a Siemens scan and 32 patients on a GE scan). Then, LIFEx software was used to delineate brain tumours using two different semi-automatic and user-independent segmentation approaches and to extract 44 radiomics features for each segmentation. A novel mixed descriptive-inferential sequential approach was used to identify a subset of relevant features that correlate with the grading of disease confirmed by pathological examination and clinical outcome. Finally, a machine learning model based on discriminant analysis was used in the evaluation of grading prediction (low grade CNS vs. high-grade CNS) of 11[C]-MET PET/CT. Results: The proposed machine learning model based on (i) two semi-automatic and user-independent segmentation processes, (ii) an innovative feature selection and reduction process, and (iii) the discriminant analysis, showed good performance in the prediction of tumour grade when the volumetric segmentation was used for feature extraction. In this case, the proposed model obtained an accuracy of ~85% (AUC ~79%) in the subgroup of patients who underwent Siemens tomography scans, of 80.51% (AUC 65.73%) in patients who underwent GE tomography scans, and of 70.31% (AUC 64.13%) in the whole patients’ dataset (Siemens and GE scans). Conclusions: This preliminary study on the use of an ML model demonstrated to be feasible and able to select radiomics features of 11[C]-MET PET with potential value in prediction of grading of disease. Further studies are needed to improve radiomics algorithms to personalize predictive and prognostic models and potentially support the medical decision process.

Evaluation of conventional and deep learning based image harmonization methods in radiomics studies

Objective.To evaluate the impact of image harmonization on outcome prediction models using radiomics.Approach.234 patients from the Brain Tumor Image Segmentation Benchmark (BRATS) dataset with T1 MRI were enrolled in this study. Images were harmonized to a reference image using histogram matching (H_HM) and a generative adversarial network (GAN)-based method (H_GAN). 88 radiomics features were extracted on H_HM, H_GANand original (H_NONE) images. Wilcoxon paired test was used to identify features significantly impacted by the harmonization protocol used. Radiomic prediction models were built using feature selection with the Least Absolute Shrinkage and Selection Operator (LASSO) and Kaplan-Meier analysis.Main results.More than 50% of the features (49/88) were statistically modified by the harmonization with H_HMand 55 with H_GAN(adjustedp-value < 0.05). The contribution of histogram and texture features selected by the LASSO, in comparison to shape features that were not impacted by harmonization, was higher in harmonized datasets (47% for H_none, 62% for H_HMand 71% for H_GAN). Both image-based harmonization methods allowed to split patients into two groups with significantly different survival (p<0.05). With the H_GANimages, we were also able to build and validate a model using only features impacted by the harmonization (median survivals of 189 versus 437 days,p= 0.006)Significance.Data harmonization in a multi-institutional cohort allows to recover the predictive value of some radiomics features that was lost due to differences in the image properties across centers. In terms of ability to build survival prediction models in the BRATS dataset, the loss of power from impacted histogram and heterogeneity features was compensated by the selection of additional shape features. The harmonization using a GAN-based approach outperformed the histogram matching technique, supporting the interest for the development of new advanced harmonization techniques for radiomic analysis purposes.

Impact of Interobserver Variability in Manual Segmentation of Non-Small Cell Lung Cancer (NSCLC) Applying Low-Rank Radiomic Representation on Computed Tomography

Simple Summary

Discovery of predictive and prognostic radiomic features in cancer is currently of great interest to the radiologic and oncologic community. Tumor phenotypic and prognostic information can be obtained by extracting features on tumor segmentations, and it is typically imaging analysts, physician trainees, and attending physicians who provide these labeled datasets for analysis. The potential impact of level and type of specialty training on interobserver variability in manual segmentation of NSCLC was examined. Although there was some variability in segmentation between readers, the subsequently extracted radiomic features were overall well correlated. High fidelity radiomic feature extraction relies on accurate feature extraction from imaging that produce robust prognostic and predictive radiomic NSCLC biomarkers. This study concludes that this goal can be obtained using segmenters of different levels of training and clinical experience.

