Radiomics: the bridge between medical imaging and personalized medicine. Nat Rev Clin Oncol. 2017;14:749-762. [PMID: 28975929 DOI: 10.1038/nrclinonc.2017.141]

The European Federation of Organisations for Medical Physics (EFOMP) White Paper: Big data and deep learning in medical imaging and in relation to medical physics profession

Big data and deep learning will profoundly change various areas of professions and research in the future. This will also happen in medicine and medical imaging in particular. As medical physicists, we should pursue beyond the concept of technical quality to extend our methodology and competence towards measuring and optimising the diagnostic value in terms of how it is connected to care outcome. Functional implementation of such methodology requires data processing utilities starting from data collection and management and culminating in the data analysis methods. Data quality control and validation are prerequisites for the deep learning application in order to provide reliable further analysis, classification, interpretation, probabilistic and predictive modelling from the vast heterogeneous big data. Challenges in practical data analytics relate to both horizontal and longitudinal analysis aspects. Quantitative aspects of data validation, quality control, physically meaningful measures, parameter connections and system modelling for the future artificial intelligence (AI) methods are positioned firmly in the field of Medical Physics profession. It is our interest to ensure that our professional education, continuous training and competence will follow this significant global development.

Magnetic Resonance, Vendor-independent, Intensity Histogram Analysis Predicting Pathologic Complete Response After Radiochemotherapy of Rectal Cancer

PURPOSE

The objective of this study is finding an intensity based histogram (IBH) signature to predict pathologic complete response (pCR) probability using only pre-treatment magnetic resonance (MR) and validate it externally in order to create a workflow for the external validation of an MR IBH signature and to apply the model out of the environment where it has been tuned. The impact of pCR and the final predictors on the survival outcome were also evaluated.

METHODS AND MATERIALS

Three centers using different MR scanners were involved in this retrospective study. The first center recruited 162 patients for model training, and the second and third centers provided 34 plus 25 patients for external validation. Patients provided written consent. Accrual period was from May 2008 to December 2014. After surgery pathologic response was defined. T2-weighted MR scans acquired before chemoradiation therapy (CRT) were used for analysis addressed on primary lesions. Images were pre-processed using Laplacian of Gaussian (LoG) filter with multiple σ, and first order intensity histogram-based features (kurtosis, skewness, and entropy) were extracted. Features selection was performed using Mann-Whitney test. Tumor staging (cT, cN) was added to build a logistic regression model and predict pCR. Model performance was evaluated with internal and external validation using area under the curve (AUC) of the receiver operator characteristic (ROC) and calibration with Hosmer-Lemeshow test. The linear cross-correlation matrix (Pearson's coefficient) and the variance inflation factor (VIF) were used to check the correlation and the co-linearity among the final predictors. The amount of the information added through the radiomics features was estimated by using the DeLong's test, and the impact of pCR and the final predictors on survival outcomes were evaluated through the Kaplan-Meier curves by using the log-rank test and the multivariate Cox model.

RESULTS

Candidate-to-analysis features were skewness (σ = 0.485, P value = .01) and entropy (σ = 0.344, P value < .05). Logistic regression analysis showed as significant covariates cT (P value < .01), skewness-σ = 0.485 (P value = .01), and entropy-σ = 0.344 (P value < .05). Model AUCs were 0.73 (internal) and 0.75 (external).

CONCLUSIONS

This MR-based, vendor-independent model can be helpful for predicting pCR probability in locally advanced rectal cancer (LARC) patients only using pre-treatment imaging.

Feasibility of CT radiomics to predict treatment response of individual liver metastases in esophagogastric cancer patients

In this study we investigate a CT radiomics approach to predict response to chemotherapy of individual liver metastases in patients with esophagogastric cancer (EGC). In eighteen patients with metastatic EGC treated with chemotherapy, all liver metastases were manually delineated in 3D on the pre-treatment and evaluation CT. From the pre-treatment CT scans 370 radiomics features were extracted per lesion. Random forest (RF) models were generated to discriminate partial responding (PR, >65% volume decrease, including 100% volume decrease), and complete remission (CR, only 100% volume decrease) lesions from other lesions. RF-models were build using a leave one out strategy where all lesions of a single patient were removed from the dataset and used as validation set for a model trained on the lesions of the remaining patients. This process was repeated for all patients, resulting in 18 trained models and one validation set for both the PR and CR datasets. Model performance was evaluated by receiver operating characteristics with corresponding area under the curve (AUC). In total 196 liver metastases were delineated on the pre-treatment CT, of which 99 (51%) lesions showed a decrease in size of more than 65% (PR). From the PR set a total of 47 (47% of RL, 24% of initial) lesions were no longer detected in CT scan 2 (CR). The RF-model for PR lesions showed an average training AUC of 0.79 (range: 0.74-0.83) and 0.65 (95% ci: 0.57-0.73) for the combined validation set. The RF-model for CR lesions had an average training AUC of 0.87 (range: 0.83-0.90) and 0.79 (95% ci 0.72-0.87) for the validation set. Our findings show that individual response of liver metastases varies greatly within and between patients. A CT radiomics approach shows potential in discriminating responding from non-responding liver metastases based on the pre-treatment CT scan, although further validation in an independent patient cohort is needed to validate these findings.

A radiomics-based nomogram for the preoperative prediction of posthepatectomy liver failure in patients with hepatocellular carcinoma

OBJECTIVES

To develop and validate a radiomics-based nomogram for the preoperative prediction of posthepatectomy liver failure (PHLF) in patients with hepatocellular carcinoma (HCC).

METHODS

One hundred twelve consecutive HCC patients who underwent hepatectomy were included in the study pool (training cohort: n = 80, validation cohort: n = 32), and another 13 patients were included in a pilot prospective analysis. A total of 713 radiomics features were extracted from portal-phase computed tomography (CT) images. A logistic regression was used to construct a radiomics score (Rad-score). Then a nomogram, including Rad-score and other risk factors, was built with a multivariate logistic regression model. The discrimination, calibration and clinical utility of nomogram were evaluated.

RESULTS

The Rad-score could predict PHLF with an AUC of 0.822 (95% CI, 0.726-0.917) in the training cohort and of 0.762 (95% CI, 0.576-0.948) in the validation cohort; however, the approach could not completely outmatch the existing methods (CP [Child-Pugh], MELD [Model of End Stage Liver Disease], ALBI [albumin-bilirubin]). The individual predictive nomogram that included the Rad-score, MELD and performance status (PS) showed better discrimination with an AUC of 0.864 (95% CI, 0.786-0.942), which was higher than the AUCs of the conventional methods (nomogram vs CP, MELD, and ALBI at P < 0.001, P < 0.005, and P < 0.005, respectively). In the validation cohort, the nomogram discrimination was also superior to those of the other three methods (AUC: 0.896; 95% CI, 0.774-1.000). The calibration curves showed good agreement in both cohorts, and the decision curve analysis of the entire cohort revealed that the nomogram was clinically useful. A pilot prospective analysis showed that the radiomics nomogram could predict PHLF with an AUC of 0.833 (95% CI, 0.591-1.000).

CONCLUSIONS

A nomogram based on the Rad-score, MELD, and PS can predict PHLF.

Radiomics and liquid biopsy in oncology: the holons of systems medicine

Abstract

Radiomics is a process of extraction and analysis of quantitative features from diagnostic images. Liquid biopsy is a test done on a sample of blood to look for cancer cells or for pieces of tumourigenic DNA circulating in the blood. Radiomics and liquid biopsy have great potential in oncology, since both are minimally invasive, easy to perform, and can be repeated in patient follow-up visits, enabling the extraction of valuable information regarding tumour type, aggressiveness, progression, and response to treatment. Both methods are in their infancy, with major evidence of application in lung and gastrointestinal cancer, while still undergoing evaluation in other cancer types. In this paper, the main oncologic applications of radiomics and liquid biopsy are reviewed, and a synergistic approach incorporating both tests for cancer diagnosis and follow-up is discussed within the context of systems medicine.

Teaching Points

• Radiomics is a process of extraction and analysis of quantitative features from diagnostic images.

• Most clinical applications of radiomics are in the field of oncologic imaging.

• Radiomics applies to all imaging modalities.

• A cluster of radiomic features is a “radiomic signature”.

• Machine learning may improve the efficacy of radiomics analysis.

