Promises and challenges for the implementation of computational medical imaging (radiomics) in oncology

Computed tomography-based deep-learning prediction of lymph node metastasis risk in locally advanced gastric cancer

Purpose

Preoperative evaluation of lymph node metastasis (LNM) is the basis of personalized treatment of locally advanced gastric cancer (LAGC). We aim to develop and evaluate CT-based model using deep learning features to preoperatively predict LNM in LAGC.

Methods

A combined size of 523 patients who had pathologically confirmed LAGC were retrospectively collected between August 2012 and July 2019 from our hospital. Five pre-trained convolutional neural networks were exploited to extract deep learning features from pretreatment CT images. And the support vector machine (SVM) was employed as the classifier. We assessed the performance using the area under the receiver operating characteristics curve (AUC) and selected an optimal model, which was compared with a radiomics model developed from the training cohort. A clinical model was built with clinical factors only for baseline comparison.

Results

The optimal model with features extracted from ResNet yielded better performance with AUC of 0.796 [95% confidence interval (95% CI), 0.715-0.865] and accuracy of 75.2% (95% CI, 67.2%-81.5%) in the testing cohort, compared with 0.704 (0.625-0.783) and 61.8% (54.5%-69.9%) for the radiomics model. The predictive performance of all the radiological models were significantly better than the clinical model.

Conclusion

The novel and noninvasive deep learning approach could provide efficient and accurate prediction of lymph node metastasis in LAGC, and benefit clinical decision making of therapeutic strategy.

Nomograms for prediction of breast cancer in breast imaging reporting and data system (BI-RADS) ultrasound category 4 or 5 lesions: A single-center retrospective study based on radiomics features

Purpose

To develop nomograms for predicting breast malignancy in BI-RADS ultrasound (US) category 4 or 5 lesions based on radiomics features.

Methods

Between January 2020 and January 2022, we prospectively collected and retrospectively analyzed the medical records of 496 patients pathologically proven breast lesions in our hospital. The data set was divided into model training group and validation testing group with a 75/25 split. Radiomics features were obtained using the PyRadiomics package, and the radiomics score was established by least absolute shrinkage and selection operator regression. A nomogram was developed for BI-RADS US category 4 or 5 lesions according to the results of multivariate regression analysis from the training group.

Result

The AUCs of radiomics score consisting of 31 US features was 0.886. The AUC of the model constructed with radiomics score, patient age, lesion diameter identified by US and BI-RADS category involved was 0.956 (95% CI, 0.910–0.972) for the training group and 0.937 (95% CI, 0.893–0.965) for the validation cohort. The calibration curves showed good agreement between the predictions and observations.

Conclusions

Both nomogram and radiomics score can be used as methods to assist radiologists and clinicians in predicting breast malignancy in BI-RADS US category 4 or 5 lesions.

Radiomics-based infarct features on CT predict hemorrhagic transformation in patients with acute ischemic stroke

Objective

To develop and validate a model based on the radiomics features of the infarct areas on non-contrast-enhanced CT to predict hemorrhagic transformation (HT) in acute ischemic stroke.

Materials and methods

A total of 118 patients diagnosed with acute ischemic stroke in two centers from January 2019 to February 2022 were included. The radiomics features of infarcted areas on non-contrast-enhanced CT were extracted using 3D-Slicer. A univariate analysis and the least absolute shrinkage and selection operator (LASSO) were used to select features, and the radiomics score (Rad-score) was then constructed. The predictive model of HT was constructed by analyzing the Rad-score and clinical and imaging features in the training cohort, and it was verified in the validation cohort. The model was evaluated with the receiver operating characteristic curve, calibration curve and decision curve, and the prediction performance of the model in different scenarios was further discussed hierarchically.

Results

Of the 118 patients, 52 developed HT, including 21 cases of hemorrhagic infarct (HI) and 31 cases of parenchymal hematoma (PH). The Rad-score was constructed from five radiomics features and was the only independent predictor for HT. The predictive model was constructed from the Rad-score. The area under the curve (AUCs) of the model for predicting HT in the training and validation cohorts were 0.845 and 0.750, respectively. Calibration curve and decision curve analyses showed that the model performed well. Further analysis found that the model predicted HT for different infarct sizes or treatment methods in the training and validation cohorts with 78.3 and 71.4% accuracy, respectively. For all samples, the model predicted an AUC of 0.754 for HT in patients within 4.5 h since stroke onset, and predicted an AUC of 0.648 for PH.

