Set Of Radiomic Features Research Articles

Investigating Alzheimer’s disease progression using a radiomics approach: The hippocampal‐amygdala border in FDG‐PET Scans

AbstractBackgroundAlzheimer’s disease (AD) is a progressive neurodegenerative disorder characterized by memory loss and cognitive decline. Detecting AD early and tracking its development is essential for better management. This study delves into the potential of radiomics to identify the AD stages and monitor its progression using FDG PET images. The particular objective is to explore the potential of tracking progression using a reduced and meaningful set of imaging biomarkers at the borders of neighboring brain regions.MethodOur study analyzed 18FDG‐PET scans from 931 participants across five stages of AD from the ADNI database. Focusing on 95 brain regions, we employed the PyRadiomics tool and eight feature selection techniques. The Hippocampus, Amygdala, and Entorhinal regions emerged as consistently important across these methods. We delineated the connecting regions between these regions using a sliding window approach and identified the best two features that help detect AD by the use of a random forest classifier. We then plotted the values of the two features to track progression.ResultThe connectivity area between the hippocampus and amygdala demonstrated superior efficacy for diagnosing AD progression. The two most important features for this region, namely “Shape Mesh Volume“ and SDLGLE ”GLDM Small Dependence Low Gray Level Emphasis" exhibited the highest classification capability for predicting AD versus control normal individuals (Accuracy = 0.77, sensitivity = 0.73, specificity = 0.81). We plotted these two features across deteriorating stages. The features mean values were able to demonstrate the incremental deterioration. The variance of the feature values increased for deteriorating cases.ConclusionWe found that the small connection border between the hippocampus and the amygdala can serve as a biomarker to help us partially detect and interpret the presence and severity of AD. In particular, we found that changes in the volume and a measure called SDLGLE within this small border region were strongly linked to AD progression. The SDLGLE increase suggests a potential weakening of the tissue structure, possibly due to neurodegeneration. This preliminary finding highlights the power of a minimal set of radiomic features to not only detect and potentially explain the mechanisms behind AD progression but also to help track disease severity.

Automatic identification and classification of pediatric glomerulonephritis on ultrasound images based on deep learning and radiomics

BackgroundGlomerulonephritis (GN) encompasses a heterogeneous group of kidney diseases, often presenting with subclinical manifestations in children, leading to frequent missed diagnoses. Renal biopsy, while considered the gold standard, is invasive, prone to sampling errors, and time-consuming, thus hindering rapid diagnosis. This study aimed to develop a noninvasive diagnostic model for childhood GN using renal ultrasound images through the integration of deep learning and radiomics techniques.MethodsUltrasound images were acquired from children undergoing ultrasound-guided biopsy. A total of 469 renal ultrasound images were selected and divided into training and validation sets at a ratio of 8:2 to train a U-Net model for precise kidney image segmentation. Using radiomics, a comprehensive set of radiomic features were extracted from the segmented kidney regions. The extracted features were categorized based on GN types: IgA nephropathy (127 cases), minimal change disease (83 cases), and Henoch–Schönlein purpura nephritis (103 cases). These categories were further randomly split into training and validation sets at a ratio of 8:2. Within the training set, analysis of variance (ANOVA) was used for feature selection, followed by supervised Least Absolute Shrinkage and Selection Operator (LASSO) regression for dimensionality reduction, resulting in the selection of 37 features. These features were then integrated with a random forest algorithm to develop a GN classification model. The model's performance was comprehensively evaluated using the validation set.ResultsThe segmentation model exhibited remarkable performance during training, achieving an accuracy of 95.19% in the validation set. Thirty-seven features were identified through feature selection, leading to the development of a robust classification model. Evaluation on the validation set revealed high accuracy and predictive power across different GN categories, with Area Under the Curve (AUC) values ranging from 0.91 to 0.98.ConclusionsThe combined use of deep learning and radiomics techniques utilizing renal ultrasound images demonstrates significant potential for classifying childhood GN subtypes. This noninvasive approach holds promise for improving diagnostic efficiency and patient outcomes in GN.

Radiomics-Based Risk Assessment Correlates With Outcomes in Patients With Acute Type B Aortic Dissection Undergoing Thoracic Endovascular Repair.

