Radiomics: a new application from established techniques

A predictive model using the mesoscopic architecture of the living brain to detect Alzheimer's disease

Background

Alzheimer's disease, the most common cause of dementia, causes a progressive and irreversible deterioration of cognition that can sometimes be difficult to diagnose, leading to suboptimal patient care.

Methods

We developed a predictive model that computes multi-regional statistical morpho-functional mesoscopic traits from T1-weighted MRI scans, with or without cognitive scores. For each patient, a biomarker called "Alzheimer's Predictive Vector" (ApV) was derived using a two-stage least absolute shrinkage and selection operator (LASSO).

Results

The ApV reliably discriminates between people with (ADrp) and without (nADrp) Alzheimer's related pathologies (98% and 81% accuracy between ADrp - including the early form, mild cognitive impairment - and nADrp in internal and external hold-out test sets, respectively), without any a priori assumptions or need for neuroradiology reads. The new test is superior to standard hippocampal atrophy (26% accuracy) and cerebrospinal fluid beta amyloid measure (62% accuracy). A multiparametric analysis compared DTI-MRI derived fractional anisotropy, whose readout of neuronal loss agrees with ADrp phenotype, and SNPrs2075650 is significantly altered in patients with ADrp-like phenotype.

Conclusions

This new data analytic method demonstrates potential for increasing accuracy of Alzheimer diagnosis.

Testing the Ability of Convolutional Neural Networks to Learn Radiomic Features

BACKGROUND AND OBJECTIVE

Radiomics and deep learning have emerged as two distinct approaches to medical image analysis. However, their relative expressive power remains largely unknown. Theoretically, hand-crafted radiomic features represent a mere subset of features that neural networks can approximate, thus making deep learning a more powerful approach. On the other hand, automated learning of hand-crafted features may require a prohibitively large number of training samples. Here we directly test the ability of convolutional neural networks (CNNs) to learn and predict the intensity, shape, and texture properties of tumors as defined by standardized radiomic features.

METHODS

Conventional 2D and 3D CNN architectures with an increasing number of convolutional layers were trained to predict the values of 16 standardized radiomic features from real and synthetic PET images of tumors, and tested. In addition, several ImageNet-pretrained advanced networks were tested. A total of 4000 images were used for training, 500 for validation, and 500 for testing.

RESULTS

Features quantifying size and intensity were predicted with high accuracy, while shape irregularity and heterogeneity features had very high prediction errors and generalized poorly. For example, mean normalized prediction error of tumor diameter with a 5-layer CNN was 4.23 ± 0.25, while the error for tumor sphericity was 15.64 ± 0.93. We additionally found that learning shape features required an order of magnitude more samples compared to intensity and size features.

CONCLUSIONS

Our findings imply that CNNs trained to perform various image-based clinical tasks may generally under-utilize the shape and texture information that is more easily captured by radiomics. We speculate that to improve the CNN performance, shape and texture features can be computed explicitly and added as auxiliary variables to the networks, or supplied as synthetic inputs.

Radiomics utilization to differentiate nonfunctional adenoma in essential hypertension and functional adenoma in primary aldosteronism

We performed the present study to investigate the role of computed tomography (CT) radiomics in differentiating nonfunctional adenoma and aldosterone-producing adenoma (APA) and outcome prediction in patients with clinically suspected primary aldosteronism (PA). This study included 60 patients diagnosed with essential hypertension (EH) with nonfunctional adenoma on CT and 91 patients with unilateral surgically proven APA. Each whole nodule on unenhanced and venous phase CT images was segmented manually and randomly split into training and test sets at a ratio of 8:2. Radiomic models for nodule discrimination and outcome prediction of APA after adrenalectomy were established separately using the training set by least absolute shrinkage and selection operator (LASSO) logistic regression, and the performance was evaluated on test sets. The model can differentiate adrenal nodules in EH and PA with a sensitivity, specificity, and accuracy of 83.3%, 78.9% and 80.6% (AUC = 0.91 [0.72, 0.97]) in unenhanced CT and 81.2%, 100% and 87.5% (AUC = 0.98 [0.77, 1.00]) in venous phase CT, respectively. In the outcome after adrenalectomy, the models showed a favorable ability to predict biochemical success (Unenhanced/venous CT: AUC = 0.67 [0.52, 0.79]/0.62 [0.46, 0.76]) and clinical success (Unenhanced/venous CT: AUC = 0.59 [0.47, 0.70]/0.64 [0.51, 0.74]). The results showed that CT-based radiomic models hold promise to discriminate APA and nonfunctional adenoma when an adrenal incidentaloma was detected on CT images of hypertensive patients in clinical practice, while the role of radiomic analysis in outcome prediction after adrenalectomy needs further investigation.

A Radiomics Nomogram for Non-Invasive Prediction of Progression-Free Survival in Esophageal Squamous Cell Carcinoma

To construct a prognostic model for preoperative prediction on computed tomography (CT) images of esophageal squamous cell carcinoma (ESCC), we created radiomics signature with high throughput radiomics features extracted from CT images of 272 patients (204 in training and 68 in validation cohort). Multivariable logistic regression was applied to build the radiomics signature and the predictive nomogram model, which was composed of radiomics signature, traditional TNM stage, and clinical features. A total of 21 radiomics features were selected from 954 to build a radiomics signature which was significantly associated with progression-free survival (p < 0.001). The area under the curve of performance was 0.878 (95% CI: 0.831–0.924) for the training cohort and 0.857 (95% CI: 0.767–0.947) for the validation cohort. The radscore of signatures' combination showed significant discrimination for survival status. Radiomics nomogram combined radscore with TNM staging and showed considerable improvement over TNM staging alone in the training cohort (C-index, 0.770 vs. 0.603; p < 0.05), and it is the same with clinical data (C-index, 0.792 vs. 0.680; p < 0.05), which were confirmed in the validation cohort. Decision curve analysis showed that the model would receive a benefit when the threshold probability was between 0 and 0.9. Collectively, multiparametric CT-based radiomics nomograms provided improved prognostic ability in ESCC.

A Radiomics Nomogram for Distinguishing Benign From Malignant Round-Like Breast Tumors

OBJECTIVE

The objective of this study is to develop a radiomics nomogram for the presurgical distinction of benign and malignant round-like solid tumors.

METHODS

This retrospective trial enrolled patients with round-like tumors who had received preoperative digital mammography (DM) no sooner than 20 days prior to surgery. Breast tumors were segmented manually on DM images in order to extract radiomic features. Four machine learning classification models were constructed, and their corresponding areas under the receiver operating characteristic (ROC) curves (AUCs) for differential tumor diagnosis were calculated. The optimal classifier was then selected for the validation set. After this, predictive machine learning models that employed radiomic features and/or patient features were applied for tumor assessment. The models' AUC, accuracy, negative (NPV) and positive (PPV) predictive values, sensitivity, and specificity were then derived.

RESULTS

In total 129 cases with benign and malignant tumors confirmed by pathological analysis were enrolled in the study, including 91 and 38 in the training and test sets, respectively. The DM images yielded 1,370 features per patient. For the machine learning models, the Least Absolute Shrinkage and Selection Operator for Gradient Boosting Classifier turned out to be the optimal classifier (AUC=0.87, 95% CI 0.76-0.99), and ROC curves for the radiomics nomogram and the DM-only model were statistically different (P<0.001). The radiomics nomogram achieved an AUC of 0.90 (95% CI 0.80-1.00) in the test cohort and was statistically higher than the DM-based model (AUC=0.67, 95% CI 0.51-0.84). The radiomics nomogram was highly efficient in detecting malignancy, with accuracy, sensitivity, specificity, PPV, and NPV in the validation set of 0.868, 0.950, 0.778, 0.826, and 0.933, respectively.

CONCLUSIONS

This radiomics nomogram that combines radiomics signatures and clinical characteristics represents a noninvasive, cost-efficient presurgical prediction technique.

Radiomics for the Prediction of Response to Antifibrotic Treatment in Patients with Idiopathic Pulmonary Fibrosis: A Pilot Study

Antifibrotic therapy has changed the treatment paradigm for idiopathic pulmonary fibrosis (IPF); however, a subset of patients still experienced rapid disease progression despite treatment. This study aimed to determine whether CT-based radiomic features can predict therapeutic response to antifibrotic agents. In this retrospective study, 35 patients with IPF on antifibrotic treatment enrolled from two centers were divided into training (n = 26) and external validation (n = 9) sets. Clinical and pulmonary function data were collected. The patients were categorized into stable disease (SD) and progressive disease (PD) groups based on functional or radiologic criteria. From pretreatment non-enhanced high-resolution CT (HRCT) images, twenty-six radiomic features were extracted through whole-lung texture analysis, and six parenchymal patterns were quantified using dedicated imaging platforms. The predictive factors for PD were determined via univariate and multivariate logistic regression analyses. In the training set (SD/PD: 12/14), univariate analysis identified eight radiomic features and ground-glass opacity percentage (GGO%) as potential predicators of PD. However, multivariate analysis found that the single independent predictor was the sum entropy (accuracy, 80.77%; AUC, 0.75). The combined sum entropy-GGO% model improved the predictive performance in the training set (accuracy, 88.46%; AUC, 0.77). The overall accuracy of the combined model in the validation set (SD/PD: 7/2) was 66.67%. Our preliminary results demonstrated that radiomic features based on pretreatment HRCT could predict the response of patients with IPF to antifibrotic treatment.