Conclusion

This model, which was based on CT radiomics features, could help to predict HT in the setting of acute ischemic stroke for any infarct size and provide guiding suggestions for clinical treatment and prognosis evaluation.

Prediction of Histologic Neoadjuvant Chemotherapy Response in Osteosarcoma Using Pretherapeutic MRI Radiomics

Histologic response to chemotherapy for osteosarcoma is one of the most important prognostic factors for survival, but assessment occurs after surgery. Although tumor imaging is used for surgical planning and follow-up, it lacks predictive value. Therefore, a radiomics model was developed to predict the response to neoadjuvant chemotherapy based on pretreatment T1-weighted contrast-enhanced MRI. A total of 176 patients (median age, 20 years [range, 5-71 years]; 107 male patients) with osteosarcoma treated with neoadjuvant chemotherapy and surgery between January 2007 and December 2018 in three different centers in France (Centre Léon Bérard in Lyon, Centre Hospitalier Universitaire de Nantes in Nantes, and Hôpital Cochin in Paris) were retrospectively analyzed. Various models were trained from different configurations of the data sets. Two different methods of feature selection were tested with and without ComBat harmonization (ReliefF and t test) to select the most relevant features, and two different classifiers were used to build the models (an artificial neural network and a support vector machine). Sixteen radiomics models were built using the different combinations of feature selection and classifier applied on the various data sets. The most predictive model had an area under the receiver operating characteristic curve of 0.95, a sensitivity of 91%, and a specificity 92% in the training set; respective values in the validation set were 0.97, 91%, and 92%. In conclusion, MRI-based radiomics may be useful to stratify patients receiving neoadjuvant chemotherapy for osteosarcomas. Keywords: MRI, Skeletal-Axial, Oncology, Radiomics, Osteosarcoma, Pediatrics Supplemental material is available for this article. © RSNA, 2022.

Artificial intelligence and radiomics: fundamentals, applications, and challenges in immunotherapy

Immunotherapy offers the potential for durable clinical benefit but calls into question the association between tumor size and outcome that currently forms the basis for imaging-guided treatment. Artificial intelligence (AI) and radiomics allow for discovery of novel patterns in medical images that can increase radiology’s role in management of patients with cancer, although methodological issues in the literature limit its clinical application. Using keywords related to immunotherapy and radiomics, we performed a literature review of MEDLINE, CENTRAL, and Embase from database inception through February 2022. We removed all duplicates, non-English language reports, abstracts, reviews, editorials, perspectives, case reports, book chapters, and non-relevant studies. From the remaining articles, the following information was extracted: publication information, sample size, primary tumor site, imaging modality, primary and secondary study objectives, data collection strategy (retrospective vs prospective, single center vs multicenter), radiomic signature validation strategy, signature performance, and metrics for calculation of a Radiomics Quality Score (RQS). We identified 351 studies, of which 87 were unique reports relevant to our research question. The median (IQR) of cohort sizes was 101 (57–180). Primary stated goals for radiomics model development were prognostication (n=29, 33.3%), treatment response prediction (n=24, 27.6%), and characterization of tumor phenotype (n=14, 16.1%) or immune environment (n=13, 14.9%). Most studies were retrospective (n=75, 86.2%) and recruited patients from a single center (n=57, 65.5%). For studies with available information on model testing, most (n=54, 65.9%) used a validation set or better. Performance metrics were generally highest for radiomics signatures predicting treatment response or tumor phenotype, as opposed to immune environment and overall prognosis. Out of a possible maximum of 36 points, the median (IQR) of RQS was 12 (10–16). While a rapidly increasing number of promising results offer proof of concept that AI and radiomics could drive precision medicine approaches for a wide range of indications, standardizing the data collection as well as optimizing the methodological quality and rigor are necessary before these results can be translated into clinical practice.