Our study aimed to investigate the correlations between radiomics-based assessment and outcomes, including positive aortic remodeling (PAR), reintervention for dissection at 1 year, and overall survival, in patients with Type B aortic dissection (TBAD) who underwent thoracic endovascular aortic repair (TEVAR). This was a single-center, retrospective, cohort study. The cohort comprised 104 patients who had undergone TEVAR of TBAD in our institution between January 2010 and October 2022. We segmented preoperative computed tomography (CT) images of the patients' descending aorta regions, then extracted a comprehensive set of radiomic features, including first-order features, shape features (2D and 3D), gray-level co-occurrence matrix (GLCM), gray-level size zone matrix, gray-level run length matrix, gray-level dependence matrix, neighborhood gray-tone difference matrix, from the regions of interest. Next, we selected radiomics features associated with total descending aorta positive aortic remodeling (TDA-PAR) and reintervention by least absolute shrinkage and selection operator (LASSO) regression and features associated with survival by LASSO-Cox regression. This enabled us to calculate radiomics-based risk scores for each patient. We then allocated the patients to high and low radiomics-based risk groups, the cutoff being the median score. We used 3 different models to validate the radiomics-based risk scores. The patients' baseline characteristics did not differ between those who achieved TDA-PAR and those who did not. The radiomics-based risk scores were significantly and independently associated with all 3 outcomes. As to the impact of specific radiomics features, we found that GLSZM_SmallAreaLowGrayLevelEmphasis and shape_Maximum2DDiameterColumn had positive impacts on both reintervention and survival outcomes, whereas GLCM_Idmn positively affected survival but negatively affected reintervention. We found that radiomics-based risk for TDA-PAR correlated most significantly with zone 6 PAR. Radiomics-based risk scores were significantly associated with the outcomes of TDA-PAR, reintervention, and overall survival. Radiomics has the potential to make significant contributions to prediction of outcomes in patients with TBAD undergoing TEVAR. In this study of 104 patients with Type B aortic dissection, we demonstrated associations between radiomics-based risk and postoperative outcomes, including total descending aorta positive aortic remodeling, reintervention and survival. These findings highlight radiomics' potential as a tool for risk stratification and prognostication in acute Type B aortic dissection management.

Utility of radiomic features in predicting clinical outcomes in stage II-III pancreatic cancer.

74 Background: We identified computed tomography (CT)-derived radiomic features predictive of tumor progression within three months, then examined their ability to prognosticate overall survival (OS) along with clinical features in pancreatic cancer. We evaluated these features in patients with unresected pancreatic cancer who underwent stereotactic body radiation therapy (SBRT) in sequence with chemotherapy, but not surgery. Methods: In this retrospective study, we examined a cohort of 101 patients with stage II-III pancreatic cancer who underwent SBRT with sequential chemotherapy at a single institution (Stanford Health Care) between 1999-2020. From their pre-SBRT contrast-enhanced CT images with segmented tumors, delineating regions-of-interest, we extracted 900 radiomic (quantitative pixel-level imaging characteristic) features. In the first phase, we identified radiomic features that predicted rapid tumor progression within three months following SBRT. We divided the dataset into a training set (n = 53) for model development and a test set (n = 48) for evaluation. Using logistic regression with the Least Absolute Shrinkage and Selection Operator algorithm for feature selection and classification, we built a binary prediction model on the training set to identify patients at risk of progression within three months of SBRT. To fine-tune parameters, we performed five-fold cross-validation (CV) on the training set, repeating each set of parameters five times. We assessed model performance on the test set using the area under the curve (AUC). We selected the model with the best AUC, generating the predictive radiomic feature set. In the second phase, we conducted univariate and multivariate Cox regression analyses to assess the relationship between OS and individual clinical variables (age, sex, stage, vessel involvement, tumor location, performance status, body mass index, biological equivalent dose of radiation) and the radiomic feature set as high versus low risk. Results: Our cohort consisted of 48 men (mean age, 70 years ± 11 [SD]) and 53 women (mean age, 67 years ± 13 [SD]). From the first phase, 32 textural features comprised the radiomic feature set that best predicted rapid tumor progression, with mean AUCs of 0.852 (CV, n=53) and 0.814 (test, n=48). In the univariate Cox model, only the radiomic feature set was predictive of OS (hazard ratio, HR, 1.724, p=0.011). In the multivariate Cox model, radiomic features and age were significant predictors of OS, with HR of 1.819 (p=0.007) and 1.024 (p=0.024), respectively. Conclusions: CT-derived radiomic features predict rapid tumor progression following SBRT, confer nearly a twofold increase in mortality risk, and, along with patient age, enhance the identification of patients with stage II-III pancreatic cancer with poor OS. [Table: see text]

Evaluating the association between radiomic features on breast MRI and dormant tumor cell presence in the SURMOUNT trial.