Abstract

This study tackles interobserver variability with respect to specialty training in manual segmentation of non-small cell lung cancer (NSCLC). Four readers included for segmentation are: a data scientist (BY), a medical student (LS), a radiology trainee (MH), and a specialty-trained radiologist (SK) for a total of 293 patients from two publicly available databases. Sørensen–Dice (SD) coefficients and low rank Pearson correlation coefficients (CC) of 429 radiomics were calculated to assess interobserver variability. Cox proportional hazard (CPH) models and Kaplan-Meier (KM) curves of overall survival (OS) prediction for each dataset were also generated. SD and CC for segmentations demonstrated high similarities, yielding, SD: 0.79 and CC: 0.92 (BY-SK), SD: 0.81 and CC: 0.83 (LS-SK), and SD: 0.84 and CC: 0.91 (MH-SK) in average for both databases, respectively. OS through the maximal CPH model for the two datasets yielded c-statistics of 0.7 (95% CI) and 0.69 (95% CI), while adding radiomic and clinical variables (sex, stage/morphological status, and histology) together. KM curves also showed significant discrimination between high- and low-risk patients (p-value < 0.005). This supports that readers’ level of training and clinical experience may not significantly influence the ability to extract accurate radiomic features for NSCLC on CT. This potentially allows flexibility in the training required to produce robust prognostic imaging biomarkers for potential clinical translation.

AI-Based Detection, Classification and Prediction/Prognosis in Medical Imaging:: Towards Radiophenomics

Artificial intelligence (AI) techniques have significant potential to enable effective, robust, and automated image phenotyping including the identification of subtle patterns. AI-based detection searches the image space to find the regions of interest based on patterns and features. There is a spectrum of tumor histologies from benign to malignant that can be identified by AI-based classification approaches using image features. The extraction of minable information from images gives way to the field of "radiomics" and can be explored via explicit (handcrafted/engineered) and deep radiomics frameworks. Radiomics analysis has the potential to be used as a noninvasive technique for the accurate characterization of tumors to improve diagnosis and treatment monitoring. This work reviews AI-based techniques, with a special focus on oncological PET and PET/CT imaging, for different detection, classification, and prediction/prognosis tasks. We also discuss needed efforts to enable the translation of AI techniques to routine clinical workflows, and potential improvements and complementary techniques such as the use of natural language processing on electronic health records and neuro-symbolic AI techniques.

Immunotherapy treatment outcome prediction in metastatic melanoma through an automated multi-objective delta-radiomics model

Based on recent studies, immunotherapy led by immune checkpoint inhibitors has significantly improved the patient survival rate and effectively reduced the recurrence risk. However, immunotherapy has different therapeutic effects for different patients, leading to difficulties in predicting the treatment response. Conversely, delta-radiomic features, which measure the difference between pre- and post-treatment through quantitative image features, have proven to be promising descriptors for treatment outcome prediction. Consequently, we developed an effective model termed as the automated multi-objective delta-radiomics (Auto-MODR) model for the prediction of immunotherapy response in metastatic melanoma. In Auto-MODR, delta-radiomic features and traditional radiomic features were used as inputs. Furthermore, a novel automated multi-objective model was developed to obtain more reliable and balanced results between sensitivity and specificity. We conducted extensive comparisons with existing studies on treatment outcome prediction. Our method achieved an area under the curve (AUC) of 0.86 in a cross-validation study and an AUC of 0.73 in an independent study. Compared with the model using conventional radiomic features (pre- and post-treatment) only, better performance can be obtained when conventional radiomic and delta-radiomic features are combined. Furthermore, Auto-MODR outperformed the currently available radiomic strategies.