Pancreatic cancer, radiomics and artificial intelligence

Patients with pancreatic ductal adenocarcinoma (PDAC) are generally classified into four categories based on contrast-enhanced CT at diagnosis: resectable, borderline resectable, unresectable, and metastatic disease. In the initial grading and staging of PDAC, structured radiological templates are useful but limited, as there is a need to define the aggressiveness and microscopic disease stage of these tumours to ensure adequate treatment allocation. Quantitative imaging analysis allows radiomics and dynamic imaging features to provide information of clinical outcomes, and to construct clinical models based on radiomics signatures or imaging phenotypes. These quantitative features may be used as prognostic and predictive biomarkers in clinical decision-making, enabling personalised management of advanced PDAC. Deep learning and convolutional neural networks also provide high level bioinformatics tools that can help define features associated with a given aspect of PDAC biology and aggressiveness, paving the way to define outcomes based on these features. Thus, the prediction of tumour phenotype, treatment response and patient prognosis may be feasible by using such comprehensive and integrated radiomics models. Despite these promising results, quantitative imaging is not ready for clinical implementation in PDAC. Limitations include the instability of metrics and lack of external validation. Large properly annotated datasets, including relevant semantic features (demographics, blood markers, genomics), image harmonisation, robust radiomics analysis, clinically significant tasks as outputs, comparisons with gold-standards (such as TNM or pretreatment classifications) and fully independent validation cohorts, will be required for the development of trustworthy radiomics and artificial intelligence solutions to predict PDAC aggressiveness in a clinical setting.

Noninvasive imaging of the tumor immune microenvironment correlates with response to immunotherapy in gastric cancer

The tumor immune microenvironment (TIME) is associated with tumor prognosis and immunotherapy response. Here we develop and validate a CT-based radiomics score (RS) using 2272 gastric cancer (GC) patients to investigate the relationship between the radiomics imaging biomarker and the neutrophil-to-lymphocyte ratio (NLR) in the TIME, including its correlation with prognosis and immunotherapy response in advanced GC. The RS achieves an AUC of 0.795-0.861 in predicting the NLR in the TIME. Notably, the radiomics imaging biomarker is indistinguishable from the IHC-derived NLR status in predicting DFS and OS in each cohort (HR range: 1.694-3.394, P < 0.001). We find the objective responses of a cohort of anti-PD-1 immunotherapy patients is significantly higher in the low-RS group (60.9% and 42.9%) than in the high-RS group (8.1% and 14.3%). The radiomics imaging biomarker is a noninvasive method to evaluate TIME, and may correlate with prognosis and anti PD-1 immunotherapy response in GC patients.

[18F]FDG-PET/CT radiomics for the identification of genetic clusters in pheochromocytomas and paragangliomas

Objectives

Based on germline and somatic mutation profiles, pheochromocytomas and paragangliomas (PPGLs) can be classified into different clusters. We investigated the use of [¹⁸F]FDG-PET/CT radiomics, SUV_max and biochemical profile for the identification of the genetic clusters of PPGLs.

Methods

In this single-centre cohort, 40 PPGLs (13 cluster 1, 18 cluster 2, 9 sporadic) were delineated using a 41% adaptive threshold of SUV_peak ([¹⁸F]FDG-PET) and manually (low-dose CT; ldCT). Using PyRadiomics, 211 radiomic features were extracted. Stratified 5-fold cross-validation for the identification of the genetic cluster was performed using multinomial logistic regression with dimensionality reduction incorporated per fold. Classification performances of biochemistry, SUV_max and PET(/CT) radiomic models were compared and presented as mean (multiclass) test AUCs over the five folds. Results were validated using a sham experiment, randomly shuffling the outcome labels.

Results

The model with biochemistry only could identify the genetic cluster (multiclass AUC 0.60). The three-factor PET model had the best classification performance (multiclass AUC 0.88). A simplified model with only SUV_max performed almost similarly. Addition of ldCT features and biochemistry decreased the classification performances. All sham AUCs were approximately 0.50.

Conclusion

PET radiomics achieves a better identification of PPGLs compared to biochemistry, SUV_max, ldCT radiomics and combined approaches, especially for the differentiation of sporadic PPGLs. Nevertheless, a model with SUV_max alone might be preferred clinically, weighing model performances against laborious radiomic analysis. The limited added value of radiomics to the overall classification performance for PPGL should be validated in a larger external cohort.

Key Points

• Radiomics derived from [¹⁸F]FDG-PET/CT has the potential to improve the identification of the genetic clusters of pheochromocytomas and paragangliomas.

• A simplified model with SUV_maxonly might be preferred clinically, weighing model performances against the laborious radiomic analysis.

• Cluster 1 and 2 PPGLs generally present distinctive characteristics that can be captured using [¹⁸F]FDG-PET imaging. Sporadic PPGLs appear more heterogeneous, frequently resembling cluster 2 PPGLs and occasionally resembling cluster 1 PPGLs.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00330-022-09034-5.