Identification of Radiomic Features as an Imaging Marker to Differentiate Benign and Malignant Breast Masses Based on Magnetic Resonance Imaging

Background - Breast cancer is one of the most common cancers among women globally and early identification is known to increase patient outcomes. Therefore, the main aim of this study is to identify the essential radiomic features as an image marker and compare the diagnostic feasibility of feature parameters derived from radiomics analysis and conventional Magnetic Resonance Imaging (MRI) to differentiate benign and malignant breast masses.

Methods and Material - T1-weighted Dynamic Contrast-Enhanced (DCE) breast MR axial images of 151 (benign (79) and malignant (72)) patients were chosen. Regions of interest were selected using both manual and semi-automatic segmentation from each lesion. 382 radiomic features computed on the selected regions. A random forest model was employed to detect the most important features that differentiate benign and malignant breast masses. The ten most important radiomics features were obtained from manual and semi-automatic segmentation based on the Gini index to train a support vector machine. MATLAB and IBM SPSS Statistics Subscription software used for statistical analysis.

Results - The accuracy (sensitivity) of the models built from the ten most significant features obtained from manual and semi-automatic segmentation were 0.815 (0.84), 0.821 (0.87), respectively. The top 10 features obtained from manual delineation and semi-automatic segmentation showed a significant difference (p<0.05) between benign and malignant breast lesions.

Conclusion - This radiomics analysis based on DCE-BMRI revealed distinct radiomic features to differentiate benign and malignant breast masses. Therefore, the radiomics analysis can be used as a supporting tool in detecting breast MRI lesions.

Radiomics analysis based on CT's greater omental caking for predicting pathological grading of pseudomyxoma peritonei

The objective of this study was to predict the preoperative pathological grading and survival period of Pseudomyxoma peritonei (PMP) by establishing models, including a radiomics model with greater omental caking as the imaging observation index, a clinical model including clinical indexes, and a combined model of these two. A total of 88 PMP patients were selected. Clinical data of patients, including age, sex, preoperative serum tumor markers [CEA, CA125, and CA199], survival time, and preoperative computed tomography (CT) images were analyzed. Three models (clinical model, radiomics model and combined model) were used to predict PMP pathological grading. The models’ diagnostic efficiency was compared and analyzed by building the receiver operating characteristic (ROC) curve. Simultaneously, the impact of PMP’s different pathological grades was evaluated. The results showed that the radiomics model based on the CT’s greater omental caking, an area under the ROC curve ([AUC] = 0.878), and the combined model (AUC = 0.899) had diagnostic power for determining PMP pathological grading. The imaging radiomics model based on CT greater omental caking can be used to predict PMP pathological grading, which is important in the treatment selection method and prognosis assessment.

Tumor Connectomics: Mapping the Intra-Tumoral Complex Interaction Network Using Machine Learning

The high-level relationships that form complex networks within tumors and between surrounding tissue is challenging and not fully understood. To better understand these tumoral networks, we developed a tumor connectomics framework (TCF) based on graph theory with machine learning to model the complex interactions within and around the tumor microenvironment that are detectable on imaging. The TCF characterization model was tested with independent datasets of breast, brain, and prostate lesions with corresponding validation datasets in breast and brain cancer. The TCF network connections were modeled using graph metrics of centrality, average path length (APL), and clustering from multiparametric MRI with IsoSVM. The Matthews Correlation Coefficient (MCC), Area Under the Curve-ROC, and Precision-Recall (AUC-ROC and AUC-PR) were used for statistical analysis. The TCF classified the breast and brain tumor cohorts with an IsoSVM AUC-PR and MCC of 0.86, 0.63 and 0.85, 0.65, respectively. The TCF benign breast lesions had a significantly higher clustering coefficient and degree centrality than malignant TCFs. Grade 2 brain tumors demonstrated higher connectivity compared to Grade 4 tumors with increased degree centrality and clustering coefficients. Gleason 7 prostate lesions had increased betweenness centrality and APL compared to Gleason 6 lesions with AUC-PR and MCC ranging from 0.90 to 0.99 and 0.73 to 0.87, respectively. These TCF findings were similar in the validation breast and brain datasets. In conclusion, we present a new method for tumor characterization and visualization that results in a better understanding of the global and regional connections within the lesion and surrounding tissue.

Radiomics analysis enables fatal outcome prediction for hospitalized patients with coronavirus disease 2019 (COVID-19)

BACKGROUND

In December 2019, a rare respiratory disease named coronavirus disease 2019 (COVID-19) broke out, leading to great concern around the world.

PURPOSE

To develop and validate a radiomics nomogram for predicting the fatal outcome of COVID-19 pneumonia.

MATERIAL AND METHODS

The present study consisted of a training dataset (n = 66) and a validation dataset (n = 30) with COVID-19 from January 2020 to March 2020. A radiomics signature was generated using the least absolute shrinkage and selection operator (LASSO) Cox regression model. A radiomics score (Rad-score) was developed from the training cohort. The radiomics model, clinical model, and integrated model were built to assess the association between radiomics signature/clinical characteristics and the mortality of COVID-19 cases. The radiomics signature combined with the Rad-score and the independent clinical factors and radiomics nomogram were constructed.

RESULTS

Seven stable radiomics features associated with the mortality of COVID-19 were finally selected. A radiomics nomogram was based on a combined model consisting of the radiomics signature and the clinical risk factors indicating optimal predictive performance for the fatal outcome of patients with COVID-19 with a C-index of 0.912 (95% confidence interval [CI] 0.867-0.957) in the training dataset and 0.907 (95% CI 0.849-0.966) in the validation dataset. The calibration curves indicated optimal consistency between the prediction and the observation in both training and validation cohorts.

CONCLUSION

The CT-based radiomics nomogram indicated favorable predictive efficacy for the overall survival risk of patients with COVID-19, which could help clinicians intensively follow up high-risk patients and make timely diagnoses.

Rapid Progress in Intelligent Radiotherapy and Future Implementation

Radiotherapy is one of the major approaches to cancer treatment. Artificial intelligence in radiotherapy (shortly, Intelligent radiotherapy) mainly involves big data, deep learning, extended reality, digital twin, radiomics, Internet plus and Internet of Things (IoT), which establish an automatic and intelligent network platform consisting of radiotherapy preparation, target volume delineation, treatment planning, radiation delivery, quality assurance (QA) and quality control (QC), prognosis judgment and post-treatment follow-up. Intelligent radiotherapy is an interdisciplinary frontier discipline in infancy. The review aims to summary the important implements of intelligent radiotherapy in various areas and put forward the future of unmanned radiotherapy center.

Diagnostic Value of Radiomics Analysis in Contrast-Enhanced Spectral Mammography for Identifying Triple-Negative Breast Cancer

Purpose

To evaluate the value of radiomics analysis in contrast-enhanced spectral mammography (CESM) for the identification of triple-negative breast cancer (TNBC).

Method

CESM images of 367 pathologically confirmed breast cancer patients (training set: 218, testing set: 149) were retrospectively analyzed. Cranial caudal (CC), mediolateral oblique (MLO), and combined models were built on the basis of the features extracted from subtracted images on CC, MLO, and the combination of CC and MLO, respectively, in the tumour region. The performance of the models was evaluated through receiver operating characteristic (ROC) curve analysis, the Hosmer-Lemeshow test, and decision curve analysis (DCA). The areas under ROC curves (AUCs) were compared through the DeLong test.

Results

The combined CC and MLO model had the best AUC and sensitivity of 0.90 (95% confidence interval: 0.85–0.96) and 0.97, respectively. The Hosmer–Lemeshow test yielded a non-significant statistic with p-value of 0.59. The clinical usefulness of the combined CC and MLO model was confirmed if the threshold was between 0.02 and 0.81 in the DCA.

Conclusions

Machine learning models based on subtracted images in CESM images were valuable for distinguishing TNBC and NTNBC. The model with the combined CC and MLO features had the best performance compared with models that used CC or MLO features alone.

Trends in Imaging Patterns of Bronchogenic Carcinoma: Reality or a Statistical Variation? A Single-Center Cross-Sectional Analysis of Outcomes

INTRODUCTION

Bronchogenic carcinoma accounts for more cancer-related deaths than any other malignancy and is the most frequently diagnosed cancer in the world. Bronchogenic carcinoma is by far the leading cause of cancer death among both men and women, making up almost 25% of all cancer deaths. The objective of this study was to identify the changing trends, if any, in radiological patterns of bronchogenic carcinoma to document the various computed tomography (CT) appearances of bronchogenic carcinoma with histopathologic correlation.

METHODS

This was a single-center cross-sectional study on 162 patients with clinical or radiological suspicion of bronchogenic carcinoma with histopathological confirmation of diagnosis.