561 Background: While breast cancer treatment is effective at eradicating primary disease, approximately 20% of patients will ultimately recur at a distant site. Recurrence has been linked to the presence of minimal residual disease (MRD) after treatment, which presents as circulating micrometastases and dormant tumor cells (DTCs) located primarily in the bone marrow. Radiomic features on MRI are imaging biomarkers that have been linked with lesion heterogeneity, which in turn has been associated with disease aggressiveness. Here, we examine the association between radiomic features from the primary lesion on MRI and the presence of post-treatment DTCs. Noninvasively identifying patients at high risk of MRD would facilitate optimized treatment strategies to reduce recurrence risk. Methods: Dynamic contrast-enhanced MRI data from breast cancer survivors enrolled in the SURMOUNT trial (N=104) were retrospectively analyzed. Participants with MRIs collected within 90 days of diagnosis who consented to provide a bone marrow aspirate (BMA) sample were eligible. Working with a board-certified radiologist, we successfully identified and segmented the primary lesion for N=70 patients with BMA outcomes. Using the publicly available CaPTk software, we computed 33 radiomic features for each segmented lesion based on pixel intensity, morphology, and grey level textures. To reduce the dimensionality of the feature space, we used these feature values to group the participants into clusters—termed radiomic phenotypes — via agglomerative hierarchical clustering with significance testing. Fisher’s exact test was applied to test the statistical significance of the association between radiomic phenotype and BMA outcome (± DTC). For all statistical tests, p<0.05 was considered significant. Results: Of the N=70 participants analyzed, 58 were negative and 12 were positive for DTCs present in the bone marrow. Two statistically significant radiomic phenotypes ( p<0.05) were identified that were strongly associated with BMA outcome ( p=0.0251). 11 out of 12 positive outcomes were grouped into the second radiomic phenotype (Table). Conclusions: We found a statistically significant association between DTC presence and radiomic phenotypes derived from MRI data. Given the large percentage of positive outcomes assigned to the second radiomic phenotype, the radiomic phenotypes may indicate low and high risk of MRD, respectively. Importantly, this analysis shows promise for the ability to predict MRD from noninvasive imaging features. These results motivate continued data collection to increase sample size and identify a robust set of radiomic features, which may ultimately be used as predictors in potential models of MRD risk. [Table: see text]

Abstract 3510: A radiogenomic approach for triple-negative breast cancer risk stratification

Abstract Background: Triple-negative breast cancer (TNBC) is an aggressive disease that accounts for 15-20% of all breast cancers. Expressions of ER, PR and HER2 receptors are lacking in this disease, and thus targeted therapies are not effective. TNBC has a shorter relapse-free survival, higher metastasis rate and decreased overall survival compared with other breast cancers. However, when undergoing standard treatment, some patients respond well, while others have poor outcome, suggesting TNBC heterogeneity. Early stratification of patients with long versus short survival could identify the subgroup of patients who would not benefit from exposure to toxicity of chemotherapy treatment. Here, we developed a non-invasive radiogenomic approach for TNBC risk stratification. Methods: A transcriptomic-based prognostic gene signature was previously developed using the TCGA-BRCA cohort (n=860). Briefly, LASSO Cox regression model analysis with the ‘glmnet’ R package was used to identify the transcriptomic signature gene-set consisting of 50 genes. We tested this signature to prognosticate overall survival in a Stanford cohort (n=63) and a previously published SCANB cohort (n=604). The patients were stratified into high- and low-risk groups based on the median risk-score. Next, we developed a machine learning model that identified a radiomic feature set to predict the prognostic transcriptomic risk-groups. Radiomic features were extracted from pre-treatment breast MRI. Radiomics features were extracted using PyRadiomics. The model utilized Decision Tree Classifier and LeaveOneOut method was used for cross-validation. Results: The transcriptomic signature low-risk group was significantly associated with improved overall survival in the two TNBC cohorts, with hazard ratios of 0.11 [95% CI: 0.01-0.88] for the Stanford cohort and 0.71 [95% CI: 0.52-0.97] for the SCANB cohort (log-rank p-values p=0.012 and p=0.032, respectively). Including this transcriptomic signature in a multivariate analysis, which adjusted for clinical features (patient age, grade, stage and Ki67%), the transcriptomic prognostic signature remained a significant prognostic factor (p<0.05). The radiomic feature set (consisting of 20 features) predicted the high- and low-risk transcriptomic groups with a mean accuracy of 72.2% and a mean AUROC of 71%. The precision, F1 and recall scores were 67%, 74% and 82%, respectively. In an independent dataset consisting of 116 Stanford TNBC patients, we used this model to predict risk groups based on the MRI radiomics features, and evaluated the prognostic effects of predicted risk groups. The overall survival of the predicted high-risk group was significantly poorer than the predicted low-risk group (p=0.013). Conclusions: We present a prognostic model that can non-invasively stratify TNBC patients for low versus high mortality risk using radiomic features derived from pre-treatment patient MRI data. Citation Format: Humaira Noor, Yuanning Zheng, Adam Mantz, Ryle Zhou, Andrew Kozlov, Wendy B. DeMartini, Shu-tian Chen, Satoko Okamoto, Debra Ikeda, Sarah Mattonen, Sandy Napel, Melinda L. Telli, George Sledge, Allison Kurian, Mina Satoyoshi, Olivier Gevaert, Haruka Itakura. A radiogenomic approach for triple-negative breast cancer risk stratification [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3510.