Preoperative Prediction of Axillary Lymph Node Metastasis in Breast Cancer Based on Intratumoral and Peritumoral DCE-MRI Radiomics Nomogram

Objective

To investigate the value of preoperative prediction of breast cancer axillary lymph node metastasis based on intratumoral and peritumoral dynamic contrast enhancement magnetic resonance imaging (DCE-MRI) radiomics nomogram. Material and Methods. In this study, a radiomics model was developed based on a training cohort involving 250 patients with breast cancer (BC) who had undergone axillary lymph node (ALN) dissection between June 2019 and January 2021. The intratumoral and peritumoral radiomics features were extracted from the second postcontrast images of DCE-MRI. Based on filtered radiomics features, the radiomics signature was built by using the least absolute shrinkage and selection operator method. The Support Vector Machines (SVM) learning algorithm was used to construct intratumoral, periatumoral, and intratumoral combined periatumoral models for predicting axillary lymph node metastasis (ALNM) in BC. Nomogram performance was determined by its discrimination, calibration, and clinical value. Multivariable logistic regression was adopted to establish a radiomics nomogram.

Results

The intratumoral combined peritumoral radiomics signature, which was composed of fifteen ALN status-related features, showed the best predictive performance and was associated with ALNM in both the training and validation cohorts (P < 0.001). The prediction efficiency of the intratumoral combined peritumoral radiomics model was higher than that of the intratumoral radiomics model and the peritumoral radiomics model. The AUCs of the training and verification cohorts were 0.867 and 0.785, respectively. The radiomics nomogram, which incorporated the radiomics signature, MR-reported ALN status, and MR-reported maximum diameter of the lesion, showed good calibration and discrimination in the training (AUC = 0.872) and validation cohorts (AUC = 0.863).

Conclusion

The intratumoral combined peritumoral radiomics model derived from DCE-MRI showed great predictive value for ALNM and may help to improve clinical decision-making for BC.

Development of an ultrasound-based radiomics nomogram to preoperatively predict Ki-67 expression level in patients with breast cancer

Objective

To develop and validate a radiomics nomogram that could incorporate clinicopathological characteristics and ultrasound (US)-based radiomics signature to non-invasively predict Ki-67 expression level in patients with breast cancer (BC) preoperatively.

Methods

A total of 328 breast lesions from 324 patients with BC who were pathologically confirmed in our hospital from June 2019 to October 2020 were included, and they were divided into high Ki-67 expression level group and low Ki-67 expression level group. Routine US and shear wave elastography (SWE) were performed for each lesion, and the ipsilateral axillary lymph nodes (ALNs) were scanned for abnormal changes. The datasets were randomly divided into training and validation cohorts with a ratio of 7:3. Correlation analysis and the least absolute shrinkage and selection operator (LASSO) were used to select the radiomics features obtained from gray-scale US images of BC patients, and each radiomics score (Rad-score) was calculated. Afterwards, multivariate logistic regression analysis was used to establish a radiomics nomogram based on the radiomics signature and clinicopathological characteristics. The prediction performance of the nomogram was assessed by the area under the receiver operating characteristic curve (AUC), the calibration curve, and decision curve analysis (DCA) using the results of immunohistochemistry as the gold standard.

Results

The radiomics signature, consisted of eight selected radiomics features, achieved a nearly moderate prediction efficacy with AUC of 0.821 (95% CI:0.764-0.880) and 0.713 (95% CI:0.612-0.814) in the training and validation cohorts, respectively. The radiomics nomogram, incorporating maximum diameter of lesions, stiff rim sign, US-reported ALN status, and radiomics signature showed a promising performance for prediction of Ki-67 expression level, with AUC of 0.904 (95% CI:0.860-0.948) and 0.890 (95% CI:0.817-0.964) in the training and validation cohorts, respectively. The calibration curve and DCA indicated promising consistency and clinical applicability.

Conclusion

The proposed US-based radiomics nomogram could be used to non-invasively predict Ki-67 expression level in BC patients preoperatively, and to assist clinicians in making reliable clinical decisions.

An automated approach for predicting glioma grade and survival of LGG patients using CNN and radiomics

Objectives

To develop and validate an efficient and automatically computational approach for stratifying glioma grades and predicting survival of lower-grade glioma (LGG) patients using an integration of state-of-the-art convolutional neural network (CNN) and radiomics.