Benchmarking Various Radiomic Toolkit Features While Applying the Image Biomarker Standardization Initiative toward Clinical Translation of Radiomic Analysis

The image biomarkers standardization initiative (IBSI) was formed to address the standardization of extraction of quantifiable imaging metrics. Despite its effort, there remains a lack of consensus or established guidelines regarding radiomic feature terminology, the underlying mathematics and their implementation across various software programs. This creates a scenario where features extracted using different toolboxes cannot be used to build or validate the same model leading to a non-generalization of radiomic results. In this study, IBSI-established phantom and benchmark values were used to compare the variation of the radiomic features while using 6 publicly available software programs and 1 in-house radiomics pipeline. All IBSI-standardized features (11 classes, 173 in total) were extracted. The relative differences between the extracted feature values from the different software programs and the IBSI benchmark values were calculated to measure the inter-software agreement. To better understand the variations, features are further grouped into 3 categories according to their properties: 1) morphology, 2) statistic/histogram and 3)texture features. While a good agreement was observed for a majority of radiomics features across the various tested programs, relatively poor agreement was observed for morphology features. Significant differences were also found in programs that use different gray-level discretization approaches. Since these software programs do not include all IBSI features, the level of quantitative assessment for each category was analyzed using Venn and UpSet diagrams and quantified using two ad hoc metrics. Morphology features earned lowest scores for both metrics, indicating that morphological features are not consistently evaluated among software programs. We conclude that radiomic features calculated using different software programs may not be interchangeable. Further studies are needed to standardize the workflow of radiomic feature extraction.

Fluorine-18 fluorodeoxyglucose positron emission tomography-based textural features for prediction of event prone carotid atherosclerotic plaques

BACKGROUND

Texture analysis has been increasingly used in the field of positron emission tomography (PET)/computed tomography (CT) imaging with Fluorine-18 fluorodeoxyglucose (18F-FDG), aiming at assessing tumor heterogeneity. The purpose of the present study is to examine the feasibility of performing texture analysis in carotid arteries, investigate the value of textural features as predictors of potential plaque vulnerability using as reference standards histological and immunohistochemical data and compare their performance with conventional uptake measurements.

METHODS

67 different 18F-FDG PET-based textural features were extracted from carotid images of 21 patients with high-grade carotid stenosis undergoing endarterectomy. To identify the more reliable predictors, univariate logistic regression analysis was performed. The accuracy was satisfactory in case of an Area Under the Receiver Operating Characteristic (ROC) curve (AUC) ≥ 0.80.

RESULTS

First measure of information correlation (AUC = 0.87, P < 0.001), large zone low gray level emphasis (AUC = 0.87, P < 0.001), and normalized run length non-uniformity (AUC = 0.84, P < 0.001) were the most optimal textural features for identifying characteristics of plaque vulnerability based on histological analysis. Addition of textural features to target-to-background ratio (TBR) (AUC = 0.74, P = 0.031) resulted in an AUC = 0.92 (P < 0.001), however, this did not reach statistical significance (Pdiff = 0.09). Intensity histogram standard deviation (AUC = 0.87, P < 0.001) and joint variance (AUC = 0.81, P = 0.001) were the most efficient features for signal differential in relation to immunohistochemical findings and provided incremental value compared to TBR (Pdiff = 0.02).

CONCLUSION

Texture analysis can be applied in 18F-FDG PET carotid imaging providing valuable information for plaque characterization.

Feasibility of prognosis assessment for cancer of unknown primary origin using texture analysis of 18F-fluorodeoxyglucose PET/computed tomography images of largest metastatic lymph node

OBJECTIVE

Cancers of unknown primary origin cannot be staged using images, making the prognosis difficult. We attempted to predict prognosis of patients with unknown primary origin using tumour heterogeneity recently introduced in F-fluorodeoxyglucose (F-FDG) PET/computed tomography (CT).

METHODS

Overall, 30 patients with unknown primary origin who underwent whole-body F-FDG PET/CT scans were retrospectively enrolled for texture analysis. The volume of interest was placed in the largest metastatic lymph nodes and conventional parameters and grey-level co-occurrence matrix (GLCM) were calculated. Statistical analysis of image-based variables was performed using Cox regression analyses. Patients were stratified into two groups based on cutoff values of GLCMentropy obtained using receiver operating characteristics (ROCs). Patients were analyzed, and overall survival (OS) was compared using Kaplan-Meier analysis.