RESULTS

There was a male preponderance with bronchogenic carcinoma and smoking being the most common risk factor. Squamous cell carcinoma followed by adenocarcinoma and small cell carcinoma is the most common histologic subtype. Squamous cell carcinoma was noted to be present predominantly in the peripheral location (55.5%), and adenocarcinoma was noted to be present predominantly in the central location (68.4%).

CONCLUSION

CT is the imaging modality of choice for evaluating bronchogenic carcinoma and provides for precise characterization of the size, extent, and staging of the carcinoma. Among 162 bronchogenic carcinoma cases evaluated in the current study, a definite changing trend in the radiological pattern of squamous cell carcinoma and adenocarcinoma was observed. Squamous cell carcinoma was predominantly noted to be a peripheral tumor, and adenocarcinoma is predominantly noted to be a central tumor. Surveillance or restaging scans are recommended, considering the high mortality rate in patients with bronchogenic carcinoma.

Multiparametric radiomic tissue signature and machine learning for distinguishing radiation necrosis from tumor progression after stereotactic radiosurgery

Background

Stereotactic radiosurgery (SRS) may cause radiation necrosis (RN) that is difficult to distinguish from tumor progression (TP) by conventional MRI. We hypothesize that MRI-based multiparametric radiomics (mpRad) and machine learning (ML) can differentiate TP from RN in a multi-institutional cohort.

Methods

Patients with growing brain metastases after SRS at 2 institutions underwent surgery, and RN or TP were confirmed by histopathology. A radiomic tissue signature (RTS) was selected from mpRad, as well as single T₁ post-contrast (T₁c) and T₂ fluid-attenuated inversion recovery (T₂-FLAIR) radiomic features. Feature selection and supervised ML were performed in a randomly selected training cohort (N = 95) and validated in the remaining cases (N = 40) using surgical pathology as the gold standard.

Results

One hundred and thirty-five discrete lesions (37 RN, 98 TP) from 109 patients were included. Radiographic diagnoses by an experienced neuroradiologist were concordant with histopathology in 67% of cases (sensitivity 69%, specificity 59% for TP). Radiomic analysis indicated institutional origin as a significant confounding factor for diagnosis. A random forest model incorporating 1 mpRad, 4 T₁c, and 4 T₂-FLAIR features had an AUC of 0.77 (95% confidence interval [CI]: 0.66–0.88), sensitivity of 67% and specificity of 86% in the training cohort, and AUC of 0.71 (95% CI: 0.51–0.91), sensitivity of 52% and specificity of 90% in the validation cohort.

Conclusions

MRI-based mpRad and ML can distinguish TP from RN with high specificity, which may facilitate the triage of patients with growing brain metastases after SRS for repeat radiation versus surgical intervention.

Building an Artificial Intelligence Laboratory Based on Real World Data: The Experience of Gemelli Generator

The problem of transforming Real World Data into Real World Evidence is becoming increasingly important in the frameworks of Digital Health and Personalized Medicine, especially with the availability of modern algorithms of Artificial Intelligence high computing power, and large storage facilities.Even where Real World Data are well maintained in a hospital data warehouse and are made available for research purposes, many aspects need to be addressed to build an effective architecture enabling researchers to extract knowledge from data.We describe the first year of activity at Gemelli Generator RWD, the challenges we faced and the solutions we put in place to build a Real World Data laboratory at the service of patients and health researchers. Three classes of services are available today: retrospective analysis of existing patient data for descriptive and clustering purposes; automation of knowledge extraction, ranging from text mining, patient selection for trials, to generation of new research hypotheses; and finally the creation of Decision Support Systems, with the integration of data from the hospital data warehouse, apps, and Internet of Things.

Development and Validation of an MRI-Based Nomogram Model for Predicting Disease-Free Survival in Locally Advanced Rectal Cancer Treated With Neoadjuvant Radiotherapy

Objectives

To develop a prognostic prediction MRI-based nomogram model for locally advanced rectal cancer (LARC) treated with neoadjuvant therapy.

Methods

This was a retrospective analysis of 233 LARC (MRI-T stage 3-4 (mrT) and/or MRI-N stage 1-2 (mrN), M0) patients who had undergone neoadjuvant radiotherapy and total mesorectal excision (TME) surgery with baseline MRI and operative pathology assessments at our institution from March 2015 to March 2018. The patients were sequentially allocated to training and validation cohorts at a ratio of 4:3 based on the image examination date. A nomogram model was developed based on the univariate logistic regression analysis and multivariable Cox regression analysis results of the training cohort for disease-free survival (DFS). To evaluate the clinical usefulness of the nomogram, Harrell’s concordance index (C-index), calibration plot, receiver operating characteristic (ROC) curve analysis, and decision curve analysis (DCA) were conducted in both cohorts.

Results

The median follow-up times were 43.2 months (13.3–61.3 months) and 32.0 months (12.3–39.5 months) in the training and validation cohorts. Multivariate Cox regression analysis identified MRI-detected extramural vascular invasion (mrEMVI), pathological T stage (ypT) and perineural invasion (PNI) as independent predictors. Lymphovascular invasion (LVI) (which almost reached statistical significance in multivariate regression analysis) and three other independent predictors were included in the nomogram model. The nomogram showed the best predictive ability for DFS (C-index: 0.769 (training cohort) and 0.776 (validation cohort)). It had a good 3-year DFS predictive capacity [area under the curve, AUC=0.843 (training cohort) and 0.771 (validation cohort)]. DCA revealed that the use of the nomogram model was associated with benefits for the prediction of 3-year DFS in both cohorts.

Conclusion

We developed and validated a novel nomogram model based on MRI factors and pathological factors for predicting DFS in LARC treated with neoadjuvant therapy. This model has good predictive value for prognosis, which could improve the risk stratification and individual treatment of LARC patients.

Radiomic Features Associated with Extent of Resection in Glioma Surgery

Radiomics defines a set of techniques for extraction and quantification of digital medical data in an automated and reproducible way. Its goal is to detect features potentially related to a clinical task, like classification, diagnosis, prognosis, and response to treatment, going beyond the intrinsic limits of operator-dependency and qualitative description of conventional radiological evaluation on a mesoscopic scale. In the field of neuro-oncology, researchers have tried to create prognostic models for a better tumor diagnosis, histological and biomolecular classification, prediction of response to treatment, and identification of disease relapse. Concerning glioma surgery, the most significant aid that radiomics can give to surgery is to improve tumor extension detection and identify areas that are more prone to recurrence to increase the extent of tumor resection, thereby ameliorating the patients' prognosis. This chapter aims to review the fundamentals of radiomics models' creation, the latest advance of radiomics in neuro-oncology, and possible radiomic features associated with the extent of resection in the brain gliomas.

Radiomics in surgical oncology: applications and challenges

Surgery is a curative treatment option for many patients with malignant tumors. Increased attention has focused on the combination of surgery with chemotherapy, as multimodality treatment has been associated with promising results in certain cancer types. Despite these data, there remains clinical equipoise on optimal timing and patient selection for neoadjuvant or adjuvant strategies. Radiomics, an emerging field involving the extraction of advanced features from radiographic images, has the potential to revolutionize oncologic treatment and contribute to the advance of personalized therapy by helping predict tumor behavior and response to therapy. This review analyzes and summarizes studies that use radiomics with machine learning in patients who have received neoadjuvant and/or adjuvant chemotherapy to predict prognosis, recurrence, survival, and therapeutic response for various cancer types. While studies in both neoadjuvant and adjuvant settings demonstrate above average performance on ability to predict progression-free and overall survival, there remain many challenges and limitations to widespread implementation of this technology. The lack of standardization of common practices to analyze radiomics, limited data sharing, and absence of auto-segmentation have hindered the inclusion and rapid adoption of radiomics in prospective, clinical studies.

Baseline 18F-FDG PET textural features as predictors of response to chemotherapy in diffuse large B-cell lymphoma

PURPOSE

We sought to investigate the performance of radiomics analysis on baseline 18F-FDG PET/CT for predicting response to first-line chemotherapy in diffuse large B-cell lymphoma (DLBCL).

MATERIAL AND METHODS

Forty-five patients who received first-line rituximab, cyclophosphamide, doxorubicin, vincristine and prednisone (R-CHOP) chemotherapy for DLBCL were included in the study. Radiomics features and standard uptake value (SUV)-based measurements were extracted from baseline PET images for a total of 147 lesions. The selection of the most relevant features was made using the recursive feature elimination algorithm. A machine-learning model was trained using the logistic regression classifier with cross-validation to predict treatment response. The independent predictors of incomplete response were evaluated with multivariable regression analysis.

RESULTS

A total of 14 textural features were selected by the recursive elimination algorithm, achieving a feature-to-lesion ratio of 1:10. The accuracy and area under the receiver operating characteristic curve of the model for predicting incomplete response were 0.87 and 0.81, respectively. Multivariable analysis revealed that SUVmax and gray level co-occurrence matrix dissimilarity were independent predictors of lesions with incomplete response to first-line R-CHOP chemotherapy.