Radiomics: the facts and the challenges of image analysis

Radiomics is an emerging translational field of research aiming to extract mineable high-dimensional data from clinical images. The radiomic process can be divided into distinct steps with definable inputs and outputs, such as image acquisition and reconstruction, image segmentation, features extraction and qualification, analysis, and model building. Each step needs careful evaluation for the construction of robust and reliable models to be transferred into clinical practice for the purposes of prognosis, non-invasive disease tracking, and evaluation of disease response to treatment. After the definition of texture parameters (shape features; first-, second-, and higher-order features), we briefly discuss the origin of the term radiomics and the methods for selecting the parameters useful for a radiomic approach, including cluster analysis, principal component analysis, random forest, neural network, linear/logistic regression, and other. Reproducibility and clinical value of parameters should be firstly tested with internal cross-validation and then validated on independent external cohorts. This article summarises the major issues regarding this multi-step process, focussing in particular on challenges of the extraction of radiomic features from data sets provided by computed tomography, positron emission tomography, and magnetic resonance imaging.

Radiomics: the facts and the challenges of image analysis

Radiomics is an emerging translational field of research aiming to extract mineable high-dimensional data from clinical images. The radiomic process can be divided into distinct steps with definable inputs and outputs, such as image acquisition and reconstruction, image segmentation, features extraction and qualification, analysis, and model building. Each step needs careful evaluation for the construction of robust and reliable models to be transferred into clinical practice for the purposes of prognosis, non-invasive disease tracking, and evaluation of disease response to treatment. After the definition of texture parameters (shape features; first-, second-, and higher-order features), we briefly discuss the origin of the term radiomics and the methods for selecting the parameters useful for a radiomic approach, including cluster analysis, principal component analysis, random forest, neural network, linear/logistic regression, and other. Reproducibility and clinical value of parameters should be firstly tested with internal cross-validation and then validated on independent external cohorts. This article summarises the major issues regarding this multi-step process, focussing in particular on challenges of the extraction of radiomic features from data sets provided by computed tomography, positron emission tomography, and magnetic resonance imaging.

Multiregional radiomics profiling from multiparametric MRI: Identifying an imaging predictor of IDH1 mutation status in glioblastoma

Purpose

Isocitrate dehydrogenase 1 (IDH1) has been proven as a prognostic and predictive marker in glioblastoma (GBM) patients. The purpose was to preoperatively predict IDH mutation status in GBM using multiregional radiomics features from multiparametric magnetic resonance imaging (MRI).

Methods

In this retrospective multicenter study, 225 patients were included. A total of 1614 multiregional features were extracted from enhancement area, non‐enhancement area, necrosis, edema, tumor core, and whole tumor in multiparametric MRI. Three multiregional radiomics models were built from tumor core, whole tumor, and all regions using an all‐relevant feature selection and a random forest classification for predicting IDH1. Four single‐region models and a model combining all‐region features with clinical factors (age, sex, and Karnofsky performance status) were also built. All models were built from a training cohort (118 patients) and tested on an independent validation cohort (107 patients).

Results

Among the four single‐region radiomics models, the edema model achieved the best accuracy of 96% and the best F1‐score of 0.75 while the non‐enhancement model achieved the best area under the receiver operating characteristic curve (AUC) of 0.88 in the validation cohort. The overall performance of the tumor‐core model (accuracy 0.96, AUC 0.86 and F1‐score 0.75) and the whole‐tumor model (accuracy 0.96, AUC 0.88 and F1‐score 0.75) was slightly better than the single‐regional models. The 8‐feature all‐region radiomics model achieved an improved overall performance of an accuracy 96%, an AUC 0.90, and an F1‐score 0.78. Among all models, the model combining all‐region imaging features with age achieved the best performance of an accuracy 97%, an AUC 0.96, and an F1‐score 0.84.

Conclusions

The radiomics model built with multiregional features from multiparametric MRI has the potential to preoperatively detect the IDH1 mutation status in GBM patients. The multiregional model built with all‐region features performed better than the single‐region models, while combining age with all‐region features achieved the best performance.

Effect of image reconstruction algorithms on volumetric and radiomic parameters of coronary plaques

BACKGROUND

Volumetric and radiomic analysis of atherosclerotic plaques on coronary CT angiography have been shown to predict high-risk plaque morphology and to predict patient outcomes. However, there is limited information whether image reconstruction algorithms and preprocessing steps (type of binning, number of bins used for discretization) may influence parameter values.

METHODS

We retrospectively identified 60 coronary lesions on coronary CT angiography (CTA). All images were reconstructed using filtered back projection (FBP), hybrid (HIR) and model-based (MIR) iterative reconstruction. Plaques were segmented manually on HIR images and copied to FBP and MIR images to ensure identical voxels were analyzed. Overall, 4 volumetric and 169 radiomic parameters were calculated. Intra-class correlation coefficient (ICC) was used to assess reproducibility between image reconstructions, while linear regression analysis was used to assess the effect of preprocessing steps done before calculating radiomic metrics.

RESULTS

All volumetric and radiomic metrics had ICC>0.90 except for first-order statistics: mode, harmonic mean, minimum (0.45, 0.76, 0.84; respectively) and gray level co-occurrence (GLCM) parameters: inverse difference sum and sum variance (0.01, 0.04; respectively). Among GLCM parameters 90% were significantly affected by the type of binning and 100% by the number of bins. In case of gray level run length matrix parameters 100% of metrics were affected by both preprocessing steps.

CONCLUSIONS

Volumetric and radiomic statistics are robust to image reconstruction algorithms. However, all radiomic variables were affected by preprocessing steps therefore, showing the need for standardization before being implemented into everyday clinical practice.

MRI radiomics analysis for predicting preoperative synchronous distant metastasis in patients with rectal cancer

OBJECTIVES

To investigate the value of MRI radiomics based on T2-weighted (T2W) images in predicting preoperative synchronous distant metastasis (SDM) in patients with rectal cancer.

METHODS

This retrospective study enrolled 177 patients with histopathology-confirmed rectal adenocarcinoma (123 patients in the training cohort and 54 in the validation cohort). A total of 385 radiomics features were extracted from pretreatment T2W images. Five steps, including univariate statistical tests and a random forest algorithm, were performed to select the best preforming features for predicting SDM. Multivariate logistic regression analysis was conducted to build the clinical and clinical-radiomics combined models in the training cohort. The predictive performance was validated by receiver operating characteristics curve (ROC) analysis and clinical utility implementing a nomogram and decision curve analysis.

RESULTS

Fifty-nine patients (33.3%) were confirmed to have SDM. Six radiomics features and four clinical characteristics were selected for predicting SDM. The clinical-radiomics combined model performed better than the clinical model in both the training and validation datasets. A threshold of 0.44 yielded an area under the ROC (AUC) value of 0.827 (95% confidence interval (CI), 0.6963-0.9580), a sensitivity of 72.2%, a specificity of 94.4%, and an accuracy of 87.0% in the validation cohort for the combined model. A clinical-radiomics nomogram and decision curve analysis confirmed the clinical utility of the combined model.

CONCLUSIONS

Our proposed clinical-radiomics combined model could be utilized as a noninvasive biomarker for identifying patients at high risk of SDM, which could aid in tailoring treatment strategies.

KEY POINTS

• T2WI-based radiomics analysis helps predict synchronous distant metastasis (SDM) of rectal cancer. • The clinical-radiomics combined model could be utilized as a noninvasive biomarker for predicting SDM. • Personalized treatment can be carried out with greater confidence based on the risk stratification for SDM in rectal cancer.

Radiomics model of contrast-enhanced computed tomography for predicting the recurrence of acute pancreatitis

OBJECTIVES

To predict the recurrence of acute pancreatitis (AP) by constructing a radiomics model of contrast-enhanced computed tomography (CECT) at AP first attack.

METHODS

We retrospectively enrolled 389 first-attack AP patients (271 in the primary cohort and 118 in the validation cohort) from three tertiary referral centers; 126 and 55 patients endured recurrent attacks in each cohort. Four hundred twelve radiomics features were extracted from arterial and venous phase CECT images, and clinical characteristics were gathered to develop a clinical model. An optimal radiomics signature was chosen using a multivariable logistic regression or support vector machine. The radiomics model was developed and validated by incorporating the optimal radiomics signature and clinical characteristics. The performance of the radiomics model was assessed based on its calibration and classification metrics.

RESULTS

The optimal radiomics signature was developed based on a multivariable logistic regression with 10 radiomics features. The classification accuracy of the radiomics model well predicted the recurrence of AP for both the primary and validation cohorts (87.1% and 89.0%, respectively). The area under the receiver operating characteristic curve (AUC) of the radiomics model was significantly better than that of the clinical model for both the primary (0.941 vs. 0.712, p = 0.000) and validation (0.929 vs. 0.671, p = 0.000) cohorts. Good calibration was observed for all the models (p > 0.05).