RESULTS

Univariate Cox regression analysis showed significant differences in prognosis for parenchymal organ metastasis (P < 0.01), GLCM homogeneity (P = 0.01), GLCMcontrast (P < 0.01), GLCMentropy (P < 0.01) and GLCMdissimilarity (P < 0.01). Multivariate Cox regression analysis showed a significant difference in reduced prognosis for GLCMentropy positive (P < 0.01). Stratification was performed based on the GLCMentropy cutoff value, determined using ROCs analysis, with smaller groups showing better OS.

CONCLUSIONS

Despite previous difficulties in predicting prognosis in patients with unknown primary origin, F-FDG PET/CT texture features may enable stratification of prognosis. This could be useful for appropriate patient selection and management and help identify a subset of patients with favourable outcomes. These novel findings may be helpful for prognostication and improving patient care.

Radiomics in the Setting of Neoadjuvant Radiotherapy: A New Approach for Tailored Treatment

Simple Summary

This review based on a literature search aims at showing the impact of Texture Analysis in the prediction of response to neoadjuvant radiotherapy and/or chemoradiotherapy. The manuscript explores radiomics approaches in different fields of neoadjuvant radiotherapy, including esophageal cancer, lung cancer, sarcoma and rectal cancer in order to shed a light in the setting of neoadjuvant radiotherapy that can be used to tailor the best subsequent therapeutical strategy.

Abstract

Introduction: Neoadjuvant radiotherapy is currently used mainly in locally advanced rectal cancer and sarcoma and in a subset of non-small cell lung cancer and esophageal cancer, whereas in other diseases it is under investigation. The evaluation of the efficacy of the induction strategy is made possible by performing imaging investigations before and after the neoadjuvant therapy and is usually challenging. In the last decade, texture analysis (TA) has been developed to help the radiologist to quantify and identify the parameters related to tumor heterogeneity, which cannot be appreciated by the naked eye. The aim of this narrative is to review the impact of TA on the prediction of response to neoadjuvant radiotherapy and or chemoradiotherapy. Materials and Methods: Key references were derived from a PubMed query. Hand searching and ClinicalTrials.gov were also used. Results: This paper contains a narrative report and a critical discussion of radiomics approaches in different fields of neoadjuvant radiotherapy, including esophageal cancer, lung cancer, sarcoma, and rectal cancer. Conclusions: Radiomics can shed a light on the setting of neoadjuvant therapies that can be used to tailor subsequent approaches or even to avoid surgery in the future. At the same, these results need to be validated in prospective and multicenter trials.

MRI-based radiomics: promise for locally advanced cervical cancer treated with a tailored integrated therapeutic approach

OBJECTIVE

To assess prognostic factors by analyzing clinical and radiomic data of patients with locally advanced cervical cancer (LACC) treated with definitive concurrent cisplatin-based chemoradiotherapy (CCRT) using magnetic resonance imaging (MRI).

METHODS

We analyzed radiomic features from MRI in 60 women with FIGO (International Federation of Gynecology and Obstetrics) stage IB2-IVA cervical cancer who underwent definitive CCRT 45-50.4 Gy (in 25-28 fractions). Thirty-nine (65.0%) received EBRT sequential boost (4-20 Gy) on primary tumor site and 56 (93.3%) received high-dose-rate brachytherapy boost (6-28 Gy) (daily fractions of 5-7 Gy). Moreover, 71.7% of patients received dose-dense neoadjuvant chemotherapy for 6 cycles. The gross tumor volume was defined on T2-weighted sequences and 29 features were extracted from each MRI performed before and after CCRT, using dedicated software, and their prognostic value was correlated with clinical information.