CONCLUSION

Increased textural heterogeneity in baseline PET images was found to be associated with incomplete response in DLBCL.

Framework for Machine Learning of CT and PET Radiomics to Predict Local Failure after Radiotherapy in Locally Advanced Head and Neck Cancers

Context:

Cancer Radiomics is an emerging field in medical imaging and refers to the process of converting routine radiological images that are typically qualitatively interpreted to quantifiable descriptions of the tumor phenotypes and when combined with statistical analytics can improve the accuracy of clinical outcome prediction models. However, to understand the radiomic features and their correlation to molecular changes in the tumor, first, there is a need for the development of robust image analysis methods, software tools and statistical prediction models which is often limited in low- and middle-income countries (LMIC).

Aims:

The aim is to build a framework for machine learning of radiomic features of planning computed tomography (CT) and positron emission tomography (PET) using open source radiomics and data analytics platforms to make it widely accessible to clinical groups. The framework is tested in a small cohort to predict local disease failure following radiation treatment for head-and-neck cancer (HNC). The predictors were also compared with the existing Aerts HNC radiomics signature.

Settings and Design:

Retrospective analysis of patients with locally advanced HNC between 2017 and 2018 and 31 patients with both pre- and post-radiation CT and evaluation PET were selected.

Subjects and Methods:

Tumor volumes were delineated on baseline PET using the semi-automatic adaptive-threshold algorithm and propagated to CT; PyRadiomics features (total of 110 under shape/intensity/texture classes) were extracted. Two feature-selection methods were tested for model stability. Models were built based on least absolute shrinkage and selection operator-logistic and Ridge regression of the top pretreatment radiomic features and compared to Aerts' HNC-signature. Average model performance across all internal validation test folds was summarized by the area under the receiver operator curve (ROC).

Results:

Both feature selection methods selected CT features MCC (GLCM), SumEntropy (GLCM) and Sphericity (Shape) that could predict the binary failure status in the cross-validated group and achieved an AUC >0.7. However, models using Aerts' signature features (Energy, Compactness, GLRLM-GrayLevelNonUniformity and GrayLevelNonUniformity-HLH wavelet) could not achieve a clear separation between outcomes (AUC = 0.51–0.54).

Conclusions:

Radiomics pipeline included open-source workflows which makes it adoptable in LMIC countries. Additional independent validation of data is crucial for the implementation of radiomic models for clinical risk stratification.

Evolving Role and Translation of Radiomics and Radiogenomics in Adult and Pediatric Neuro-Oncology

Exponential technologic advancements in imaging, high-performance computing, and artificial intelligence, in addition to increasing access to vast amounts of diverse data, have revolutionized the role of imaging in medicine. Radiomics is defined as a high-throughput feature-extraction method that unlocks microscale quantitative data hidden within standard-of-care medical imaging. Radiogenomics is defined as the linkage between imaging and genomics information. Multiple radiomics and radiogenomics studies performed on conventional and advanced neuro-oncology image modalities show that they have the potential to differentiate pseudoprogression from true progression, classify tumor subgroups, and predict recurrence, survival, and mutation status with high accuracy. In this article, we outline the technical steps involved in radiomics and radiogenomics analyses with the use of artificial intelligence methods and review current applications in adult and pediatric neuro-oncology.

Technical overview of magnetic resonance fingerprinting and its applications in radiation therapy

Magnetic resonance fingerprinting (MRF) is an emerging imaging technique for rapid and simultaneous quantification of multiple tissue properties. The technique has been developed for quantitative imaging of different organs. The obtained quantitative measures have the potential to improve multiple steps of a typical radiotherapy workflow and potentially further improve integration of magnetic resonance imaging guided clinical decision making. In this review paper, we first provide a technical overview of the MRF method from data acquisition to postprocessing, along with recent development in advanced reconstruction methods. We further discuss critical aspects that could influence its usage in radiation therapy, such as accuracy and precision, repeatability and reproducibility, geometric distortion, and motion robustness. Finally, future directions for MRF application in radiation therapy are discussed.

Radiomics Analysis of Fat-Saturated T2-Weighted MRI Sequences for the Prediction of Prognosis in Soft Tissue Sarcoma of the Extremities and Trunk Treated With Neoadjuvant Radiotherapy

Purpose

To create a prognostic prediction radiomics model for soft tissue sarcoma (STS) of the extremities and trunk treated with neoadjuvant radiotherapy.

Methods

This study included 62 patients with STS of the extremities and trunk who underwent magnetic resonance imaging (MRI) before neoadjuvant radiotherapy. After tumour segmentation and preprocessing, 851 radiomics features were extracted. The radiomics score was constructed according to the least absolute shrinkage and selection operator (LASSO) method. Survival analysis (disease-free survival; DFS) was performed using the log-rank test and Cox’s proportional hazards regression model. The nomogram model was established based on the log-rank test and Cox regression model. Harrell’s concordance index (C-index), calibration curve and receiver operating characteristic (ROC) curve analysis were used to evaluate the prognostic factors. The clinical utility of the model was assessed by decision curve analysis (DCA).

Results

The univariate survival analysis showed that tumour location (p = 0.032), clinical stage (p = 0.022), tumour size (p = 0.005) and the radiomics score were correlated with DFS (p < 0.05). The multivariate analysis showed that tumour location, tumour size, and the radiomics score were independent prognostic factors for DFS (p < 0.05). The combined clinical-radiomics model based on the multivariate analysis showed the best predictive ability for DFS (C-index: 0.781; Area Under Curve: 0.791). DCA revealed that the use of the radiomics score-based nomogram was associated with better benefit gains relative to the prediction of 2-year DFS events than other models in the threshold probability range between 0.12 and 0.38.

Conclusion

The radiomics score from pretreatment MRI is an independent prognostic factor for DFS in patients with STS of the extremities and trunk. The radiomics score-based nomogram could improve prognostic stratification ability and thus contribute to individualized therapy for STS patients.

A deep look into radiomics

Radiomics is a process that allows the extraction and analysis of quantitative data from medical images. It is an evolving field of research with many potential applications in medical imaging. The purpose of this review is to offer a deep look into radiomics, from the basis, deeply discussed from a technical point of view, through the main applications, to the challenges that have to be addressed to translate this process in clinical practice. A detailed description of the main techniques used in the various steps of radiomics workflow, which includes image acquisition, reconstruction, pre-processing, segmentation, features extraction and analysis, is here proposed, as well as an overview of the main promising results achieved in various applications, focusing on the limitations and possible solutions for clinical implementation. Only an in-depth and comprehensive description of current methods and applications can suggest the potential power of radiomics in fostering precision medicine and thus the care of patients, especially in cancer detection, diagnosis, prognosis and treatment evaluation.

The Wavelia Microwave Breast Imaging system-tumour discriminating features and their clinical usefulness

OBJECTIVE

The Wavelia Microwave Breast Imaging (MBI) system, based on non-ionising imaging technology, has demonstrated exciting potential in the detection and localisation of breast pathology in symptomatic patients. In this study, the ability of the system to accurately estimate the size and likelihood of malignancy of breast lesions is detailed, and its clinical usefulness determined.

METHODS

Institutional review board and Health Products Regulatory Authority (HPRA) approval were obtained. Patients were recruited from the symptomatic unit to three groups; breast cancer (Group-1), unaspirated cysts (Group-2) and biopsied benign lesions (Group-3). MBI, radiological and histopathological findings were reviewed. MBI size estimations were compared with the sizes determined by conventional imaging and histopathology. A Quadratic Discriminant Analysis (QDA) classifier was trained in a 3D feature space to discriminate malignant from benign lesions. An independent review was performed by two independent breast radiologists.

RESULTS

24 patients (11 Group-1, 8 Group-2 and 5 Group-3) underwent MBI. The Wavelia system was more accurate than conventional imaging in size estimation of breast cancers. The QDA accurately separated benign from malignant breast lesions in 88.5% of cases. The addition of MBI and the Wavelia malignancy risk calculation was deemed useful by the two radiologists in 70.6% of cases.

CONCLUSION

The results from this MBI investigation demonstrate the potential of this novel system in estimating size and malignancy risk of breast lesions. This system holds significant promise as a potential non-invasive, comfortable, and harmless adjunct for breast cancer diagnosis. Further larger studies are under preparation to validate the findings of this study.

ADVANCES IN KNOWLEDGE

This study details the potential of the Wavelia MBI system in delineating size and malignancy risk of benign and malignant breast lesions in a symptomatic cohort. The usefulness of the Wavelia system is assessed in the clinical setting.

Prognostic role of pre-treatment magnetic resonance imaging (MRI)-based radiomic analysis in effectively cured head and neck squamous cell carcinoma (HNSCC) patients

OBJECTIVES

To identify and validate baseline magnetic resonance imaging (b-MRI) radiomic features (RFs) as predictors of disease outcomes in effectively cured head and neck squamous cell carcinoma (HNSCC) patients.