CONCLUSIONS

The radiomics model based on CECT performed well in predicting AP recurrence. As a quantitative method, radiomics exhibits promising performance in terms of alerting recurrent patients to potential precautions.

KEY POINTS

• The incidence of recurrence after an initial episode of acute pancreatitis is high, and quantitative methods for predicting recurrence are lacking. • The radiomics model based on contrast-enhanced computed tomography performed well in predicting the recurrence of acute pancreatitis. • As a quantitative method, radiomics exhibits promising performance in terms of alerting recurrent patients to the potential need to take precautions.

Prospects and challenges for clinical decision support in the era of big data

Recently, there has been burgeoning interest in developing more effective and robust clinical decision support systems (CDSSs) for oncology. This has been primarily driven by the demands for more personalized and precise medical practice in oncology in the era of so-called Big Data (BD); an era that promises to harness the power of large-scale data flow to revolutionize cancer treatment. This interest in BD analytics has created new opportunities as well as new unmet challenges. These include: routine aggregation and standardization of clinical data; patient privacy; transformation of current analytical approaches to handle such noisy and heterogeneous data; and expanded use of advanced statistical learning methods based on confluence of modern statistical methods and machine learning algorithms. In this review, we present the current status of CDSSs in oncology, the prospects and current challenges of BD analytics, and the promising role of integrated modern statistics and machine learning algorithms in predicting complex clinical endpoints, individualizing treatment rules, and optimizing dynamic personalized treatment regimens. We discuss issues pertaining to these topics and present application examples from an aggregate of experiences. We also discuss the role of human factors in improving the utilization and acceptance of such enhanced CDSSs and how to mitigate possible sources of human error to achieve optimal performance and wider acceptance.

Machine Learning in Neurooncology Imaging: From Study Request to Diagnosis and Treatment

OBJECTIVE

Machine learning has potential to play a key role across a variety of medical imaging applications. This review seeks to elucidate the ways in which machine learning can aid and enhance diagnosis, treatment, and follow-up in neurooncology.

CONCLUSION

Given the rapid pace of development in machine learning over the past several years, a basic proficiency of the key tenets and use cases in the field is critical to assessing potential opportunities and challenges of this exciting new technology.

Pre-treatment CT radiomics to predict 3-year overall survival following chemoradiotherapy of esophageal cancer

BACKGROUND

Radiomic features retrieved from standard CT-images have shown prognostic power in several tumor sites. In this study, we investigated the prognostic value of pretreatment CT radiomic features to predict overall survival of esophageal cancer patients after chemoradiotherapy.

MATERIAL AND METHODS

Two datasets of independent centers were analyzed, consisting of esophageal cancer patients treated with concurrent chemotherapy (Carboplatin/Paclitaxel) and 41.4Gy radiotherapy, followed by surgery if feasible. In total, 1049 radiomic features were calculated from the primary tumor volume. Recursive feature elimination was performed to select the 40 most relevant predictors. Using these 40 features and six clinical variables as input, two random forest (RF) models predicting 3-year overall survival were developed.

RESULTS

In total 165 patients from center 1 and 74 patients from center 2 were used. The radiomics-based RF model yielded an area under the curve (AUC) of 0.69 (95%CI 0.61-0.77), with the top-5 most important features for 3-year survival describing tumor heterogeneity after wavelet filtering. In the validation dataset, the RF model yielded an AUC of 0.61 (95%CI 0.47-0.75). Kaplan Meier plots were significantly different between risk groups in the training dataset (p = .027) and borderline significant in the validation dataset (p = .053). The clinical RF model yielded AUCs of 0.63 (95%CI 0.54-0.71) and 0.62 (95%CI 0.49-0.76) in the training and validation dataset, respectively. Risk groups did not reach a significant correlation with pathological response in the primary tumor.

CONCLUSIONS

A RF model predicting 3-year overall survival based on pretreatment CT radiomic features was developed and validated in two independent datasets of esophageal cancer patients. The radiomics model had better prognostic power compared to the model using standard clinical variables.

Potential of serum microRNAs as biomarkers of radiation injury and tools for individualization of radiotherapy

Due to tremendous technological advances, radiation oncologists are now capable of personalized treatment plans and deliver the dose in a highly precise manner. However, a crucial challenge is how to escalate radiation doses to cancer cells while reducing damage to surrounding healthy tissues. This determines the probability of achieving therapeutic success whilst safeguarding patients from complications. The current dose constraints rely on observational data. Therefore, incidental toxicity observed in a minority of patients limits the admissible dose thresholds for the whole population, theoretically narrowing down the curative potential of radiotherapy. Future tools for measurements of individual's radiosensitivity before and during treatment would allow proper treatment personalization. Variation in tissue tolerance is at least partially genetically-determined and recent progress in the field of molecular biology raises the possibility that novel assays will allow to predict the response to ionizing radiation. Recently, microRNAs have garnered interest as stable biomarkers of tumor radiation response and normal-tissue toxicity. Preclinical studies in mice and nonhuman primates have shown that serum circulating microRNAs can be used to accurately distinguish pre- and postirradiation states and predict the biological impact of high-dose irradiation. First reports from human studies are also encouraging, however biology-driven precision radiation oncology, which tailors treatment to individual patient's needs, still remains to be translated into clinical studies. In this review, we summarize current knowledge about the potential of serum microRNAs as biodosimeters and biomarkers for radiation injury to lung and hematopoietic cells.

A review on radiomics and the future of theranostics for patient selection in precision medicine

The growing complexity and volume of clinical data and the associated decision-making processes in oncology promote the advent of precision medicine. Precision (or personalised) medicine describes preventive and/or treatment procedures that take individual patient variability into account when proscribing treatment, and has been hindered in the past by the strict requirements of accurate, robust, repeatable and preferably non-invasive biomarkers to stratify both the patient and the disease. In oncology, tumour subtypes are traditionally measured through repeated invasive biopsies, which are taxing for the patient and are cost and labour intensive. Quantitative analysis of routine clinical imaging provides an opportunity to capture tumour heterogeneity non-invasively, cost-effectively and on large scale. In current clinical practice radiological images are qualitatively analysed by expert radiologists whose interpretation is known to suffer from inter- and intra-operator variability. Radiomics, the high-throughput mining of image features from medical images, provides a quantitative and robust method to assess tumour heterogeneity, and radiomics-based signatures provide a powerful tool for precision medicine in cancer treatment. This study aims to provide an overview of the current state of radiomics as a precision medicine decision support tool. We first provide an overview of the requirements and challenges radiomics currently faces in being incorporated as a tool for precision medicine, followed by an outline of radiomics' current applications in the treatment of various types of cancer. We finish with a discussion of possible future advances that can further develop radiomics as a precision medicine tool.

Deep learning for lung cancer prognostication: A retrospective multi-cohort radiomics study

BACKGROUND

Non-small-cell lung cancer (NSCLC) patients often demonstrate varying clinical courses and outcomes, even within the same tumor stage. This study explores deep learning applications in medical imaging allowing for the automated quantification of radiographic characteristics and potentially improving patient stratification.

METHODS AND FINDINGS

We performed an integrative analysis on 7 independent datasets across 5 institutions totaling 1,194 NSCLC patients (age median = 68.3 years [range 32.5-93.3], survival median = 1.7 years [range 0.0-11.7]). Using external validation in computed tomography (CT) data, we identified prognostic signatures using a 3D convolutional neural network (CNN) for patients treated with radiotherapy (n = 771, age median = 68.0 years [range 32.5-93.3], survival median = 1.3 years [range 0.0-11.7]). We then employed a transfer learning approach to achieve the same for surgery patients (n = 391, age median = 69.1 years [range 37.2-88.0], survival median = 3.1 years [range 0.0-8.8]). We found that the CNN predictions were significantly associated with 2-year overall survival from the start of respective treatment for radiotherapy (area under the receiver operating characteristic curve [AUC] = 0.70 [95% CI 0.63-0.78], p < 0.001) and surgery (AUC = 0.71 [95% CI 0.60-0.82], p < 0.001) patients. The CNN was also able to significantly stratify patients into low and high mortality risk groups in both the radiotherapy (p < 0.001) and surgery (p = 0.03) datasets. Additionally, the CNN was found to significantly outperform random forest models built on clinical parameters-including age, sex, and tumor node metastasis stage-as well as demonstrate high robustness against test-retest (intraclass correlation coefficient = 0.91) and inter-reader (Spearman's rank-order correlation = 0.88) variations. To gain a better understanding of the characteristics captured by the CNN, we identified regions with the most contribution towards predictions and highlighted the importance of tumor-surrounding tissue in patient stratification. We also present preliminary findings on the biological basis of the captured phenotypes as being linked to cell cycle and transcriptional processes. Limitations include the retrospective nature of this study as well as the opaque black box nature of deep learning networks.