Radiomics analysis of patellofemoral joint improves knee replacement risk prediction: Data from the Multicenter Osteoarthritis Study (MOST)

ObjectiveKnee replacement (KR) is the last-resort treatment for knee osteoarthritis. Although radiographic evidence of tibiofemoral joint has been widely adopted for prognostication, patellofemoral joint has gained little attention and may hold additional value for further improvements. We aimed to quantitatively analyse patellofemoral joint through radiomics analysis of lateral view radiographs for improved KR risk prediction. DesignFrom the Multicenter Osteoarthritis Study dataset, we retrospectively retrieved the initial-visit lateral left knee radiographs of 2943 patients aged 50 to 79. They were split into training and test cohorts at a 2:1 ratio. A comprehensive set of radiomic features were extracted within the best-performing subregion of patellofemoral joint and combined into a radiomics score (RadScore). A KR risk score, derived from Kellgren-Lawrence grade (KLG) of tibiofemoral joint and RadScore of patellofemoral joint, was developed by multivariate Cox regression and assessed using time-dependent area under receiver operating characteristic curve (AUC). ResultsWhile patellofemoral osteoarthritis (PFOA) was insignificant during multivariate analysis, RadScore was identified as an independent risk factor (multivariate Cox p-value < 0.001) for KR. The subgroup analysis revealed that RadScore was particularly effective in predicting rapid progressor (KR occurrence before 30 months) among early- (KLG < 2) and mid-stage (KLG ​= ​2) patients. Combining two joints radiographic information, the AUC reached 0.89/0.87 for predicting 60-month KR occurrence. ConclusionsThe RadScore of the patellofemoral joint on lateral radiographs emerges as an independent prognostic factor for improving KR prognosis prediction. The KR risk score could be instrumental in managing progressive knee osteoarthritis interventions.

Predicting rapid progression and overall survival in stage II-III pancreatic cancer using a CT-based radiomic signature.

706 Background: We sought to assess the ability of a computed tomography (CT)-derived radiomic signature (RS) feature set to predict rapid tumor progression and overall survival (OS) in patients diagnosed with stage II-III pancreatic cancer undergoing stereotactic body radiation therapy (SBRT) in sequence with chemotherapy. Methods: We conducted a retrospective study on a cohort of 240 stage II-III pancreatic cancer patients who underwent SBRT in sequence with chemotherapy at a single institution (Stanford Hospital and Clinics). Among them, 83 had pre-SBRT contrast-enhanced CT images with segmented tumors, forming the imaging set, from which we extracted 900 radiomic (quantitative pixel-level imaging characteristic) features from delineated tumor regions-of-interest. For RS development, we divided the imaging set into a training set (n = 53) for model development and a test set (n = 30) for evaluation. We built a binary prediction model on the training set to identify patients at risk of rapid tumor progression within three months after SBRT. We utilized logistic regression with the Least Absolute Shrinkage and Selection Operator algorithm for feature selection and classification. To fine-tune parameters, we performed five-fold cross-validation on the training set, repeating each set of parameters five times. Finally, we assessed the model's performance on the test set (n = 30) using the area under the curve (AUC) value. We selected the model with the best AUC, while also generating the predictive radiomic feature set. For OS prognostication, we conducted both univariate and multivariate Cox proportional-hazards analyses using RS and clinical features (age, sex, stage, vessel involvement, tumor location, performance status, body mass index, biological equivalent dose of radiation) as potential predictors. Results: The enrolled cohort consisted of 127 men (mean age, 69 years ± 12 [SD]) and 113 women (mean age, 69 years ± 13 [SD]). Seventeen textural radiomic features were identified as the RS, which demonstrated a high AUC in the test set for the prediction of rapid progression (AUC 0.850; 95% CI: 0.725, 0.975). Age was associated with OS in the multivariate model while high RS (&gt;0.5) was independently associated with OS in both univariate and multivariate models. Conclusions: CT-derived RS, combined with age, represent the most prognostic factors for stage II-III pancreatic cancer. RS, which likely reflects underlying biology or molecular alterations, contributes to the highest HR of poor OS and provide the strongest indicator of outcome following commonly used treatments in this patient population. [Table: see text]

Diffusion Kurtosis Imaging and Radiomics in Diffuse Axonal Injury

This study aimed to assess the feasibility of radiomic features derived from diffusion kurtosis imaging (DK MRI) in identifying microstructural brain damage in diffuse axonal injury (DAI) and predicting its outcome. We hypothesized that radiomic features, computed from parametric DK MRI maps, may differ between healthy individuals and those with trauma, and may be related to DAI outcomes. The study included 31 DAI patients and 12 healthy volunteers. A total of 342,300 radiomic features were calculated (2282 features for each combination of 10 parametric DK maps with 15 regions of interest). Our findings suggest that the set of radiomic features effectively distinguishes between healthy and damaged brain tissues, and can predict DAI outcome. A broad spectrum of radiomic parameters based on DK MRI data showed high diagnostic and prognostic potential in DAI, presenting advantages beyond the traditionally used average values for the regions of interest on parametric DK MRI maps.