MATERIALS AND METHODS

Training set (TS) and validation set (VS) were retrieved from preexisting datasets (HETeCo and BD2Decide trials, respectively). Only patients with both pre- and post-contrast enhancement T1 and T2-weighted b-MRI and at least 2 years of follow-up (FUP) were selected. The combination of the best extracted RFs was used to classify low risk (LR) vs. high risk (HR) of disease recurrence. Sensitivity, specificity, and area under the curve (AUC) of the radiomic model were computed on both TS and VS. Overall survival (OS) and 5-year disease-free survival (DFS) Kaplan-Meier (KM) curves were compared for LR vs. HR. The radiomic-based risk class was used in a multivariate Cox model, including well-established clinical prognostic factors (TNM, sub-site, human papillomavirus [HPV]).

RESULTS

In total, 57 patients of TS and 137 of VS were included. Three RFs were selected for the signature. Sensitivity of recurrence risk classifier was 0.82 and 0.77, specificity 0.78 and 0.81, AUC 0.83 and 0.78 for TS and VS, respectively. VS KM curves for LR vs. HR groups significantly differed both for 5-year DFS (p<.0001) and OS (p=.0004). A combined model of RFs plus TNM improved prognostic performance as compared to TNM alone, both for VS 5-year DFS (C-index: 0.76 vs. 0.60) and OS (C-index: 0.74 vs. 0.64).

CONCLUSIONS

Radiomics of b-MRI can help to predict recurrence and survival outcomes in HNSCC.

Machine learning for the prediction of pathologic pneumatosis intestinalis

BACKGROUND

The radiographic finding of pneumatosis intestinalis can indicate a spectrum of underlying processes ranging from a benign finding to a life-threatening condition. Although radiographic pneumatosis intestinalis is relatively common, there is no validated clinical tool to guide surgical management.

METHODS

Using a retrospective cohort of 300 pneumatosis intestinalis cases from a single institution, we developed 3 machine learning models for 2 clinical tasks: (1) the distinction of benign from pathologic pneumatosis intestinalis cases and (2) the determination of patients who would benefit from an operation. The 3 models are (1) an imaging model based on radiomic features extracted from computed tomography scans, (2) a clinical model based on clinical variables, and (3) a combination model using both the imaging and clinical variables.

RESULTS

The combination model achieves an area under the curve of 0.91 (confidence interval: 0.87-0.94) for task I and an area under the curve of 0.84 (confidence interval: 0.79-0.88) for task II. The combination model significantly (P < .05) outperforms the imaging model and the clinical model for both tasks. The imaging model achieves an area under the curve of 0.72 (confidence interval: 0.57-0.87) for task I and 0.68 (confidence interval: 0.61-0.74) for task II. The clinical model achieves an area under the curve of 0.87 (confidence interval: 0.83-0.91) for task I and 0.76 (confidence interval: 0.70-0.81) for task II.

CONCLUSION

This study suggests that combined radiographic and clinical features can identify pathologic pneumatosis intestinalis and aid in patient selection for surgery. This tool may better inform the surgical decision-making process for patients with pneumatosis intestinalis.

Artificial intelligence in medical imaging: A radiomic guide to precision phenotyping of cardiovascular disease

ABSTRACT

Rapid technological advances in non-invasive imaging, coupled with the availability of large data sets and the expansion of computational models and power, have revolutionized the role of imaging in medicine. Non-invasive imaging is the pillar of modern cardiovascular diagnostics, with modalities such as cardiac computed tomography (CT) now recognized as first-line options for cardiovascular risk stratification and the assessment of stable or even unstable patients. To date, cardiovascular imaging has lagged behind other fields, such as oncology, in the clinical translational of artificial intelligence (AI)-based approaches. We hereby review the current status of AI in non-invasive cardiovascular imaging, using cardiac CT as a running example of how novel machine learning (ML)-based radiomic approaches can improve clinical care. The integration of ML, deep learning, and radiomic methods has revealed direct links between tissue imaging phenotyping and tissue biology, with important clinical implications. More specifically, we discuss the current evidence, strengths, limitations, and future directions for AI in cardiac imaging and CT, as well as lessons that can be learned from other areas. Finally, we propose a scientific framework in order to ensure the clinical and scientific validity of future studies in this novel, yet highly promising field. Still in its infancy, AI-based cardiovascular imaging has a lot to offer to both the patients and their doctors as it catalyzes the transition towards a more precise phenotyping of cardiovascular disease.

Delta-volume radiomics of induction chemotherapy to predict outcome of subsequent chemoradiotherapy for locally advanced hypopharyngeal cancer

INTRODUCTION

Induction chemotherapy (IC) followed by concurrent chemoradiotherapy (CCRT) is recommended for larynx-preserving treatment of locally advanced hypopharyngeal cancer (LAHC). However, the conventional evaluation of response is not robust enough to predict the outcome of subsequent treatments. This study aimed to develop an imaging biomarker using changes in radiomic features in invasive tumor front (ITF) by IC to predict treatment outcome of subsequent CCRT in LAHC.

METHODS

From 2006 to 2018, 59 computed tomography (CT) scan images before and after IC in patients with LAHC were used to contour the gross tumor volumes (GTVs). A total of 48 delta-volume radiomics features were acquired from the absolute spatial difference of GTVs (delta-GTV) before and after IC, conceptually representing a consistent portion of ITF. Least absolute shrinkage and selection operator regression (LASSO) was used to select features for establishing the model generating radiomic score (R score).

RESULTS

A model including 5 radiomic features from delta-GTV to predict better progression-free survival (PFS) of patients receiving subsequent CCRT was established. The R score was validated with all datasets (area under the curve 0.77). Low R score (<-0.16) was associated with improved PFS (p < 0.05).

CONCLUSIONS

The established radiomic model for ITF from radiomic features of delta-GTV after IC might be a potential imaging biomarker for predicting clinical outcome of subsequent CCRT in LAHC.

Combining computed tomography and biologically effective dose in radiomics and deep learning improves prediction of tumor response to robotic lung stereotactic body radiation therapy

PURPOSE

The aim of this study is to improve the performance of machine learning (ML) models in predicting response of non-small cell lung cancer (NSCLC) to stereotactic body radiation therapy (SBRT) by integrating image features from pre-treatment computed tomography (CT) with features from the biologically effective dose (BED) distribution.

MATERIALS AND METHODS

Image features, consisting of crafted radiomic features or machine-learned features extracted using a convolutional neural network, were calculated from pre-treatment CT data and from dose distributions converted into BED for 80 NSCLC lesions over 76 patients treated with robotic guided SBRT. ML models using different combinations of features were trained to predict complete or partial response according to response criteria in solid tumors, including radiomics CT (Rad_CT ), radiomics CT and BED (Rad_CT,BED ), deep learning (DL) CT (DL_CT ), and DL CT and BED (DL_CT,BED ). Training of ML included feature selection by neighborhood component analysis followed by ensemble ML using robust boosting. A model was considered as acceptable when the sum of average sensitivity and specificity on test data in repeated cross validations was at least 1.5.

RESULTS

Complete or partial response occurred in 58 out of 80 lesions. The best models to predict the tumor response were those using BED variables, achieving significantly better area under curve (AUC) and accuracy than those using only features from CT, including a Rad_CT,BED model using three radiomic features from BED, which scored an accuracy of 0.799 (95% confidence intervals (0.75-0.85)) and AUC of 0.773 (0.688-0.846), and a DL_CT,BED model also using three variables with an accuracy of 0.798 (0.649-0.829) and AUC of 0.812 (0.755-0.867).

CONCLUSION

According to our results, the inclusion of BED features improves the response prediction of ML models for lung cancer patients undergoing SBRT, regardless of the use of radiomic or DL features.

T2-FLAIR, DWI and DKI radiomics satisfactorily predicts histological grade and Ki-67 proliferation index in gliomas

OBJECTIVE

To build highly predictive performance models for glioma stratification with radiomics features from non-invasive MRI.

METHODS

T2-weighted fluid-attenuated inversion recovery (T2-FLAIR) imaging, diffusion-weighted MRI and diffusion kurtosis imaging were retrospectively collected for 139 glioma cases (2 with grade I, 67 with grade II, 36 with grade III and 34 with grade IV disease). Multi-parameter maps, including the apparent diffusion coefficient (ADC), mean diffusion coefficient (Dmean), fractional anisotropy (FA), and mean kurtosis (MK), were co-registered to T2-FLAIR, and 431 radiomics features for each were extracted within manually segmented tumour regions. Partial correlation analysis revealed correlations between radiomics features and glioma stratification factors (tumour grades and Ki-67 LI). Predictive models were built with adjusted-imbalanced logistic regression.

RESULTS

MK radiomics features were more closely correlated with glioma stratification than other modalities analysed. The maximum R in MK was 0.52 for tumour grade and 0.56 for Ki-67 index (compared with 0.36 and 0.34 in FA, and 0.43 and 0.37 in ADC, and 0.48 and 0.42 in T2-FLAIR). The best predictive models for grade II vs. III, grade II vs. IV, low-grade vs. high-grade gliomas and positive vs. highly positive Ki-67 LI were obtained by combining multi-parameter MR radiomics features with AUCs of 0.858, 0.966, 0.853 and 0.870, respectively. However, for grade III vs. IV gliomas, the model obtained from MK radiomics features achieved the highest AUC (0.947), with excellent sensitivity and specificity.