CONCLUSIONS

Our results provide evidence that deep learning networks may be used for mortality risk stratification based on standard-of-care CT images from NSCLC patients. This evidence motivates future research into better deciphering the clinical and biological basis of deep learning networks as well as validation in prospective data.

A review of radiation genomics: integrating patient radiation response with genomics for personalised and targeted radiation therapy

Background

The success of radiation therapy for cancer patients is dependent on the ability to deliver a total tumouricidal radiation dose capable of eradicating all cancer cells within the clinical target volume, however, the radiation dose tolerance of the surrounding healthy tissues becomes the main dose-limiting factor. The normal tissue adverse effects following radiotherapy are common and significantly impact the quality of life of patients. The likelihood of developing these adverse effects following radiotherapy cannot be predicted based only on the radiation treatment parameters. However, there is evidence to suggest that some common genetic variants are associated with radiotherapy response and the risk of developing adverse effects. Radiation genomics is a field that has evolved in recent years investigating the association between patient genomic data and the response to radiation therapy. This field aims to identify genetic markers that are linked to individual radiosensitivity with the potential to predict the risk of developing adverse effects due to radiotherapy using patient genomic information. It also aims to determine the relative radioresponse of patients using their genetic information for the potential prediction of patient radiation treatment response.

Methods and materials

This paper reports on a review of recent studies in the field of radiation genomics investigating the association between genomic data and patients response to radiation therapy, including the investigation of the role of genetic variants on an individual’s predisposition to enhanced radiotherapy radiosensitivity or radioresponse.

Conclusion

The potential for early prediction of treatment response and patient outcome is critical in cancer patients to make decisions regarding continuation, escalation, discontinuation, and/or change in treatment options to maximise patient survival while minimising adverse effects and maintaining patients’ quality of life.

Artificial intelligence in medical imaging: threat or opportunity? Radiologists again at the forefront of innovation in medicine

One of the most promising areas of health innovation is the application of artificial intelligence (AI), primarily in medical imaging. This article provides basic definitions of terms such as “machine/deep learning” and analyses the integration of AI into radiology. Publications on AI have drastically increased from about 100–150 per year in 2007–2008 to 700–800 per year in 2016–2017. Magnetic resonance imaging and computed tomography collectively account for more than 50% of current articles. Neuroradiology appears in about one-third of the papers, followed by musculoskeletal, cardiovascular, breast, urogenital, lung/thorax, and abdomen, each representing 6–9% of articles. With an irreversible increase in the amount of data and the possibility to use AI to identify findings either detectable or not by the human eye, radiology is now moving from a subjective perceptual skill to a more objective science. Radiologists, who were on the forefront of the digital era in medicine, can guide the introduction of AI into healthcare. Yet, they will not be replaced because radiology includes communication of diagnosis, consideration of patient’s values and preferences, medical judgment, quality assurance, education, policy-making, and interventional procedures. The higher efficiency provided by AI will allow radiologists to perform more value-added tasks, becoming more visible to patients and playing a vital role in multidisciplinary clinical teams.

A radiogenomic dataset of non-small cell lung cancer

Medical image biomarkers of cancer promise improvements in patient care through advances in precision medicine. Compared to genomic biomarkers, image biomarkers provide the advantages of being non-invasive, and characterizing a heterogeneous tumor in its entirety, as opposed to limited tissue available via biopsy. We developed a unique radiogenomic dataset from a Non-Small Cell Lung Cancer (NSCLC) cohort of 211 subjects. The dataset comprises Computed Tomography (CT), Positron Emission Tomography (PET)/CT images, semantic annotations of the tumors as observed on the medical images using a controlled vocabulary, and segmentation maps of tumors in the CT scans. Imaging data are also paired with results of gene mutation analyses, gene expression microarrays and RNA sequencing data from samples of surgically excised tumor tissue, and clinical data, including survival outcomes. This dataset was created to facilitate the discovery of the underlying relationship between tumor molecular and medical image features, as well as the development and evaluation of prognostic medical image biomarkers.

Biliary Tract Cancer at CT: A Radiomics-based Model to Predict Lymph Node Metastasis and Survival Outcomes

Purpose To evaluate a radiomics model for predicting lymph node (LN) metastasis in biliary tract cancers (BTCs) and to determine its prognostic value for disease-specific and recurrence-free survival. Materials and Methods For this retrospective study, a radiomics model was developed on the basis of a primary cohort of 177 patients with BTC who underwent resection and LN dissection between June 2010 and December 2016. Radiomic features were extracted from portal venous CT scans. A radiomics signature was built on the basis of reproducible features by using the least absolute shrinkage and selection operator method. Multivariable logistic regression model was adopted to establish a radiomics nomogram. Nomogram performance was determined by its discrimination, calibration, and clinical usefulness. The model was internally validated in 70 consecutive patients with BTC between January 2017 and February 2018. Results The radiomics signature, composed of three LN-status-related features, was associated with LN metastasis in primary and validation cohorts (P < .001). The radiomics nomogram that incorporated radiomics signature and CT-reported LN status showed good calibration and discrimination in primary cohort (area under the curve, 0.81) and validation cohort (area under the curve, 0.80). Patients at high risk of LN metastasis portended lower disease-specific and recurrence-free survival than did those at low risk after surgery (both P < .001). High-risk LN metastasis was an independent preoperative predictor of disease-specific survival (hazard ratio, 3.37; P < .001) and recurrence-free survival (hazard ratio, 1.98; P = .003). Conclusion A radiomics model derived from portal phase CT of the liver has good performance for predicting lymph node metastasis in biliary tract cancer and may help to improve clinical decision making. © RSNA, 2018 Online supplemental material is available for this article. See also the editorial by Laghi and Voena in this issue.

A biomarker basing on radiomics for the prediction of overall survival in non-small cell lung cancer patients

BACKGROUND

This study aimed at predicting the survival status on non-small cell lung cancer patients with the phenotypic radiomics features obtained from the CT images.

METHODS

A total of 186 patients' CT images were used for feature extraction via Pyradiomics. The minority group was balanced via SMOTE method. The final dataset was randomized into training set (n = 223) and validation set (n = 75) with the ratio of 3:1. Multiple random forest models were trained applying hyperparameters grid search with 10-fold cross-validation using precision or recall as evaluation standard. Then a decision threshold was searched on the selected model. The final model was evaluated through ROC curve and prediction accuracy.

RESULTS

From those segmented images of 186 patients, 1218 features were obtained via feature extraction. The preferred model was selected with recall as evaluation standard and the optimal decision threshold was set 0.56. The model had a prediction accuracy of 89.33% and the AUC score was 0.9296.

CONCLUSION

A hyperparameters tuning random forest classifier had greater performance in predicting the survival status of non-small cell lung cancer patients, which could be taken for an automated classifier promising to stratify patients.

Predicting the response to neoadjuvant therapy for early-stage breast cancer: tumor-, blood-, and imaging-related biomarkers

Neoadjuvant therapy (NAT) has been used increasingly in patients with locally advanced or early-stage breast cancer. However, the accurate evaluation and prediction of response to NAT remain the great challenge. Biomarkers could prove useful to identify responders or nonresponders, or even to distinguish between early and delayed responses. These biomarkers could include markers from the tumor itself, such as versatile proteins, genes, and ribonucleic acids, various biological factors or peripheral blood cells, and clinical and pathological features. Possible predictive markers could also include multiple features from functional imaging, such as standard uptake values in positron emission tomography, apparent diffusion coefficient in magnetic resonance, or radiomics imaging biomarkers. In addition, cells that indirectly present the immune status of tumor cells and/or their host could also potentially be used as biomarkers, eg, tumor-infiltrating lymphocytes, tumor-associated macrophages, and myeloid-derived suppressor cells. Though numerous biomarkers have been widely investigated, only estrogen and/or progesterone receptors and human epidermal growth factor receptor have been proven to be reliable biomarkers to predict the response to NAT. They are the only biomarkers recommended in several international guidelines. The other aforementioned biomarkers warrant further validation studies. Some multigene profiling assays that are commercially available, eg, Oncotype DX and MammaPrint, should be used with caution when extrapolated to NAT settings. A panel of combined multilevel biomarkers might be able to predict the response to NAT more robustly than individual biomarkers. To establish such a panel and its prediction model, reliable methods and extensive clinical validation are warranted.