Impact of Wavelet Kernels on Predictive Capability of Radiomic Features: A Case Study on COVID-19 Chest X-ray Images

Radiomic analysis allows for the detection of imaging biomarkers supporting decision-making processes in clinical environments, from diagnosis to prognosis. Frequently, the original set of radiomic features is augmented by considering high-level features, such as wavelet transforms. However, several wavelets families (so called kernels) are able to generate different multi-resolution representations of the original image, and which of them produces more salient images is not yet clear. In this study, an in-depth analysis is performed by comparing different wavelet kernels and by evaluating their impact on predictive capabilities of radiomic models. A dataset composed of 1589 chest X-ray images was used for COVID-19 prognosis prediction as a case study. Random forest, support vector machine, and XGBoost were trained (on a subset of 1103 images) after a rigorous feature selection strategy to build-up the predictive models. Next, to evaluate the models generalization capability on unseen data, a test phase was performed (on a subset of 486 images). The experimental findings showed that Bior1.5, Coif1, Haar, and Sym2 kernels guarantee better and similar performance for all three machine learning models considered. Support vector machine and random forest showed comparable performance, and they were better than XGBoost. Additionally, random forest proved to be the most stable model, ensuring an appropriate balance between sensitivity and specificity.

Prostate Gleason Score Detection by Calibrated Machine Learning Classification through Radiomic Features

The Gleason score was originally formulated to represent the heterogeneity of prostate cancer and helps to stratify the risk of patients affected by this tumor. The Gleason score assigning represents an on H&amp;E stain task performed by pathologists upon histopathological examination of needle biopsies or surgical specimens. In this paper, we propose an approach focused on the automatic Gleason score classification. We exploit a set of 18 radiomic features. The radiomic feature set is directly obtainable from segmented magnetic resonance images. We build several models considering supervised machine learning techniques, obtaining with the RandomForest classification algorithm a precision ranging from 0.803 to 0.888 and a recall from to 0.873 to 0.899. Moreover, with the aim to increase the never seen instance detection, we exploit the sigmoid calibration to better tune the built model.

NIMG-85. RADIOMIC FEATURES PREDICTIVE OF RESPONSE IN HGG-TARGETING CAR-T THERAPY

Abstract SIGNIFICANCE Radiomics may improve precision medicine in CAR-T (Chimeric Antigen Receptor T Cell) therapy patient selection. BACKGROUND High-Grade Glioma (HGG) is a heterogenous primary CNS neoplasm with a high recurrence rate and poor outcomes. Many studies are exploring CAR T cells to combat HGG. Radiomic models have shown value in identifying biomarkers predictive of tumor genetics, response, and patient prognosis. In this study, we explore radiomic features derived from four clinical sequences and volumes of edema and enhancing tumor to predict treatment response to the first three doses of CAR-T therapy. METHODS In this IRB-approved phase 1 clinical trial (IRB 13384), patients underwent surgical resection of the tumor and received CAR-T cell therapy. Of the 82 patients accrued, 59 (20 females, median age = 49) completed three cycles of therapy and had 3T field strength MRI scans with minimal imaging artifacts. T1 weighted pre-contrast, T1 weighted post-contrast, T2 weighted, and T2 FLuid Attenuated Inversion Recovery sequences were used to generate 3D and 2D radiomics features. In total 28,541 radiomic features were generated per patient using the images prior to CAR-T administration. Each patient’s response to treatment after three cycles was determined to be either stable disease (29 patients) or progression. The radiomic feature set dimensionality was reduced using Maximum Relevance Minimum Redundancy. 10-fold cross-validation XGBoost was used to determine radiomic features predictive of treatment response with a randomized grid search for hyperparameter tuning. RESULTS Six radiomic features (four shape-based), had high SHapley Additive exPlanations (SHAP)-based importance feature predictive of RANO response with an AUC &amp;gt;0.73. CONCLUSION Despite the limited study size, such imaging-based radiomic models can serve as a potential basis for optimizing clinical trial design through more precise patient screening and providing potential predictive imaging biomarkers of whether a patient will respond to CAR-T cell therapy in HGGs.