CONCLUSION

Compared with the other modalities, MK showed closer correlations with tumour grade and Ki-67 LI. Combined radiomics models integrating multi-parameter MRI allow for the generation of highly predictive models for glioma stratification.

Radiomics of hepatocellular carcinoma: promising roles in patient selection, prediction, and assessment of treatment response

Radiomics refers to the process of conversion of conventional medical images into quantifiable data ("features") which can be further mined to reveal complex patterns and relationships between the voxels in the image. These high throughput features can potentially reflect the histology of biologic tissues at macroscopic and microscopic levels. Several studies have investigated radiomics of hepatocellular carcinoma (HCC) before and after treatment. HCC is a heterogeneous disease with diverse phenotypical and genotypical landscape. Due to this inherent heterogeneity, HCC lesions can manifest variable aggressiveness with different response to treatment options, including the newer targeted therapies. Hence, radiomics can be used as a potential tool to enable patient selection for therapies and to predict response to treatments and outcome. Additionally, radiomics may serve as a tool for earlier and more efficient assessment of response to treatment. Radiomics, radiogenomics, and radio-immunoprofiling and their potential roles in management of patients with HCC will be discussed and critically reviewed in this article.

Distinct tumor signatures using deep learning-based characterization of the peritumoral microenvironment in glioblastomas and brain metastases

Tumor types are classically distinguished based on biopsies of the tumor itself, as well as a radiological interpretation using diverse MRI modalities. In the current study, the overarching goal is to demonstrate that primary (glioblastomas) and secondary (brain metastases) malignancies can be differentiated based on the microstructure of the peritumoral region. This is achieved by exploiting the extracellular water differences between vasogenic edema and infiltrative tissue and training a convolutional neural network (CNN) on the Diffusion Tensor Imaging (DTI)-derived free water volume fraction. We obtained 85% accuracy in discriminating extracellular water differences between local patches in the peritumoral area of 66 glioblastomas and 40 metastatic patients in a cross-validation setting. On an independent test cohort consisting of 20 glioblastomas and 10 metastases, we got 93% accuracy in discriminating metastases from glioblastomas using majority voting on patches. This level of accuracy surpasses CNNs trained on other conventional DTI-based measures such as fractional anisotropy (FA) and mean diffusivity (MD), that have been used in other studies. Additionally, the CNN captures the peritumoral heterogeneity better than conventional texture features, including Gabor and radiomic features. Our results demonstrate that the extracellular water content of the peritumoral tissue, as captured by the free water volume fraction, is best able to characterize the differences between infiltrative and vasogenic peritumoral regions, paving the way for its use in classifying and benchmarking peritumoral tissue with varying degrees of infiltration.

A picture is worth a thousand words: a history of diagnostic imaging for lymphoma

The journey from early drawings of Thomas Hodgkin's patients to deep learning with radiomics in lymphoma has taken nearly 200 years, and in many ways, it parallels the journey of medicine. By tracing the history of imaging in clinical lymphoma practice, we can better understand the motivations for current imaging practices. The earliest imaging modalities of the 2D era each had varied, site-dependent sensitivity, and the improved accuracy of imaging studies allowed new diagnostic and therapeutic techniques. First, we review the initial imaging technologies that were applied to understand lymphoma spread and achieve practical guidance for the earliest lymphoma treatments. Next, in the 3D era, we describe how anatomical imaging advances replaced and complemented conventional modalities. Afterward, we discuss how the PET era scans were used to understand response of tumors to treatment and risk stratification. Finally, we discuss the emergence of radiomics as a promising area of research in personalized medicine. We are now able to identify involved lymph nodes and body sites both before and after treatment to offer patients improved treatment outcomes. As imaging methods continue to improve sensitivity, we will be able to use personalized medicine approaches to give targeted and highly focused therapies at even earlier time points, and ideally, we can obtain long-term disease control and cures for lymphomas.

Radiomics AI prediction for head and neck squamous cell carcinoma (HNSCC) prognosis and recurrence with target volume approach

OBJECTIVES

To evaluate the performance of radiomics features extracted from planning target volume (PTV) and gross tumor volume (GTV) in the prediction of the death prognosis and cancer recurrence rate for head and neck squamous cell carcinoma (HNSCC).

METHODS

188 HNSCC patients' planning CT images with radiotherapy structures sets were acquired from Cancer Imaging Archive (TCIA). The 3D slicer (v. 4.10.2) with the PyRadiomics extension (Computational Imaging and Bioinformatics Lab, Harvard medical School) was used to extract radiomics features from the radiotherapy planning images. An in-house developed deep learning artificial neural networks (DL-ANN) model was used to predict death prognosis and cancer recurrence rate based on the features extracted from GTV and PTV of the CT images.

RESULTS

The PTV radiomics features with DL-ANN model could achieve 77.7% accuracy with overall AUC equal to 0.934 and 0.932 when predicting HNSCC-related death prognosis and cancer recurrence respectively. Furthermore, the DL-ANN model can achieve an accuracy of 74.3% with AUC equal to 0.947 and 0.956 for the HNSCC-related death prognosis and cancer recurrence respectively using GTV features.

CONCLUSION

Using both GTV and PTV radiomics features in the DL-ANN model, can aid in predicting HNSCC-related death prognosis and cancer recurrence. Clinicians may find it helpful in formulating different treatment regimens and facilitate personized medicine based on the predicted outcome when performing GTV and PTV delineation.

ADVANCES IN KNOWLEDGE

Radiomics features of GTV and PTV are reliable prognosis and recurrence predicting tools, which may help clinicians in GTV and PTV delineation to facilitate delivery of personalized treatment.

A Novel Application of Unsupervised Machine Learning and Supervised Machine Learning-Derived Radiomics in Anterior Cruciate Ligament Rupture

Purpose

We aim to present an unsupervised machine learning application in anterior cruciate ligament (ACL) rupture and evaluate whether supervised machine learning-derived radiomics features enable prediction of ACL rupture accurately.

Patients and Methods

Sixty-eight patients were reviewed. Their demographic features were recorded, radiomics features were extracted, and the input dataset was defined as a collection of demographic features and radiomics features. The input dataset was automatically classified by the unsupervised machine learning algorithm. Then, we used a supervised machine learning algorithm to construct a radiomics model. The t-test and least absolute shrinkage and selection operator (LASSO) method were used for feature selection, random forest and support vector machine (SVM) were used as machine learning classifiers. For each model, the sensitivity, specificity, accuracy, and the area under the curve (AUC) of receiver operating characteristic (ROC) curves were calculated to evaluate model performance.

Results

In total, 5 demographic features were recorded and 106 radiomics features were extracted. By applying the unsupervised machine learning algorithm, patients were divided into 5 groups. Group 5 had the highest incidence of ACL rupture and left knee involvement. There were significant differences in left knee involvement among the groups. Forty-three radiomics features were extracted using t-test and 7 radiomics features were extracted using LASSO method. We found that the combination of LASSO selection method and random forest classifier has the highest sensitivity, specificity, accuracy, and AUC. The 7 radiomics features extracted by LASSO method were potential predictors for ACL rupture.

Conclusion

We validated the clinical application of unsupervised machine learning involving ACL rupture. Moreover, we found 7 radiomics features which were potential predictors for ACL rupture. The study indicated that radiomics could be a valuable method in the prediction of ACL rupture.

The prognostic value of MRI-detected extramural vascular invasion (mrEMVI) for rectal cancer patients treated with neoadjuvant therapy: a meta-analysis

OBJECTIVES

The purpose of this meta-analysis was to evaluate the prognostic value of MRI-detected extramural vascular invasion (mrEMVI) and mrEMVI after neoadjuvant therapy (ymrEMVI) in rectal cancer patients receiving neoadjuvant therapy.

METHODS

A systematic search of the PubMed, Web of Science, Embase, and Cochrane Library databases was carried out up to June 2020. Studies that evaluated mrEMVI, used treatment with neoadjuvant therapy, and reported survival were included. The time-to-event outcomes (OS and DFS rates) are expressed as hazard ratios (HRs) with 95% confidence intervals (CIs). If the HR was not reported in the study, it was calculated from the survival curve using methods according to Parmar's recommendation. The Newcastle-Ottawa scale was used to assess the methodological quality of the studies included in the meta-analysis.

RESULTS

A total of 2237 patients from 11 studies were included, and the pooled analysis of the overall results from eight studies showed that patients who were mrEMVI positive at baseline had significantly worse disease-free survival (DFS) (random-effects model: HR = 2.50 [1.84, 3.14]; Z = 5.83, p < 0.00001). The pooled analysis of the overall results from six studies showed that patients who were ymrEMVI positive following neoadjuvant therapy had significantly worse DFS (random-effects model: HR = 2.24 [1.73, 2.90], Z = 6.12, p < 0.00001). Patients with mrEMVI positivity at baseline were also associated with worse overall survival (OS) (random-effects model: HR = 1.93 [1.36, 2.73]; Z = 3.71, p < 0.00001).