Comparison of radiomics machine-learning classifiers and feature selection for differentiation of sacral chordoma and sacral giant cell tumour based on 3D computed tomography features

OBJECTIVE

We aimed to identify optimal machine-learning methods for preoperative differentiation of sacral chordoma (SC) and sacral giant cell tumour (SGCT) based on 3D non-enhanced computed tomography (CT) and CT-enhanced (CTE) features.

METHODS

A total of 95 patients were divided into a training set and a validation set. Three best feature selection methods (Relief, least absolute shrinkage and selection operator (LASSO) and Random Forest (RF)) and three classification methods, including generalised linear models (GLM), support vector machines (SVM) and RF, were compared for their performance in distinguishing SC and SGCT. The performance of the radiomics model was investigated via area under the receiver-operating characteristic curve (AUC) and accuracy (ACC) analysis.

RESULTS

The selection method LASSO + classifier GLM had the highest AUC of 0.984 and ACC of 0.897 in the validating set, followed by Relief + GLM (AUC = 0.909, ACC = 0.862) and LASSO + SVM (AUC = 0.900, ACC = 0.862) based on CTE features. For CT features, RF + GLM had the highest AUC of 0.889, while LASSO + GLM achieved a high ACC of 0.793 in the validating set. Regardless of the methods, CTE features significantly outperformed those from CT for the differentiation of SC and SGCT (Z_AUC = -3.029, Z_ACC = -4.553; p < 0.05).

CONCLUSIONS

Our study demonstrated CTE features performed better than CT features. The selection method LASSO + classifier GLM had the best performance in differentiation of SC and SGCT, which could enhance the application of radiomics methods in sacral tumours.

KEY POINTS

• Sacral chordoma and sacral giant cell tumour are the two most common primary tumours of the sacrum with many common clinical and imaging characteristics. • A radiomics model helps clinicians to identify the histology of a sacral tumour. • CTE features should be preferred.

Development and validation of a radiomics signature for clinically significant portal hypertension in cirrhosis (CHESS1701): a prospective multicenter study

Clinically significant portal hypertension (CSPH) is associated with an incremental risk of esophageal varices and overt clinical decompensations. However, hepatic venous pressure gradient (HVPG) measurement, the gold standard for defining CSPH (HVPG≥10 mm Hg) is invasive and therefore not suitable for routine clinical practice. This study aims to develop and validate a radiomics-based model as a noninvasive method for accurate detection of CSPH in cirrhosis. The prospective multicenter diagnostic trial (CHESS1701, ClinicalTrials.gov identifier: NCT03138915) involved 385 patients with cirrhosis from five liver centers in China between August 2016 and September 2017. Patients who had both HVPG measurement and contrast-enhanced CT within 14 days prior to the catheterization were collected. The noninvasive radiomics model, termed rHVPG for CSPH was developed based on CT images in a training cohort consisted of 222 consecutive patients and the diagnostic performance was prospectively assessed in 163 consecutive patients in four external validation cohorts. rHVPG showed a good performance in detection of CSPH with a C-index of 0·849 (95%CI: 0·786-0·911). Application of rHVPG in four external prospective validation cohorts still gave excellent performance with the C-index of 0·889 (95%CI: 0·752-1·000, 0·800 (95%CI: 0·614-0·986), 0·917 (95%CI: 0·772-1·000), and 0·827 (95%CI: 0·618-1·000), respectively. Intraclass correlation coefficients for inter- and intra-observer agreement were 0·92-0·99 and 0·97-0·99, respectively. A radiomics signature was developed and prospectively validated as an accurate method for noninvasive detection of CSPH in cirrhosis. The tool of rHVPG assessment can facilitate the identification of CSPH rapidly when invasive transjugular procedure is not available.

Quantitative Identification of Nonmuscle-Invasive and Muscle-Invasive Bladder Carcinomas: A Multiparametric MRI Radiomics Analysis

BACKGROUND

Preoperative discrimination between nonmuscle-invasive bladder carcinomas (NMIBC) and the muscle-invasive ones (MIBC) is very crucial in the management of patients with bladder cancer (BC).

PURPOSE

To evaluate the discriminative performance of multiparametric MRI radiomics features for precise differentiation of NMIBC from MIBC, preoperatively.

STUDY TYPE

Retrospective, radiomics.

POPULATION

Fifty-four patients with postoperative pathologically proven BC lesions (24 in NMIBC and 30 in MIBC groups) were included.

FIELD STRENGTH/SEQUENCE

3.0T MRI/T₂ -weighted (T₂ W) and multi-b-value diffusion-weighted (DW) sequences.

ASSESSMENT

A total of 1104 radiomics features were extracted from carcinomatous regions of interest on T₂ W and DW images, and the apparent diffusion coefficient maps. Support vector machine with recursive feature elimination (SVM-RFE) and synthetic minority oversampling technique (SMOTE) were used to construct an optimal discriminative model, and its performance was evaluated and compared with that of using visual diagnoses by experts.

STATISTICAL TESTS

Chi-square test and Student's t-test were applied on clinical characteristics to analyze the significant differences between patient groups.

RESULTS

Of the 1104 features, an optimal subset involving 19 features was selected from T₂ W and DW sequences, which outperformed the other two subsets selected from T₂ W or DW sequence in muscle invasion discrimination. The best performance for the differentiation task was achieved by the SVM-RFE+SMOTE classifier, with averaged sensitivity, specificity, accuracy, and area under the curve of receiver operating characteristic of 92.60%, 100%, 96.30%, and 0.9857, respectively, which outperformed the diagnostic accuracy by experts.

DATA CONCLUSION

The proposed radiomics approach has potential for the accurate differentiation of muscle invasion in BC, preoperatively. The optimal feature subset selected from multiparametric MR images demonstrated better performance in identifying muscle invasiveness when compared with that from T₂ W sequence or DW sequence only.

LEVEL OF EVIDENCE

3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;49:1489-1498.

Radiomics improves efficiency for differentiating subclinical pheochromocytoma from lipid-poor adenoma: a predictive, preventive and personalized medical approach in adrenal incidentalomas

Objectives

This study aims to define a radiomic signature for pre-operative differentiation between subclinical pheochromocytoma (sPHEO) and lipid-poor adrenal adenoma (LPA) in adrenal incidentaloma. The goal was to apply a predictive, preventive, and personalized medical approach to the management of adrenal tumors.

Patients and methods

This retrospective study consisted of 265 consecutive patients (training cohort, 212 (LPA, 145; sPHEO, 67); validation cohort, 53 (LPA, 36; sPHEO, 17)). Computed tomography (CT) imaging features were evaluated, including long diameter (LD), short diameter (SD), pre-enhanced CT value (CT_pre), enhanced CT value (CT_post), shape, homogeneity, necrosis or cystic degeneration (N/C). Radiomic features were extracted and then were used to construct a radiomic signature (Rad-score) and radiomic nomogram. The area under the receiver operating characteristic curve (AUC) was used to evaluate their performance.

Results

Sixteen of three hundred forty candidate features were used to build a radiomic signature. The signature was significantly different between the sPHEO and LPA groups (AUC: training, 0.907; validation, 0.902). The radiomic nomogram based on enhanced CT features (M1) consisted of Rad-score, LD, SD, CT_pre, shape, homogeneity and N/C (AUC: training, 0.957; validation, 0.967). The pre-enhanced CT features based radiomic nomogram (M2) included Rad-score, LD, SD, CT_pre, shape, and homogeneity (AUC: training, 0.955; validation, 0.958).

Conclusions

Our radiomic nomograms based on pre-enhanced and enhanced CT images distinguished sPHEO from LPA. In addition, the promising result using pre-enhanced CT images for predictive diagnostics is important because patients could avoid the additional radiation and risk associated with enhanced CT.

Radiomics of US texture features in differential diagnosis between triple-negative breast cancer and fibroadenoma

Triple-negative breast cancer (TNBC) is sometimes mistaken for fibroadenoma due to its tendency to show benign morphology on breast ultrasound (US) albeit its aggressive nature. This study aims to develop a radiomics score based on US texture analysis for differential diagnosis between TNBC and fibroadenoma, and to evaluate its diagnostic performance compared with pathologic results. We retrospectively included 715 pathology-proven fibroadenomas and 186 pathology-proven TNBCs which were examined by three different US machines. We developed the radiomics score by using penalized logistic regression with a least absolute shrinkage and selection operator (LASSO) analysis from 730 extracted features consisting of 14 intensity-based features, 132 textural features and 584 wavelet-based features. The constructed radiomics score showed significant difference between fibroadenoma and TNBC for all three US machines (p < 0.001). Although the radiomics score showed dependency on the type of US machine, we developed more elaborate radiomics score for a subgroup in which US examinations were performed with iU22. This subsequent radiomics score also showed good diagnostic performance, even for BI-RADS category 3 or 4a lesions (AUC 0.782) which were presumed as probably benign or low suspicious of malignancy by radiologists. It was expected to assist radiologist’s diagnosis and reduce the number of invasive biopsies, although US standardization should be overcome before clinical application.