Leveraging Serial MRI Radiomics and Machine Learning to Predict Risk of Radiation Necrosis in Patients with Brain Metastases Managed with Stereotactic Radiation and Immunotherapy

<h3>Purpose/Objective(s)</h3> Radiation necrosis (RN) is the primary dose-limiting toxicity of stereotactic radiotherapy (SRT) for brain metastases (BMs), especially in patients also receiving immunotherapy (IO). We investigated the utility of serial magnetic resonance imaging (MRI) radiomic features to develop predictive quantitative imaging biomarkers of brain metastases response to SRT and IO. <h3>Materials/Methods</h3> Using an IRB-approved single-institution database, we retrospectively identified adult patients with brain metastases managed with SRT and IO between 2006 and 2021. We collected per-lesion outcomes measures of RN (radiographic and/or symptomatic), progressive disease (PD), or neither (NA). Routine pre- and post-SRT contrast-enhanced T1-weighted MRIs were pushed to MIM Software (Beachwood, OH), where individual BMs were manually contoured. A total of 1061 radiomics features were extracted per lesion using a digital gene expression panel. They were selected from the following categories: shape, intensity, neighborhood intensity difference (NID), grey-level co-occurrence matrix (GLCM), and grey-level run-length matrix (GLRLM). Feature reduction followed using Spearman's correlation coefficient (0.3 cut-off). Features with non-significantly different mean ranks pre- and post-SRT (p<0.01) per Wilcoxon signed-rank test were also excluded. Simple then multiple nominal logistic regressions (SLR and MLR) with Bonferroni correction were used to model the risk of RN, PD, or NA based on pre- and post-SRT, and delta-radiomic ([post-pre]/pre) features. 95% CI of ROC AUC was obtained by 10,000 bootstrapping. Statistical analysis was performed using software. <h3>Results</h3> Our dataset encompassed 301 BMs from 92 patients with NSCLC, melanoma, and renal cell carcinoma, all treated with brain SRT and IO. 74 (24.6%), 75 (24.9%), and 152 developed RN, PD, or NA, respectively. The final radiomic feature set included 39 features that significantly changed post-SRT (p<0.01). MLR resulted in a 3-radiomic feature model that predicted risk of RN (ROC AUC 0.71). The model included post-SRT BM surface area and texture (GLCM 3D Homogeneity) and pre-RT BM roundness. Interestingly, another 5-radiomic feature MLR model predicted RN and PD (ROC AUC 0.71) based on pre-SRT GLCM 3D texture feature, 3 post-SRT shape features (surface area, mean breadth, and roundness), in addition to delta-volume. <h3>Conclusion</h3> Our analysis showed that static and serial MR radiomic features could potentially predict BMs response to SRT and IO. Model external validation, combination with clinical variables, and integration of other MRI sequences are warranted before the large-scale adoption of this novel diagnostic approach.

Radiomics and Machine Learning for Detecting Scar Tissue on CT Delayed Enhancement Imaging.

BackgroundDelayed enhancement CT (CT-DE) has been evaluated as a tool for the detection of myocardial scar and compares well to the gold standard of MRI with late gadolinium enhancement (MRI-LGE). Prior work has established that high performance can be achieved with manual reading; however, few studies have looked at quantitative measures to differentiate scar and healthy myocardium on CT-DE or automated analysis.MethodsEighteen patients with clinically indicated MRI-LGE were recruited for CT-DE at multiple 80 and 100 kV post contrast imaging. Left ventricle segmentation was performed on both imaging modalities, along with scar segmentation on MRI-LGE. Segmentations were registered together and scar regions were estimated on CT-DE. 93 radiomic features were calculated and analysed for their ability to differentiate between scarred and non-scarred myocardium regions. Machine learning (ML) classifiers were trained using the strongest set of radiomic features to classify segments containing scar on CT-DE. Features and classifiers were compared across both tube voltages and combined-energy images.ResultsThere were 59 and 51 statistically significant features in the 80 and 100 kV images respectively. Combined-energy imaging increased this to 63 with more features having area under the curve (AUC) above 0.9. The 10 highest AUC features for each image were used in the ML classifiers. The 100 kV images produced the best ML classifier, a support vector machine with an AUC of 0.88 (95% CI 0.87–0.90). Comparable performance was achieved with both the 80 kV and combined-energy images.ConclusionsCT-DE can be quantitatively analyzed using radiomic feature calculations. These features may be suitable for ML classification techniques to prospectively identify AHA segments with performance comparable to previously reported manual reading. Future work on larger CT-DE datasets is warranted to establish optimum imaging parameters and features.

A novel methodology for head and neck carcinoma treatment stage detection by means of model checking

ContextHead and neck cancers are diagnosed at an annual rate of 3% to 7% with respect to the total number of cancers, and 50% to 75% of such new tumours occur in the upper aerodigestive tract. PurposeIn this paper we propose formal methods based approach aimed to identify the head and neck tumour treatment stage by means of model checking. We exploit a set of radiomic features to model medical imaging as a labelled transition system to verify treatment stage properties. Main findingsWe experiment the proposed method using a public dataset related to computed tomography images obtained in different treatment stages, reaching an accuracy ranging from 0.924 to 0.978 in treatment stage detection. Principal conclusionsThe study confirms the effectiveness of the adoption of formal methods in the head and neck carcinoma treatment stage detection to support radiologists and pathologists.