CONCLUSION

mrEMVI and ymrEMVI positivity are poor prognostic factors for rectal cancer patients treated with neoadjuvant therapy. The precise evaluation of EMVI may contribute to designing individualised treatments and improving patient outcomes.

KEY POINTS

• Extramural vascular invasion (EMVI) is a prognostic factor for rectal cancer. • MRI can be used to evaluate EMVI status before (mrEMVI) and after neoadjuvant therapy (ymrEMVI). • The evaluation of mrEMVI and ymrEMVI in neoadjuvant therapy would provide an early assessment of patient prognosis.

Radiomic biomarkers of tumor immune biology and immunotherapy response

Immunotherapies are leading to improved outcomes for many cancers, including those with devastating prognoses. As therapies like immune checkpoint inhibitors (ICI) become a mainstay in treatment regimens, many concurrent challenges have arisen - for instance, delineating clinical responders from non-responders. Predicting response has proven to be difficult given a lack of consistent and accurate biomarkers, heterogeneity of the tumor microenvironment (TME), and a poor understanding of resistance mechanisms. For the most part, imaging data have remained an untapped, yet abundant, resource to address these challenges. In recent years, quantitative image analyses have highlighted the utility of medical imaging in predicting tumor phenotypes, prognosis, and therapeutic response. These studies have been fueled by an explosion of resources in high-throughput mining of image features (i.e. radiomics) and artificial intelligence. In this review, we highlight current progress in radiomics to understand tumor immune biology and predict clinical responses to immunotherapies. We also discuss limitations in these studies and future directions for the field, particularly if high-dimensional imaging data are to play a larger role in precision medicine.

A radiomics approach for automated diagnosis of ovarian neoplasm malignancy in computed tomography

This paper develops a two-dimensional (2D) radiomics approach with computed tomography (CT) to differentiate between benign and malignant ovarian neoplasms. A retrospective study was conducted from July 2017 to June 2019 for 134 patients with surgically-verified benign or malignant ovarian tumors. The patients were randomly divided in a ratio of 7:3 into two sets, namely a training set (of n = 95) and a test set (of n = 39). The ITK-SNAP software was used to delineate the regions of interest (ROI) associated with lesions of the largest diameters in plain CT image slices. Texture features were extracted by the Analysis Kit (AK) software. The training set was used to select the best features according to the maximum-relevance minimum-redundancy (mRMR) criterion, in addition to the algorithm of the least absolute shrinkage and selection operator (LASSO). Then, we employed a radiomics model for classification via multivariate logistic regression. Finally, we evaluated the overall performance of our method using the receiver operating characteristics (ROC), the DeLong test. and tested in an external validation test sample of patients of ovarian neoplasm. We created a radiomics prediction model from 14 selected features. The radiomic signature was found to be highly discriminative according to the area under the ROC curve (AUC) for both the training set (AUC = 0.88), and the test set (AUC = 0.87). The radiomics nomogram also demonstrated good calibration and differentiation for both the training (AUC = 0.95) and test (AUC = 0.96) samples. External validation tests gave a good performance in radiomic signature (AUC = 0.83) and radiomics nomogram (AUC = 0.95). The decision curve explicitly indicated the clinical usefulness of our nomogram method in the sense that it can influence major clinical events such as the ordering or abortion of other tests, treatments or invasive procedures. Our radiomics model based on plain CT images has a high diagnostic efficiency, which is helpful for the identification and prediction of benign and malignant ovarian neoplasms.

Advanced image analytics predicting clinical outcomes in patients with colorectal liver metastases: A systematic review of the literature

BACKGROUND

To better select patients with colorectal liver metastases (CRLM) for an optimal selection of treatment strategy (i.e. local, systemic or combined treatment) new prognostic models are warranted. In the last decade, radiomics has emerged as a field to create predictive models based on imaging features. This systematic review aims to investigate the current state and potential of radiomics to predict clinical outcomes in patients with CRLM.

METHODS

A comprehensive literature search was conducted in the electronic databases of PubMed, Embase, and Cochrane Library, according to PRISMA guidelines. Original studies reporting on radiomics predicting clinical outcome in patients diagnosed with CRLM were included. Clinical outcomes were defined as response to systemic treatment, recurrence of disease, and survival (overall, progression-free, disease-free). Primary outcome was the predictive performance of radiomics. A narrative synthesis of the results was made. Methodological quality was assessed using the radiomics quality score.

RESULTS

In 11 out of 14 included studies, radiomics was predictive for response to treatment, recurrence of disease, survival, or a combination of outcomes. Combining clinical parameters and radiomic features in multivariate modelling often improved the predictive performance. Different types of individual features were found prognostic. Noticeable were the contrary levels of heterogeneous and homogeneous features in patients with good response. The methodological quality as assessed by the radiomics quality score varied considerably between studies.

CONCLUSION

Radiomics appears a promising non-invasive method to predict clinical outcome and improve personalized decision-making in patients with CRLM. However, results were contradictory and difficult to compare. Standardized prospective studies are warranted to establish the added value of radiomics in patients with CRLM.

Association Between the Size and 3D CT-Based Radiomic Features of Breast Cancer Hepatic Metastasis

PURPOSE

To evaluate the effect of the anatomic size on 3D radiomic imaging features of the breast cancer hepatic metastases.

MATERIALS AND METHODS

CT scans of 81 liver metastases from 54 patients with breast cancer were evaluated. Ten most common 3D radiomic features from the histogram and gray level co-occurrence matrix (GLCM) categories were calculated for the hepatic metastases (HM) and compared to normal liver (NL). The effect of size was evaluated by using linear mixed-effects regression models. The effect of size on different radiomic features was analyzed for both liver lesions and background liver.

RESULTS

Three-dimensional radiomic features from GLCM demonstrate an important size dependence. The texture-feature size dependence was found to be different among feature categories and between the HM and NL, thus demonstrating a discriminatory power for the tissue type. Significant difference in the slope was found for GLCM homogeneity (NL slope = 0.004, slope difference 95% confidence interval [CI] 0.06-0.1, p <0.001), contrast (NL slope = 45, slope difference 95% CI 205-305, p <0.001), correlation (NL slope = 0.04, slope difference 95% CI 0.11-0.21, p <0.001), and dissimilarity (NL slope = 0.7, slope difference 95% CI 3.6-5.4, p <0.001). The GLCM energy (NL slope = 0.002, slope difference 95% CI -0.0005 to -0.0003, p <0.007), and entropy (NL slope = 1.49, slope difference 95% CI 0.07-0.52, p <0.009) exhibited size-dependence for both NL and HM, although demonstrating a difference in the slope between themselves.

CONCLUSION

Radiomic features of breast cancer hepatic metastasis exhibited significant correlation with tumor size. This finding demonstrates the complex behavior of imaging features and the need to include feature-specific properties into radiomic models.

A Clinical-Radiomics Nomogram Based on Computed Tomography for Predicting Risk of Local Recurrence After Radiotherapy in Nasopharyngeal Carcinoma

Purpose: We aimed to establish a nomogram model based on computed tomography (CT) imaging radiomic signature and clinical factors to predict the risk of local recurrence in nasopharyngeal carcinoma (NPC) after intensity-modulated radiotherapy (IMRT). Methods: This was a retrospective study consisting of 156 NPC patients treated with IMRT. Radiomics features were extracted from the gross tumor volume for nasopharynx (GTVnx) in pretreatment CT images for patients with or without local recurrence. Discriminative radiomics features were selected after t-test and the least absolute shrinkage and selection operator (LASSO) analysis. The most stable model was obtained to generate radiomics signature (Rad_Score) by using machine learning models including Logistic Regression, K-Nearest neighbor, Naive Bayes, Decision Tree, Stochastic Gradient Descent, Gradient Booting Tree and Linear Support Vector Classification. A nomogram for local recurrence was established based on Rad_Score and clinical factors. The predictive performance of nomogram was evaluated by discrimination ability and calibration ability. Decision Curve Analysis (DCA) was used to evaluate the clinical benefits of the multi-factor nomogram in predicting local recurrence after IMRT. Results: Local recurrence occurred in 42 patients. A total of 1,452 radiomics features were initially extracted and seven stable features finally selected after LASSO analysis were used for machine learning algorithm modeling to generate Rad_Score. The nomogram showed that the greater Rad_Score was associated with the higher risk of local recurrence. The concordance index, specificity and sensitivity in the training cohort were 0.931 (95%CI:0.8765-0.9856), 91.2 and 82.8%, respectively; whereas, in the validation cohort, they were 0.799 (95%CI: 0.6458-0.9515), 79.4, and 69.2%, respectively. Conclusion: The nomogram based on radiomics signature and clinical factors can predict the risk of local recurrence after IMRT in patients with NPC and provide evidence for early clinical intervention.