Incidental finding of a breast carcinoma on Ga-68-DOTA-1-Nal3-octreotide positron emission tomography/computed tomography performed for the evaluation of a pancreatic neuroendocrine tumor: A case report

RATIONALE

Functional imaging using radiolabeled somatostatin analogues plays an important role in the management of patients with neuroendocrine tumors, and it is a promising tool in the new era of theragnosis and personalized medicine.

PATIENTS CONCERNS

The authors present the case of a 63-year-old woman referred for evaluation of a suspected pancreatic neuroendocrine tumor by Ga-68-DOTA-1-Nal3-octreotide positron emission tomography/computed tomography (Ga-68-DOTA-NOC PET/CT).

DIAGNOSES

PET/CT confirmed increased uptake of Ga-68-DOTA-NOC in a pancreatic lesion compatible with hyperexpression of somatostatin receptors in a neuroendocrine tumor. Furthermore, PET/CT revealed increased uptake in a breast lesion and in lymphadenomegalies (less intense than in the pancreatic tumor), which conducted to the incidental diagnosis of a breast carcinoma with lymph node metastases.

INTERVENTIONS

For the breast cancer, the patient underwentneoadjuvant chemotherapy and anti-HER2 monoclonal antibody, after which she was submitted to surgery. Regarding thepancreatic neuroendocrine tumor, it was decided to maintain itunder surveillance.

OUTCOMES

Breast carcinomas are known to express somatostatin receptors and this is the first report of Ga-68-DOTA-NOC uptake in a breast tumor.

LESSONS

Ga-68-DOTA-NOC PET/CT could be useful for the management of breast cancer patients in the new era of theragnosis and personalized medicine.

Preoperative prediction of sentinel lymph node metastasis in breast cancer by radiomic signatures from dynamic contrast-enhanced MRI

BACKGROUND

Sentinel lymph node (SLN) status is an important prognostic factor for patients with breast cancer, which is currently determined in clinical practice by invasive SLN biopsy.

PURPOSE

To noninvasively predict SLN metastasis in breast cancer using dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) intra- and peritumoral radiomics features combined with or without clinicopathologic characteristics of the primary tumor.

STUDY TYPE

Retrospective.

POPULATION

A total of 163 breast cancer patients (55 positive SLN and 108 negative SLN).

FIELD STRENGTH/SEQUENCE

1.5T, T₁ -weighted DCE-MRI.

ASSESSMENT

A total of 590 radiomic features were extracted for each patient from both intratumoral and peritumoral regions of interest. To avoid overfitting, the dataset was randomly separated into a training set (∼67%) and a validation set (∼33%). The prediction models were built with the training set using logistic regression on the most significant radiomic features in the training set combined with or without clinicopathologic characteristics. The prediction performance was further evaluated in the independent validation set.

STATISTICAL TESTS

Mann-Whitney U-test, Spearman correlation, least absolute shrinkage selection operator (LASSO) regression, logistic regression, and receiver operating characteristic (ROC) analysis were performed.

RESULTS

Combining radiomic features with clinicopathologic characteristics, six features were automatically selected in the training set to establish the prediction model of SLN metastasis. In the independent validation set, the area under ROC curve (AUC) was 0.869 (NPV = 0.886). Using radiomic features alone in the same procedure, 4 features were selected and the validation set AUC was 0.806 (NPV = 0.824).

DATA CONCLUSION

This is the first attempt to demonstrate the feasibility of using DCE-MRI radiomics to predict SLN metastasis in breast cancer. Clinicopathologic characteristics improved the prediction performance. This study provides noninvasive methods to evaluate SLN status for guiding further treatment of breast cancer patients, and can potentially benefit those with negative SLN, by eliminating unnecessary invasive lymph node removal and the associated complications, which is a step further towards precision medicine.

LEVEL OF EVIDENCE

1 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;49:131-140.

A Deep Look Into the Future of Quantitative Imaging in Oncology: A Statement of Working Principles and Proposal for Change

The adoption of enterprise digital imaging, along with the development of quantitative imaging methods and the re-emergence of statistical learning, has opened the opportunity for more personalized cancer treatments through transformative data science research. In the last 5 years, accumulating evidence has indicated that noninvasive advanced imaging analytics (i.e., radiomics) can reveal key components of tumor phenotype for multiple lesions at multiple time points over the course of treatment. Many groups using homegrown software have extracted engineered and deep quantitative features on 3-dimensional medical images for better spatial and longitudinal understanding of tumor biology and for the prediction of diverse outcomes. These developments could augment patient stratification and prognostication, buttressing emerging targeted therapeutic approaches. Unfortunately, the rapid growth in popularity of this immature scientific discipline has resulted in many early publications that miss key information or use underpowered patient data sets, without production of generalizable results. Quantitative imaging research is complex, and key principles should be followed to realize its full potential. The fields of quantitative imaging and radiomics in particular require a renewed focus on optimal study design and reporting practices, standardization, interpretability, data sharing, and clinical trials. Standardization of image acquisition, feature calculation, and statistical analysis (i.e., machine learning) are required for the field to move forward. A new data-sharing paradigm enacted among open and diverse participants (medical institutions, vendors and associations) should be embraced for faster development and comprehensive clinical validation of imaging biomarkers. In this review and critique of the field, we propose working principles and fundamental changes to the current scientific approach, with the goal of high-impact research and development of actionable prediction models that will yield more meaningful applications of precision cancer medicine.

Diagnostic accuracy of MRI texture analysis for grading gliomas

PURPOSE

Texture analysis (TA) can quantify variations in surface intensity or patterns, including some that are imperceptible to the human visual system. The purpose of this study was to determine the diagnostic accuracy of radiomic based filtration-histogram TA to differentiate high-grade from low-grade gliomas by assessing tumor heterogeneity.

METHODS

Patients with a histopathological diagnosis of glioma and preoperative 3T MRI imaging were included in this retrospective study. A region of interest was manually delineated on post-contrast T1 images. TA was performed using commercially available research software. The histogram parameters including mean, standard deviation, entropy, mean of the positive pixels, skewness, and kurtosis were analyzed at spatial scaling factors ranging from 0 to 6 mm. The parameters were correlated with WHO glioma grade using Spearman correlation. Areas under the curve (AUC) were calculated using ROC curve analysis to distinguish tumor grades.

RESULTS

Of a total of 94 patients, 14 had low-grade gliomas and 80 had high-grade gliomas. Mean, SD, MPP, entropy and kurtosis each showed significant differences between glioma grades for different spatial scaling filters. Low and high-grade gliomas were best-discriminated using mean of 2 mm fine texture scale, with a sensitivity and specificity of 93% and 86% (AUC of 0.90).

CONCLUSIONS

Quantitative measurement of heterogeneity using TA can discriminate high versus low-grade gliomas. Radiomic data of texture features can provide complementary diagnostic information for gliomas.

Deep Learning and Radiomics predict complete response after neo-adjuvant chemoradiation for locally advanced rectal cancer

Treatment of locally advanced rectal cancer involves chemoradiation, followed by total mesorectum excision. Complete response after chemoradiation is an accurate surrogate for long-term local control. Predicting complete response from pre-treatment features could represent a major step towards conservative treatment. Patients with a T2-4 N0-1 rectal adenocarcinoma treated between June 2010 and October 2016 with neo-adjuvant chemoradiation from three academic institutions were included. All clinical and treatment data was integrated in our clinical data warehouse, from which we extracted the features. Radiomics features were extracted from the tumor volume from the treatment planning CT Scan. A Deep Neural Network (DNN) was created to predict complete response, as a methodological proof-of-principle. The results were compared to a baseline Linear Regression model using only the TNM stage as a predictor and a second model created with Support Vector Machine on the same features used in the DNN. Ninety-five patients were included in the final analysis. There were 49 males (52%) and 46 females (48%). Median tumour size was 48 mm (15-130). Twenty-two patients (23%) had pathologic complete response after chemoradiation. One thousand six hundred eighty-three radiomics features were extracted. The DNN predicted complete response with an 80% accuracy, which was better than the Linear Regression model (69.5%) and the SVM model (71.58%). Our model correctly predicted complete response after neo-adjuvant rectal chemoradiotherapy in 80% of the patients of this multicenter cohort. Our results may help to identify patients who would benefit from a conservative treatment, rather than a radical resection.