Imaging Radiomic Biomarkers of Mandibular Osteoradionecrosis for Head and Neck Cancer

<h3>Purpose/Objective(s)</h3> This study aims to identify radiomic features extracted from contrast-enhanced CT scans that differentiate osteoradionecrosis (ORN) from normal mandibular bone in head and neck cancer patients treated with radiotherapy. <h3>Materials/Methods</h3> Contrast-enhanced CT images were collected for patients with confirmed ORN diagnosis at MD Anderson Cancer Center between 2008 and 2018. The ORN regions of interest (ROIs) were segmented manually in each image. The control ROIs of the contralateral health mandible were generated by a Python script then adjusted manually in each image. Commercial software was then used to extract the radiomic features from both ORN and control ROIs after the application of intrinsic filters. The pairwise correlation filter was used to remove radiomic features whose pairwise correlation was ≥0.99. Filter algorithms were then used to further reduce the number of radiomic features. After that, wrapper and embedded methods were applied on the resulting radiomic features. Finally, Gini importance and Recursive Feature Elimination (RFE) were used to select the final radiomic features for the predictive model. The support vector machine (SVM) with linear kernel was used for the binary classification of ORN and normal mandibular bone. The performance of the model was evaluated using the Area Under Curve (AUC). <h3>Results</h3> A total of 150 patients with radiologically established ORN were included in our study. The mean age was 62.3 years (range 27-82). The mean duration between the end of RT and ORN diagnosis was 32.6 months. The pairwise correlation omitted 432 features with a correlation ≥ 0.99. After that, the first step of the radiomic features engineering (using the filter algorithm) resulted in the selection of 33 radiomic features with statistically significant results in all the following three statistical methods: Pearson correlation, Chi-square test, and F-score. The RFE based on the Gini index selected 5 radiomics features. The final classifier used SVM with linear Kernel. The input for this classifier was the final set of radiomic features (N=5). We validated this binary classification model using 5-fold cross-validation. During this validation, the range of AUC was (0.84–0.95) & the average AUC was 0.90. <h3>Conclusion</h3> We successfully used imaging radiomic features to construct an accurate model (AUC= 0.90) to discriminate ORN versus normal mandibular bone in head and neck cancer patients. Future studies are needed to validate this model in prospective studies to early detect ORN in head and neck cancer patients after radiation treatment.

A radiomics approach for lung nodule detection in thoracic CT images based on the dynamic patterns of morphological variation.

To propose and evaluate a set of radiomic features, called morphological dynamics features, for pulmonary nodule detection, which were rooted in the dynamic patterns of morphological variation and needless precise lesion segmentation. Two datasets were involved, namely, university hospital (UH) and LIDC datasets, comprising 72 CT scans (360 nodules) and 888 CT scans (2230 nodules), respectively. Each nodule was annotated by multiple radiologists. Denoted the category of nodules identified by at least k radiologists as ALk. A nodule detection algorithm, called CAD-MD algorithm, was proposed based on the morphological dynamics radiomic features, characterizing a lesion by ten sets of the same features with different values extracted from ten different thresholding results. Each nodule candidate was classified by a two-level classifier, including ten decision trees and a random forest, respectively. The CAD-MD algorithm was compared with a deep learning approach, the N-Net, using the UH dataset. On the AL1 and AL2 of the UH dataset, the AUC of the AFROC curves were 0.777 and 0.851 for the CAD-MD algorithm and 0.478 and 0.472 for the N-Net, respectively. The CAD-MD algorithm achieved the sensitivities of 84.4% and 91.4% with 2.98 and 3.69 FPs/scan and the N-Net 74.4% and 80.7% with 3.90 and 4.49 FPs/scan, respectively. On the LIDC dataset, the CAD-MD algorithm attained the sensitivities of 87.6%, 89.2%, 92.2%, and 95.0% with 4 FPs/scan for AL1-AL4, respectively. The morphological dynamics radiomic features might serve as an effective set of radiomic features for lung nodule detection. • Texture features varied with such CT system settings as reconstruction kernels of CT images, CT scanner models, and parameter settings, and so on. • Shape and first-order statistics were shown to be the most robust features against variation in CT imaging parameters. • The morphological dynamics radiomic features, which mainly characterized the dynamic patterns of morphological variation, were shown to be effective for lung nodule detection.