Method

This retrospective study reviewed 470 preoperative MR images of glioma from BraTs public dataset (n=269) and Jinling hospital (n=201). A fully automated pipeline incorporating tumor segmentation and grading was developed, which can avoid variability and subjectivity of manual segmentations. First, an integrated approach by fusing CNN features and radiomics features was employed to stratify glioma grades. Then, a deep-radiomics signature based on the integrated approach for predicting survival of LGG patients was developed and subsequently validated in an independent cohort.

Results

The performance of tumor segmentation achieved a Dice coefficient of 0.81. The intraclass correlation coefficients (ICCs) of the radiomics features between the segmentation network and physicians were all over 0.75. The performance of glioma grading based on integrated approach achieved the area under the curve (AUC) of 0.958, showing the effectiveness of the integrated approach. The multivariable Cox regression results demonstrated that the deep-radiomics signature remained an independent prognostic factor and the integrated nomogram showed significantly better performance than the clinical nomogram in predicting overall survival of LGG patients (C-index: 0.865 vs. 0.796, P=0.005).

Conclusion

The proposed integrated approach can be noninvasively and efficiently applied in prediction of gliomas grade and survival. Moreover, our fully automated pipeline successfully achieved computerized segmentation instead of manual segmentation, which shows the potential to be a reproducible approach in clinical practice.

Ranking the most influential predictors of CT-based radiomics feature values in metastatic lung adenocarcinoma

PURPOSE

To investigate which acquisition, post-processing, tumor, and patient characteristics contribute the most to the value of radiomics features (RFs) in lung adenocarcinoma in order to better understand and order the potential sources of bias in radiomics studies in a multivariate setting.

METHODS

This single-center retrospective study included all consecutive patients with newly-diagnosed lung adenocarcinoma treated between December 2016 and September 2018 who had pre-treatment contrast-enhanced CT-scan showing ≥ 2 target lesions per response evaluation criteria in solid tumors (RECIST) v1.1. All measurable lesions were manually segmented; 49 RFs were extracted using LIFEx v7.0.0. Afterwards, we reverted the usual radiomics approach (i.e., predicting a clinical outcome base on multiple RFs). To do so, for each RF, random forests and linear regression algorithms were trained using cross-validation to predict the RF value depending on the following variables: patient, mutational status, phase of CT-scan acquisition, discretization (binsize), lesion location, lesion volume, and best response obtained during the first line of treatment (partial response per RECIST vs other). The most important contributors to the value of reproducible RFs (intra-class correlation coefficient > 0.80) according to the best random forests model (selected via R-squared) were ranked.

RESULTS

101 patients (median age: 62.3) were included, with a median of 5 target lesions per patient (range: 2-10) providing 466 segmented lesions. Twenty-nine RFs were reproducible. The most important predictors of the reproducible RFs values were, in order: tumor volume, binsize, tumor location, CT-scan phase, KRAS mutation, and treatment response (average importance: 61.7%, 57.4%, 8.1%, 3.3%, 3%, and 2.7%, respectively). The treatment response and KRAS and EGFR/ROS1/ALK mutational status remained independently correlated with the RF value for 64.3%, 32.1%, and 50% reproducible RFs, respectively.

CONCLUSION

Tumor volume, location, acquisition and post-processing parameters should systematically be incorporated in radiomics-based modeling; however, most reproducible RFs do have significant relationships with mutational status and treatment response.

Review of Radiomics- and Dosiomics-based Predicting Models for Rectal Cancer

By breaking the traditional medical image analysis framework, precision medicine-radiomics has attracted much attention in the past decade. The use of various mathematical algorithms offers radiomics the ability to extract vast amounts of detailed features from medical images for quantitative analysis and analyzes the confidential information related to the tumor in the image, which can establish valuable disease diagnosis and prognosis models to support personalized clinical decisions. This article summarizes the application of radiomics and dosiomics in radiation oncology. We focus on the application of radiomics in locally advanced rectal cancer and also summarize the latest research progress of dosiomics in radiation tumors to provide ideas for the treatment of future related diseases, especially 125I CT-guided radioactive seed implant brachytherapy.