Exploration of PET and MRI radiomic features for decoding breast cancer phenotypes and prognosis

Radiomics is an emerging technology for imaging biomarker discovery and disease-specific personalized treatment management. This paper aims to determine the benefit of using multi-modality radiomics data from PET and MR images in the characterization breast cancer phenotype and prognosis. Eighty-four features were extracted from PET and MR images of 113 breast cancer patients. Unsupervised clustering based on PET and MRI radiomic features created three subgroups. These derived subgroups were statistically significantly associated with tumor grade (p = 2.0 × 10⁻⁶), tumor overall stage (p = 0.037), breast cancer subtypes (p = 0.0085), and disease recurrence status (p = 0.0053). The PET-derived first-order statistics and gray level co-occurrence matrix (GLCM) textural features were discriminative of breast cancer tumor grade, which was confirmed by the results of L2-regularization logistic regression (with repeated nested cross-validation) with an estimated area under the receiver operating characteristic curve (AUC) of 0.76 (95% confidence interval (CI) = [0.62, 0.83]). The results of ElasticNet logistic regression indicated that PET and MR radiomics distinguished recurrence-free survival, with a mean AUC of 0.75 (95% CI = [0.62, 0.88]) and 0.68 (95% CI = [0.58, 0.81]) for 1 and 2 years, respectively. The MRI-derived GLCM inverse difference moment normalized (IDMN) and the PET-derived GLCM cluster prominence were among the key features in the predictive models for recurrence-free survival. In conclusion, radiomic features from PET and MR images could be helpful in deciphering breast cancer phenotypes and may have potential as imaging biomarkers for prediction of breast cancer recurrence-free survival.

Automated analyses of breast scans taken with two types of medical imaging technologies can help oncologists decode clinically relevant features, a finding that could help personalize cancer diagnosis and treatment. Youngho Seo from the University of California, San Francisco, USA, and coworkers extracted 84 quantitative features from positron emission tomography and magnetic resonance imaging scans performed on 113 women with breast cancer. The researchers then applied data-characterization and pattern-recognition algorithms—which included machine-learning methods and engineered features coded by experts—to create classification models that helped uncover disease characteristics that were not obvious to the naked eye. These models successfully subdivided patients according to tumor grade, overall stage, cancer subtype and disease recurrence risk, providing proof of principle that radiomic analyses of this kind could provide valuable information for personalized management of breast cancer.

Radiomic signature as a predictive factor for lymph node metastasis in early-stage cervical cancer

BACKGROUND

Lymph node metastasis (LNM) is the principal risk factor for poor outcomes in early-stage cervical cancer. Radiomics may offer a noninvasive way for predicting the stage of LNM.

PURPOSE

To evaluate a radiomic signature of LN involvement based on sagittal T₁ contrast-enhanced (CE) and T₂ MRI sequences.

STUDY TYPE

Retrospective.

POPULATION

In all, 143 patients were randomly divided into two primary and validation cohorts with 100 patients in the primary cohort and 43 patients in the validation cohort.

FIELD STRENGTH/SEQUENCE

T₁ CE and T₂ MRI sequences at 3T.

ASSESSMENT

The gold standard of LN status was based on histologic results. A radiologist with 10 years of experience used the ITK-SNAP software for 3D manual segmentation. A senior radiologist with 15 years of experience validated all segmentations. The area under the receiver operating characteristics curve (ROC AUC), classification accuracy, sensitivity, and specificity were used between LNM and non-LNM groups.

STATISTICAL TESTS

A total of 970 radiomic features and seven clinical characteristics were extracted. Minimum redundancy / maximum relevance and support vector machine algorithms were applied to select features and construct a radiomic signature. The Mann-Whitney U-test and the chi-square test were used to test the performance of clinical characteristics and potential prognostic outcomes. The results were used to assess the quantitative discrimination performance of the SVM-based radiomic signature.

RESULTS

The radiomic signatures allowed good discrimination between LNM and non-LNM groups. The ROC AUC was 0.753 (95% confidence interval [CI], 0.656-0.850) in the primary cohort and 0.754 (95% CI, 0584-0.924) in the validation cohort.

DATA CONCLUSIONS

A multiple-sequence MRI radiomic signature can be used as a noninvasive biomarker for preoperative assessment of LN status and potentially influence the therapeutic decision-making in early-stage cervical cancer patients.

LEVEL OF EVIDENCE

3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;49:304-310.

Development and Validation of an MRI-Based Radiomics Signature for the Preoperative Prediction of Lymph Node Metastasis in Bladder Cancer

Background

Preoperative lymph node (LN) status is important for the treatment of bladder cancer (BCa). However, a proportion of patients are at high risk for inaccurate clinical nodal staging by current methods. Here, we report an accurate magnetic resonance imaging (MRI)-based radiomics signature for the individual preoperative prediction of LN metastasis in BCa.

Methods

In total, 103 eligible BCa patients were divided into a training set (n = 69) and a validation set (n = 34). And 718 radiomics features were extracted from the cancerous volumes of interest (VOIs) on T2-weighted MRI images. A radiomics signature was constructed using the least absolute shrinkage and selection operator (LASSO) algorithm in the training set, whose performance was assessed and then validated in the validation set. Stratified analyses were also performed. Based on the multivariable logistic regression analysis, a radiomics nomogram was developed incorporating the radiomics signature and selected clinical predictors. Discrimination, calibration and clinical usefulness of the nomogram were assessed.

Findings

Consisting of 9 selected features, the radiomics signature showed a favorable discriminatory ability in the training set with an AUC of 0.9005, which was confirmed in the validation set with an AUC of 0.8447. Encouragingly, the radiomics signature also showed good discrimination in the MRI-reported LN negative (cN0) subgroup (AUC, 0.8406). The nomogram, consisting of the radiomics signature and the MRI-reported LN status, showed good calibration and discrimination in the training and validation sets (AUC, 0.9118 and 0.8902, respectively). The decision curve analysis indicated that the nomogram was clinically useful.

Interpretation

The MRI-based radiomics nomogram has the potential to be used as a non-invasive tool for individualized preoperative prediction of LN metastasis in BCa. External validation is further required prior to clinical implementation.

The developing role of FDG PET imaging for prognostication and radiotherapy target volume delineation in non-small cell lung cancer

Advancements in functional imaging technology have allowed new possibilities in contouring of target volumes, monitoring therapy, and predicting treatment outcome in non-small cell lung cancer (NSCLC). Consequently, the role of ¹⁸F-fluorodeoxyglucose positron emission tomography (FDG PET) has expanded in the last decades from a stand-alone diagnostic tool to a versatile instrument integrated with computed tomography (CT), with a prominent role in lung cancer radiotherapy. This review outlines the most recent literature on developments in FDG PET imaging for prognostication and radiotherapy target volume delineation (TVD) in NSCLC. We also describe the challenges facing the clinical implementation of these developments and present new ideas for future research.

Gold nanoparticle-induced sonosensitization enhances the antitumor activity of ultrasound in colon tumor-bearing mice

PURPOSE

As a noninvasive and nonionizing radiation, ultrasound can be focused remotely, transferring acoustic energy deep in the body, thereby addressing the penetration depth barrier of the light-based therapies. In cancer therapy, the effectiveness of ultrasound can be enhanced by utilizing nanomaterials that exhibit sonosensitizing properties called as nanosonosensitizers. The gold nanoparticle (AuNP) has been recently presented as a potent nanosonosensitizer with the potential to simultaneously enhance both the thermal and mechanical interactions of ultrasound with the tissue of the human body. Accordingly, this paper attempts to evaluate the in vivo antitumor efficiency of ultrasound in combination with AuNP.

METHODS

BALB/c mice-bearing CT26 colorectal tumor model was intraperitoneally injected with AuNPs and then subjected to ultrasound irradiation (1 MHz; 2 W/cm2 ; 10 min) for three sessions. Furthermore, [18 F]FDG (2-deoxy-2-[18 F]fluoro-d-glucose) positron-emission tomography (PET) imaging was performed and the radiomic features from different feature categorizes were extracted to quantify the tumors' phenotype.

RESULTS

The tumors were dramatically shrunk and the mice appeared healthy over 21 days of study span without the evidence of relapse. The animals treated with AuNP + ultrasound exhibited an obvious decline in tumor metabolic parameters such as standard uptake value (SUV), total lesion glycolysis (TLG), and metabolic tumor volume (MTV) compared to other treatment groups.