Brain tumor segmentation and overall survival period prediction in glioblastoma multiforme using radiomic features

SummaryGlioblastoma multiforme (GBM or glioblastoma) is a fast‐growing glioma that are the most invasive type of glial tumors, rapidly growing and commonly spreading into nearby brain tissue. Due to its aggressive and fast growing nature, patients suffer from high grade glioma (GBM) survive very less time as compare to other tumors. Prediction of patient survival (OS) time helps the radiologist for better systematic treatment planning and clinical decision making. The OS rate depends on the tumor size, shape, and different imaging features of brain. In this study, the OS period prediction was performed using Random Forest, SVM, XgBoost, and LGBM taking radiomic features which represents fused deep features and hand crafted features of the tumor. Efficiency of the prediction depends on the tumor volume that is segmented from the different MRI modalities. Hence the whole tumor and its sub tumor are extracted from multi‐modal MR images using U‐Net++ deep model and stacked together for deep features extraction using convolutional neural networks. To increase the accuracy, the features are reduced using PCA and then this radiomic feature set was used for OS period prediction. Prediction performance was evaluated for both 2‐class and 3‐class survival groups. The experiment was performed on well‐known dataset BraTS 2017 and achieved a classification AUC value as 63% for 3‐class classification and 2‐class group using different classifier. Segmentation DOR is computed as 1269.29, 2033.99, and 648.00 for complete tumor, enhancing tumor, and necrotic tumor extraction, respectively. To achieve even more accuracy, bio inspired optimization methods GA and PSO are used on fused feature set. Finally, the method achieves the AUC score of 0.66 using fused feature+SVM+GA (3‐class group) and 0.70 using fused feature+SVM+PSO (2‐class group) which outperforms the state‐of‐the‐art.

Predicting cell invasion in breast tumor microenvironment from radiological imaging phenotypes

BackgroundThe abundance of immune and stromal cells in the tumor microenvironment (TME) is informative of levels of inflammation, angiogenesis, and desmoplasia. Radiomics, an approach of extracting quantitative features from radiological imaging to characterize diseases, have been shown to predict molecular classification, cancer recurrence risk, and many other disease outcomes. However, the ability of radiomics methods to predict the abundance of various cell types in the TME remains unclear. In this study, we employed a radio-genomics approach and machine learning models to predict the infiltration of 10 cell types in breast cancer lesions utilizing radiomic features extracted from breast Dynamic Contrast Enhanced Magnetic Resonance Imaging.MethodsWe performed a retrospective study utilizing 73 patients from two independent institutions with imaging and gene expression data provided by The Cancer Imaging Archive (TCIA) and The Cancer Genome Atlas (TCGA), respectively. A set of 199 radiomic features including shape-based, morphological, texture, and kinetic characteristics were extracted from the lesion volumes. To capture one-to-one relationships between radiomic features and cell type abundance, we performed linear regression on each radiomic feature/cell type abundance combination. Each regression model was tested for statistical significance. In addition, multivariate models were built for the cell type infiltration status (i.e. “high” vs “low”) prediction. A feature selection process via Recursive Feature Elimination was applied to the radiomic features on the training set. The classification models took the form of a binary logistic extreme gradient boosting framework. Two evaluation methods including leave-one-out cross validation and external independent test, were used for radiomic model learning and testing. The models’ performance was measured via area under the receiver operating characteristic curve (AUC).ResultsUnivariate relationships were identified between a set of radiomic features and the abundance of fibroblasts. Multivariate models yielded leave-one-out cross validation AUCs ranging from 0.5 to 0.83, and independent test AUCs ranging from 0.5 to 0.68 for the multiple cell type invasion predictions.ConclusionsOn two independent breast cancer cohorts, breast MRI-derived radiomics are associated with the tumor’s microenvironment in terms of the abundance of several cell types. Further evaluation with larger cohorts is needed.

Using Radiomic Features to Predict Visual Prognosis of Idiopathic Macular Hole Surgery

Idiopathic macular holes (IMH) are a leading cause of vision loss in middle-aged and older people. IMH is usually treated surgically by vitrectomy. Improvement in visual acuity is used to assess the prognosis of macular hole surgery. To achieve better diagnosis and treatment of IMH, Optical coherence tomography (OCT) technology is widely used. It will improve patient outcomes if parameters that are correlated with favorable prognoses are identified in OCT images. Both 2D OCT images and 3D OCT images can be obtained, which allows for the possibility of extracting 3D features. We propose the use of radiomic features as predictors of IMH prognosis. In this study, radiomic techniques were used to identify shape and texture features of IMH. OCT images were sliced into 2D images to enable experts to segment the region of interest before feature extraction. Elements of an initial set of mined features were selected as predictors based on correlation analysis and least absolute shrinkage and selection operator data analysis. P-values between selected features and best-corrected visual acuity were calculated to identify significant features. Nine features were found that influence surgical outcomes. Receiver operating characteristic curve analyses of positive samples (post-surgical visual outcome improved more than 0.3) and negative samples (all nonpositive samples) showed the area under the ROC curve of the radiomic feature set reached 0.81. Analysis showed that radiomic features of preoperative OCT images accurately predict visual prognosis, and thus identifying the features can be a powerful diagnostic technique.