Assessing the robustness of radiomics/deep learning approach in the identification of efficacy of anti–PD-1 treatment in advanced or metastatic non-small cell lung carcinoma patients

Administration of anti–PD-1 is now a standard therapy in advanced non-small cell lung carcinoma (NSCLC) patients. The clinical application of biomarkers reflecting tumor immune microenvironment is hurdled by the invasiveness of obtaining tissues despite its importance in immunotherapy. This study aimed to develop a robust and non-invasive radiomics/deep learning machine biomarker for predicting the response to immunotherapy in NSCLC patients. Radiomics/deep learning features were exacted from computed tomography (CT) images of NSCLC patients treated with Nivolumab or Pembrolizumab. The robustness of radiomics/deep learning features was assessed against various perturbations, then robust features were selected based on the Intraclass Correlation Coefficient (ICC). Radiomics/deep learning machine-learning classifiers were constructed by combining seven feature exactors, 13 feature selection methods, and 12 classifiers. The optimal model was selected using the mean area under the curve (AUC) and relative standard deviation (RSD). The consistency of image features against various perturbations was high (the range of median ICC: 0.78–0.97), but the consistency was poor in test–retest testing (the range of median ICC: 0.42–0.67). The optimal model, InceptionV3_RELF_Nearest Neighbors classifiers, had the highest prediction efficacy (AUC: 0.96 and RSD: 0.50) for anti–PD-1/PD-L1 treatment. Accuracy (ACC), sensitivity, specificity, precision, and F1 score were 95.24%, 95.00%, 95.50%, 91.67%, and 95.30%, respectively. For successful model robustification, tailoring perturbations for robustness testing to the target dataset is key. Robust radiomics/deep learning features, when paired with machine-learning methodologies, will work on the exactness and the repeatability of anticipating immunotherapy adequacy.

A combined predictive model based on radiomics features and clinical factors for disease progression in early-stage non-small cell lung cancer treated with stereotactic ablative radiotherapy

Purpose

To accurately assess disease progression after Stereotactic Ablative Radiotherapy (SABR) of early-stage Non-Small Cell Lung Cancer (NSCLC), a combined predictive model based on pre-treatment CT radiomics features and clinical factors was established.

Methods

This study retrospectively analyzed the data of 96 patients with early-stage NSCLC treated with SABR. Clinical factors included general information (e.g. gender, age, KPS, Charlson score, lung function, smoking status), pre-treatment lesion status (e.g. diameter, location, pathological type, T stage), radiation parameters (biological effective dose, BED), the type of peritumoral radiation-induced lung injury (RILI). Independent risk factors were screened by logistic regression analysis. Radiomics features were extracted from pre-treatment CT. The minimum Redundancy Maximum Relevance (mRMR) and the Least Absolute Shrinkage and Selection Operator (LASSO) were adopted for the dimensionality reduction and feature selection. According to the weight coefficient of the features, the Radscore was calculated, and the radiomics model was constructed. Multiple logistic regression analysis was applied to establish the combined model based on radiomics features and clinical factors. Receiver Operating Characteristic (ROC) curve, DeLong test, Hosmer-Lemeshow test, and Decision Curve Analysis (DCA) were used to evaluate the model’s diagnostic efficiency and clinical practicability.

Results

With the median follow-up of 59.1 months, 29 patients developed progression and 67 remained good controlled within two years. Among the clinical factors, the type of peritumoral RILI was the only independent risk factor for progression (P< 0.05). Eleven features were selected from 1781 features to construct a radiomics model. For predicting disease progression after SABR, the Area Under the Curve (AUC) of training and validation cohorts in the radiomics model was 0.88 (95%CI 0.80-0.96) and 0.80 (95%CI 0.62-0.98), and AUC of training and validation cohorts in the combined model were 0.88 (95%CI 0.81-0.96) and 0.81 (95%CI 0.62-0.99). Both the radiomics and the combined models have good prediction efficiency in the training and validation cohorts. Still, DeLong test shows that there is no difference between them.

Conclusions

Compared with the clinical model, the radiomics model and the combined model can better predict the disease progression of early-stage NSCLC after SABR, which might contribute to individualized follow-up plans and treatment strategies.

CT-based radiomics to predict development of macrovascular invasion in hepatocellular carcinoma: A multicenter study

BACKGROUND

Macrovascular invasion (MaVI) occurs in nearly half of hepatocellular carcinoma (HCC) patients at diagnosis or during follow-up, which causes severe disease deterioration, and limits the possibility of surgical approaches. This study aimed to investigate whether computed tomography (CT)-based radiomics analysis could help predict development of MaVI in HCC.