CONCLUSION

These findings support the use of AuNP as a potent sonosensitizing agent with the potential to use the thermal and mechanical effects of ultrasound so as to cause damage to the focused tumor site, resulting in an improved antitumor efficacy.

The MR radiomic signature can predict preoperative lymph node metastasis in patients with esophageal cancer

PURPOSE

To assess the role of the MR radiomic signature in preoperative prediction of lymph node (LN) metastasis in patients with esophageal cancer (EC).

PATIENTS AND METHODS

A total of 181 EC patients were enrolled in this study between April 2015 and September 2017. Their LN metastases were pathologically confirmed. The first half of this cohort (90 patients) was set as the training cohort, and the second half (91 patients) was set as the validation cohort. A total of 1578 radiomic features were extracted from MR images (T2-TSE-BLADE and contrast-enhanced StarVIBE). The lasso and elastic net regression model was exploited for dimension reduction and selection of the feature space. The multivariable logistic regression analysis was adopted to identify the radiomic signature of pathologically involved LNs. The discriminating performance was assessed with the area under receiver-operating characteristic curve (AUC). The Mann-Whitney U test was adopted for testing the potential correlation of the radiomic signature and the LN status in both training and validation cohorts.

RESULTS

Nine radiomic features were selected to create the radiomic signature significantly associated with LN metastasis (p < 0.001). AUC of radiomic signature performance in the training cohort was 0.821 (95% CI: 0.7042-0.9376) and in the validation cohort was 0.762 (95% CI: 0.7127-0.812). This model showed good discrimination between metastatic and non-metastatic lymph nodes.

CONCLUSION

The present study showed MRI radiomic features that could potentially predict metastatic LN involvement in the preoperative evaluation of EC patients.

KEY POINTS

• The role of MRI in preoperative staging of esophageal cancer patients is increasing. • MRI radiomic features showed the ability to predict LN metastasis in EC patients. • ICCs showed excellent interreader agreement of the extracted MR features.

Voxel size and gray level normalization of CT radiomic features in lung cancer

Radiomic features are potential imaging biomarkers for therapy response assessment in oncology. However, the robustness of features with respect to imaging parameters is not well established. Previously identified potential imaging biomarkers were found to be intrinsically dependent on voxel size and number of gray levels (GLs) in a recent texture phantom investigation. Here, we validate the voxel size and GL in-phantom normalizations in lung tumors. Eighteen patients with non-small cell lung cancer of varying tumor volumes were analyzed. To compare with patient data, phantom scans were acquired on eight different scanners. Twenty four previously identified features were extracted from lung tumors. The Spearman rank (r_s) and interclass correlation coefficient (ICC) were used as metrics. Eight out of 10 features showed high (r_s > 0.9) and low (r_s < 0.5) correlations with number of voxels before and after normalizations, respectively. Likewise, texture features were unstable (ICC < 0.6) and highly stable (ICC > 0.8) before and after GL normalizations, respectively. We conclude that voxel size and GL normalizations derived from a texture phantom study also apply to lung tumors. This study highlights the importance and utility of investigating the robustness of radiomic features with respect to CT imaging parameters in radiomic phantoms.

The Role of PET-Based Radiomic Features in Predicting Local Control of Esophageal Cancer Treated with Concurrent Chemoradiotherapy

This study was designed to evaluate the predictive performance of ¹⁸F-fluorodeoxyglucose positron emission tomography (PET)-based radiomic features for local control of esophageal cancer treated with concurrent chemoradiotherapy (CRT). For each of the 30 patients enrolled, 440 radiomic features were extracted from both pre-CRT and mid-CRT PET images. The top 25 features with the highest areas under the receiver operating characteristic curve for identifying local control status were selected as discriminative features. Four machine-learning methods, random forest (RF), support vector machine, logistic regression, and extreme learning machine, were used to build predictive models with clinical features, radiomic features or a combination of both. An RF model incorporating both clinical and radiomic features achieved the best predictive performance, with an accuracy of 93.3%, a specificity of 95.7%, and a sensitivity of 85.7%. Based on risk scores of local failure predicted by this model, the 2-year local control rate and PFS rate were 100.0% (95% CI 100.0–100.0%) and 52.2% (31.8–72.6%) in the low-risk group and 14.3% (0.0–40.2%) and 0.0% (0.0–40.2%) in the high-risk group, respectively. This model may have the potential to stratify patients with different risks of local failure after CRT for esophageal cancer, which may facilitate the delivery of personalized treatment.

A Novel MRI-Based Radiomics Model for Predicting Recurrence in Chordoma

Chordoma is a rare primary malignant tumor. For evaluating the related factors of postoperative recurrence probability of chordoma before surgery, we retrospective collected 80 patients to analyze by using a novel radiomics method. A total of 620 3D imaging features used for radiomics analysis were extracted, and 5 features were selected from T2-weighted (T2-w) magnetic resonance imaging (MRI) that were most strongly associated with 4-year recurrence probability to build a radiomics signature. Verification by logistic regression classification model, the area under the receiver operating characteristic curve and accuracy was 0.8600 (95% CI: 0.7226-0.9824) and 85.00% in the training cohort, respectively, while in the validation cohort was 0.8568 (95% CI: 0.7327-0.9758) and 85.00%. Experimental results show that T2-w MRI-based radiomics signature is closely associated with the recurrence of chordoma. It is possible to prejudge the recurrence of chordoma before surgery.

Predicting Lymph Node Metastasis in Head and Neck Cancer by Combining Many-objective Radiomics and 3-dimensioal Convolutional Neural Network through Evidential Reasoning

Lymph node metastasis (LNM) is a significant prognostic factor in patients with head and neck cancer, and the ability to predict it accurately is essential for treatment optimization. PET and CT imaging are routinely used for LNM identification. However, uncertainties of LNM always exist especially for small size or reactive nodes. Radiomics and deep learning are the two preferred imaging-based strategies for node malignancy prediction. Radiomics models are built based on handcrafted features, and deep learning can learn the features automatically. We proposed a hybrid predictive model that combines many-objective radiomics (MO-radiomics) and 3-dimensional convolutional neural network (3D-CNN) through evidential reasoning (ER) approach. To build a more reliable model, we proposed a new many-objective radiomics model. Meanwhile, we designed a 3D-CNN that fully utilizes spatial contextual information. Finally, the outputs were fused through the ER approach. To study the predictability of the two modalities, three models were built for PET, CT, and PET& CT. The results showed that the model performed best when the two modalities were combined. Moreover, we showed that the quantitative results obtained from the hybrid model were better than those obtained from MO-radiomics and 3D-CNN.

Radiomics nomogram outperforms size criteria in discriminating lymph node metastasis in resectable esophageal squamous cell carcinoma

OBJECTIVES

To determine the value of radiomics in predicting lymph node (LN) metastasis in resectable esophageal squamous cell carcinoma (ESCC) patients.

METHODS

Data of 230 consecutive patients were retrospectively analyzed (154 in the training set and 76 in the test set). A total of 1576 radiomics features were extracted from arterial-phase CT images of the whole primary tumor. LASSO logistic regression was performed to choose the key features and construct a radiomics signature. A radiomics nomogram incorporating this signature was developed on the basis of multivariable analysis in the training set. Nomogram performance was determined and validated with respect to its discrimination, calibration and reclassification. Clinical usefulness was estimated by decision curve analysis.

RESULTS

The radiomics signature including five features was significantly associated with LN metastasis. The radiomics nomogram, which incorporated the signature and CT-reported LN status (i.e. size criteria), distinguished LN metastasis with an area under curve (AUC) of 0.758 in the training set, and performance was similar in the test set (AUC 0.773). Discrimination of the radiomics nomogram exceeded that of size criteria alone in both the training set (p <0.001) and the test set (p=0.005). Integrated discrimination improvement (IDI) and categorical net reclassification improvement (NRI) showed significant improvement in prognostic value when the radiomics signature was added to size criteria in the test set (IDI 17.3%; p<0.001; categorical NRI 52.3%; p<0.001). Decision curve analysis supported that the radiomics nomogram is superior to size criteria.

CONCLUSIONS

The radiomics nomogram provides individualized risk estimation of LN metastasis in ESCC patients and outperforms size criteria.

KEY POINTS

• A radiomics nomogram was built and validated to predict LN metastasis in resectable ESCC. • The radiomics nomogram outperformed size criteria. • Radiomics helps to unravel intratumor heterogeneity and can serve as a novel biomarker for determination of LN status in resectable ESCC.