Combining machine learning and texture analysis to differentiate mediastinal lymph nodes in lung cancer patients

Evaluate whether texture analysis associated with machine learning approaches could differentiate between malignant and benign lymph nodes. A total 18 patients with lung cancer were selected, with 39 lymph nodes, being 15 malignant and 24 benign. Retrospective computed tomography scans were utilized both with and without contrast medium. The great differential of this work was the use of 15 textures from mediastinal lymph nodes, with five different physicians as operators. First and second order statistical textures such as gray level run length and co-occurrence matrix were extracted and applied to three different machine learning classifiers. The best machine learning classifier demonstrated a variability of less than 5% among operators. The support vector machine (SVM) classifier presented 95% of the area under the ROC curve (AUC) and 89% of sensitivity for sequences without contrast medium. SVM classifier presented 93% of AUC and 86% of sensitivity for sequences with contrast medium. Texture analysis and machine learning may be helpful in the differentiation between malign and benign lymph nodes. This study can aid the physician in diagnosis and staging of lymph nodes and potentially reduce the number of invasive analysis to histopathological confirmation.

Radiomics: The Next Frontier of Cardiac Computed Tomography

Radiomics uses advanced image analysis to extract massive amounts of quantitative information from digital images, which is not otherwise distinguishable to the human eye. The mined data can be used to explore and establish new and undiscovered correlations between these imaging features and clinical end points. Cardiac computed tomography (CT) is a first-line imaging modality for evaluating coronary artery disease and has a primary role in the assessment of cardiac structures. Conventional interpretation of cardiac CT images relies mostly on subjective and qualitative analysis, as well as basic geometric quantification. To date, some proof-of-concept studies have demonstrated the feasibility and diagnostic performance of cardiac CT radiomics analysis. This review describes the current literature on radiomics in cardiac CT and discusses its advantages, challenges, and future directions. Although much evidences are needed in this field, cardiac CT radiomics has a lot to offer to patients and physicians with potential to define cardiac disease phenotypes on imaging with higher precision.

Prediction of the early recurrence in spinal giant cell tumor of bone using radiomics of preoperative CT: Long-term outcome of 62 consecutive patients

•

Characteristics of 62 patients with spinal GCTB who underwent surgery.

•

A prognostic classification model was built based on features selected by SVM.

•

The combined histogram and texture features could predict recurrence of GCTB.

Objectives

To determine if radiomics analysis based on preoperative computed tomography (CT) can predict early postoperative recurrence of giant cell tumor of bone (GCTB) in the spine.

Methods

In a retrospective review, 62 patients with pathologically confirmed spinal GCTB from March 2008 to February 2018, with a minimum follow-up of 24 months, were identified. The mean follow-up was 73.7 months (range, 28.7–152.1 months). The clinical information including age, gender, lesion location, multi-vertebral involvement, and surgical methods, were obtained. CT images acquired before the operation were retrieved for radiomics analysis. For each case, the tumor regions of interest (ROI) was manually outlined, and a total of 107 radiomics features were extracted. The features were selected via the sequential selection process by using the support vector machine (SVM), then used to construct classification models with Gaussian kernels. The differentiation between recurrence and non-recurrence groups was evaluated by ROC analysis, using 10-fold cross-validation.

Results

Of the 62 patients, 17 had recurrence with a recurrence rate of 27.4%. None of the clinical information was significantly different between the two groups. Patients receiving curettage had a higher recurrence rate (6/16 = 37.5%) compared to patients receiving TES (6/26 = 23.1%) or intralesional spondylectomy (5/20 = 25%). The final radiomics model was built using 10 selected features, which achieved an accuracy of 89% with AUC of 0.78.

Conclusions

The radiomics model developed based on pre-operative CT can achieve a high accuracy to predict the recurrence of spinal GCTB. Patients who have a high risk of early recurrence should be treated more aggressively to minimize recurrence.

Value of conventional magnetic resonance imaging texture analysis in the differential diagnosis of benign and borderline/malignant phyllodes tumors of the breast

BACKGROUND

The purpose of this study was to determine the potential value of magnetic resonance imaging (MRI) texture analysis (TA) in differentiating between benign and borderline/malignant phyllodes tumors of the breast.

METHODS

The preoperative MRI data of 25 patients with benign phyllodes tumors (BPTs) and 19 patients with borderline/malignant phyllodes tumors (BMPTs) were retrospectively analyzed. A gray-level histogram and gray-level cooccurrence matrix (GLCM) were used for TA with fat-suppressed T2-weighted imaging (FS-T2WI), diffusion-weighted imaging (DWI), apparent diffusion coefficient (ADC) images, and 2- and 7-min postcontrast T1W images on dynamic contrast-enhanced MRI (DCE-T1WI_2min and DCE-T1WI_7min) between BPTs and BMPTs. Independent sample t-test and Mann-Whitney U test were performed for intergroup comparison. A regression model was established by using binary logistic regression analysis, and receiver operating characteristic (ROC) curve analysis was carried out to evaluate diagnostic efficiency.

RESULTS

For ADC images, the texture parameters angular second moment (ASM), correlation, contrast, entropy and the minimum gray values of ADC images (ADC_Minimum) showed significant differences between the BPT group and BMPT group (all p<0.05). The parameter entropy of FS-T2WI and the maximum gray values and kurtosis of the tumor solid region of DCE-T1WI_7min also showed significant differences between these two groups. Except for ADC_Minimum, angular second moment of FS-T2WI (FS-T2WI_ASM), and the maximum gray values of DCE-T1WI_7min (DCE-T1WI_7min-Maximum) of the tumor solid region, the AUC values of other positive texture parameters mentioned above were greater than 0.75. Binary logistic regression analysis demonstrated that the contrast of ADC images (ADC_Contrast) and entropy of FS-T2WI (FS-T2WI_Entropy) could be considered independent texture variables for the differential diagnosis of BPTs and BMPTs. Combined, the AUC of these parameters was 0.891 (95% CI: 0.793-0.988), with a sensitivity of 84.2% and a specificity of up to 89.0%.

CONCLUSION

Texture analysis could be helpful in improving the diagnostic efficacy of conventional MR images in differentiating BPTs and BMPTs.

Prediction of femoral osteoporosis using machine-learning analysis with radiomics features and abdomen-pelvic CT: A retrospective single center preliminary study

Background

Osteoporosis has increased and developed into a serious public health concern worldwide. Despite the high prevalence, osteoporosis is silent before major fragility fracture and the osteoporosis screening rate is low. Abdomen-pelvic CT (APCT) is one of the most widely conducted medical tests. Artificial intelligence and radiomics analysis have recently been spotlighted. This is the first study to evaluate the prediction performance of femoral osteoporosis using machine-learning analysis with radiomics features and APCT.

Materials and methods

500 patients (M: F = 70:430; mean age, 66.5 ± 11.8yrs; range, 50–96 years) underwent both dual-energy X-ray absorptiometry and APCT within 1 month. The volume of interest of the left proximal femur was extracted and 41 radiomics features were calculated using 3D volume of interest analysis. Top 10 importance radiomic features were selected by the intraclass correlation coefficient and random forest feature selection. Study cohort was randomly divided into 70% of the samples as the training cohort and the remaining 30% of the sample as the validation cohort. Prediction performance of machine-learning analysis was calculated using diagnostic test and comparison of area under the curve (AUC) of receiver operating characteristic curve analysis was performed between training and validation cohorts.

Results

The osteoporosis prevalence of this study cohort was 20.8%. The prediction performance of the machine-learning analysis to diagnose osteoporosis in the training and validation cohorts were as follows; accuracy, 92.9% vs. 92.7%; sensitivity, 86.6% vs. 80.0%; specificity, 94.5% vs. 95.8%; positive predictive value, 78.4% vs. 82.8%; and negative predictive value, 96.7% vs. 95.0%. The AUC to predict osteoporosis in the training and validation cohorts were 95.9% [95% confidence interval (CI), 93.7%-98.1%] and 96.0% [95% CI, 93.2%-98.8%], respectively, without significant differences (P = 0.962).

Conclusion

Prediction performance of femoral osteoporosis using machine-learning analysis with radiomics features and APCT showed high validity with more than 93% accuracy, specificity, and negative predictive value.

Artificial intelligence in the diagnosis and management of hepatocellular carcinoma

Despite recent improvements in therapeutic interventions, hepatocellular carcinoma is still associated with a poor prognosis in patients with an advanced disease at diagnosis. Recently, significant progress has been made in image recognition through advances in the field of artificial intelligence (AI) (or machine learning), especially deep learning. AI is a multidisciplinary field that draws on the fields of computer science and mathematics for developing and implementing computer algorithms capable of maximizing the predictive accuracy from static or dynamic data sources using analytic or probabilistic models. Because of the multifactorial and complex nature of liver diseases, the machine learning approach to integrate multiple factors would appear to be an advantageous approach to improve the likelihood of making a precise diagnosis and predicting the response of treatment and prognosis of liver diseases. In this review, we attempted to summarize the potential use of AI in the diagnosis and management of liver diseases, especially hepatocellular carcinoma.