RESULTS

In univariate analysis, age ⩾60 years and FIGO stage IB2-IIB had significantly better progression-free survival (PFS) (p = 0.022 and p = 0.009, respectively). There was a trend for significance for worse overall survival (OS) in patients with positive nodes (p = 0.062). In multivariate analysis, only age ⩾60 years and FIGO stage IB2-IIB reached significantly better PFS (p = 0.020 and p = 0.053, respectively). In radiomic dataset, in multivariate analysis, pregray level p75 was significantly associated with PFS (p = 0.047), pre-D3D value with OS (p = 0.049), and preinformation measure of correlation value with local control (p = 0.031).

CONCLUSION

The combination of clinical and radiomics features can provide information to predict behavior and prognosis of LACC and to make more accurate treatment decisions.

The Diagnostic Value of Conventional MRI and CT Features in the Identification of the IDH1-Mutant and 1p/19q Co-Deletion in WHO Grade II Gliomas

RATIONALE AND OBJECTIVES

The classification of patients based on pathology and molecular features is important for improving WHO grade II glioma patient prognosis, especially for the initially diagnosed patients. Less invasive and more convenient methods for the prediction of the pathological type and gene status are desired.

MATERIALS AND METHODS

This study investigates the ability to use conventional magnetic resonance imaging (MRI) and computed tomography (CT) features for determining the Isocitrate Dehydrogenase (IDH)-mutant and 1p/19q-codeletion status, through a retrospective review of information obtained from 189 WHO grade II glioma patients. Diffuse astrocytoma (IDH-mutant), Diffuse astrocytoma (IDH- wildtype) and Oligodendroglioma (IDH-mutant and 1p/19q co-deletion) were included in this cohort. All patients were divided into IDH-mutant group and IDH-wildtype group according to the IDH R132H mutation status. Moreover, all patients were divided into 1p/19q co-deletion group and 1p/19q non-codeletion group according to the 1p and 19q chromosome status. Patients underwent pre-operative CT and MRI scans, followed by operation and histopathological analyses, including immunohistochemistry and polymerase chain reaction analysis for IDH mutants, and fluorescence capillary electrophoresis analysis for the 1p/19q co-deletion. The χ² test, logistical regression and receiver operating characteristic curve analysis were conducted for statistical analysis.

RESULTS

IDH-mutant group patients exhibited a higher calcification frequency (25.2% vs 2.4%, p = 0.006) and lower frequency of T1 enhancement (20.4% vs 38.1%, p = 0.028) comparing patients in IDH-wildtype group, while 1p/19q co-deletion group patients exhibited a higher calcification frequency (46.67% vs 2.6%, p < 0.001) and lower homogenous signal frequency in T2WI (12.0% vs 31.6%, p = 0.014), sharp lesion margins (14.7% vs 43.0%, p = 0.010), T2/fluid attenuated inversion recovery mismatch signs (22.7% vs 50.9%, p = 0.001), and subventricular zone involvement (64.0% vs 15.8%, p = 0.021) comparing patients in 1p/19q non-codeletion group. According to the results of receiver operating characteristic analysis, these features were observed to have certain diagnostic abilities, especially with regard to combination parameters, which had a high diagnostic capability, with an area under the curve of 0.848.

CONCLUSION

Conventional MRI and CT features, which still represent the most convenient and widely used predictive method, might be a promising noninvasive predictor for differentiating between varied WHO grade II gliomas. Patients with calcification and T1 nonenhancement are more likely to be IDH-mutant. Moreover, patients with noncalcification, homogenous signal, sharp lesion margins, subventricular zone involvement on T2 and T2/fluid attenuated inversion recovery mismatch signs are more likely to be 1p/19q non-codeletion.

Exploratory Analysis of 18F-3'-deoxy-3'-fluorothymidine (18F-FLT) PET/CT-Based Radiomics for the Early Evaluation of Response to Neoadjuvant Chemotherapy in Patients With Locally Advanced Breast Cancer

Purpose

The objective of this study was to evaluate a set of radiomics-based advanced textural features extracted from ¹⁸F-FLT-PET/CT images to predict tumor response to neoadjuvant chemotherapy (NCT) in patients with locally advanced breast cancer (BC).

Materials and Methods

Patients with operable (T2-T3, N0-N2, M0) or locally advanced (T4, N0-N2, M0) BC were enrolled. All patients underwent chemotherapy (six cycles every 3 weeks). Surgery was performed within 4 weeks of the end of NCT. The MD Anderson Residual Cancer Burden calculator was used to evaluate the pathological response. ¹⁸F-FLT-PET/CT was performed 2 weeks before the start of NCT and approximately 3 weeks after the first cycle. The evaluation of PET response was based on EORTC criteria. Standard uptake value (SUV) statistics (SUV_max, SUV_peak, SUV_mean), together with 148 textural features, were extracted from each lesion. Indices that are robust against contour variability (ICC test) were used as independent variables to logistically model tumor response. LASSO analysis was used for variable selection.

Results

Twenty patients were included in the study. Lesions from 15 patients were evaluable and analyzed: 9 with pathological complete response (pCR) and 6 with pathological partial response (pPR). Concordance between PET response and histological examination was found in 13/15 patients. LASSO logistic modelling identified a combination of SUV_max and the textural feature index IVH_VolumeIntFract_90 as the most useful to classify PET response, and a combination of PET response, ID range, and ID_Coefficient of Variation as the most useful to classify pathological response.

Conclusions

Our study suggests the potential usefulness of FLT-PET for early monitoring of response to NCT. A model based on PET radiomic characteristics could have good discriminatory capacity of early response before the end of treatment.

A Bayesian approach to tissue-fraction estimation for oncological PET segmentation

Tumor segmentation in oncological PET is challenging, a major reason being the partial-volume effects (PVEs) that arise due to low system resolution and finite voxel size. The latter results in tissue-fraction effects (TFEs), i.e. voxels contain a mixture of tissue classes. Conventional segmentation methods are typically designed to assign each image voxel as belonging to a certain tissue class. Thus, these methods are inherently limited in modeling TFEs. To address the challenge of accounting for PVEs, and in particular, TFEs, we propose a Bayesian approach to tissue-fraction estimation for oncological PET segmentation. Specifically, this Bayesian approach estimates the posterior mean of the fractional volume that the tumor occupies within each image voxel. The proposed method, implemented using a deep-learning-based technique, was first evaluated using clinically realistic 2D simulation studies with known ground truth, in the context of segmenting the primary tumor in PET images of patients with lung cancer. The evaluation studies demonstrated that the method accurately estimated the tumor-fraction areas and significantly outperformed widely used conventional PET segmentation methods, including a U-net-based method, on the task of segmenting the tumor. In addition, the proposed method was relatively insensitive to PVEs and yielded reliable tumor segmentation for different clinical-scanner configurations. The method was then evaluated using clinical images of patients with stage IIB/III non-small cell lung cancer from ACRIN 6668/RTOG 0235 multi-center clinical trial. Here, the results showed that the proposed method significantly outperformed all other considered methods and yielded accurate tumor segmentation on patient images with Dice similarity coefficient (DSC) of 0.82 (95% CI: 0.78, 0.86). In particular, the method accurately segmented relatively small tumors, yielding a high DSC of 0.77 for the smallest segmented cross-section of 1.30 cm². Overall, this study demonstrates the efficacy of the proposed method to accurately segment tumors in PET images.

Machine Learning Predicts Outcomes of Phase III Clinical Trials for Prostate Cancer

The ability to predict the individual outcomes of clinical trials could support the development of tools for precision medicine and improve the efficiency of clinical-stage drug development. However, there are no published attempts to predict individual outcomes of clinical trials for cancer. We used machine learning (ML) to predict individual responses to a two-year course of bicalutamide, a standard treatment for prostate cancer, based on data from three Phase III clinical trials (n = 3653). We developed models that used a merged dataset from all three studies. The best performing models using merged data from all three studies had an accuracy of 76%. The performance of these models was confirmed by further modeling using a merged dataset from two of the three studies, and a separate study for testing. Together, our results indicate the feasibility of ML-based tools for predicting cancer treatment outcomes, with implications for precision oncology and improving the efficiency of clinical-stage drug development.