METHODS

A cohort of 226 patients diagnosed with HCC was enrolled from 5 hospitals with complete MaVI and prognosis follow-ups. CT-based radiomics signature was built via multi-strategy machine learning methods. Afterwards, MaVI-related clinical factors and radiomics signature were integrated to construct the final prediction model (CRIM, clinical-radiomics integrated model) via random forest modeling. Cox-regression analysis was used to select independent risk factors to predict the time of MaVI development. Kaplan-Meier analysis was conducted to stratify patients according to the time of MaVI development, progression-free survival (PFS), and overall survival (OS) based on the selected risk factors.

RESULTS

The radiomics signature showed significant improvement for MaVI prediction compared with conventional clinical/radiological predictors (P < 0.001). CRIM could predict MaVI with satisfactory areas under the curve (AUC) of 0.986 and 0.979 in the training (n = 154) and external validation (n = 72) datasets, respectively. CRIM presented with excellent generalization with AUC of 0.956, 1.000, and 1.000 in each external cohort that accepted disparate CT scanning protocol/manufactory. Peel9_fos_InterquartileRange [hazard ratio (HR) = 1.98; P < 0.001] was selected as the independent risk factor. The cox-regression model successfully stratified patients into the high-risk and low-risk groups regarding the time of MaVI development (P < 0.001), PFS (P < 0.001) and OS (P = 0.002).

CONCLUSIONS

The CT-based quantitative radiomics analysis could enable high accuracy prediction of subsequent MaVI development in HCC with prognostic implications.

Applications of Artificial Intelligence Based on Medical Imaging in Glioma: Current State and Future Challenges

Glioma is one of the most fatal primary brain tumors, and it is well-known for its difficulty in diagnosis and management. Medical imaging techniques such as magnetic resonance imaging (MRI), positron emission tomography (PET), and spectral imaging can efficiently aid physicians in diagnosing, treating, and evaluating patients with gliomas. With the increasing clinical records and digital images, the application of artificial intelligence (AI) based on medical imaging has reduced the burden on physicians treating gliomas even further. This review will classify AI technologies and procedures used in medical imaging analysis. Additionally, we will discuss the applications of AI in glioma, including tumor segmentation and classification, prediction of genetic markers, and prediction of treatment response and prognosis, using MRI, PET, and spectral imaging. Despite the benefits of AI in clinical applications, several issues such as data management, incomprehension, safety, clinical efficacy evaluation, and ethical or legal considerations, remain to be solved. In the future, doctors and researchers should collaborate to solve these issues, with a particular emphasis on interdisciplinary teamwork.

AutoComBat: a generic method for harmonizing MRI-based radiomic features

The use of multicentric data is becoming essential for developing generalizable radiomic signatures. In particular, Magnetic Resonance Imaging (MRI) data used in brain oncology are often heterogeneous in terms of scanners and acquisitions, which significantly impact quantitative radiomic features. Various methods have been proposed to decrease dependency, including methods acting directly on MR images, i.e., based on the application of several preprocessing steps before feature extraction or the ComBat method, which harmonizes radiomic features themselves. The ComBat method used for radiomics may be misleading and presents some limitations, such as the need to know the labels associated with the "batch effect". In addition, a statistically representative sample is required and the applicability of a signature whose batch label is not present in the train set is not possible. This work aimed to compare a priori and a posteriori radiomic harmonization methods and propose a code adaptation to be machine learning compatible. Furthermore, we have developed AutoComBat, which aims to automatically determine the batch labels, using either MRI metadata or quality metrics as inputs of the proposed constrained clustering. A heterogeneous dataset consisting of high and low-grade gliomas coming from eight different centers was considered. The different methods were compared based on their ability to decrease relative standard deviation of radiomic features extracted from white matter and on their performance on a classification task using different machine learning models. ComBat and AutoComBat using image-derived quality metrics as inputs for batch assignment and preprocessing methods presented promising results on white matter harmonization, but with no clear consensus for all MR images. Preprocessing showed the best results on the T1w-gd images for the grading task. For T2w-flair, AutoComBat, using either metadata plus quality metrics or metadata alone as inputs, performs better than the conventional ComBat, highlighting its potential for data harmonization. Our results are MRI weighting, feature class and task dependent and require further investigations on other datasets.

GEP-NET radiomics: a systematic review and radiomics quality score assessment

OBJECTIVE

The number of radiomics studies in gastroenteropancreatic neuroendocrine tumours (GEP-NETs) is rapidly increasing. This systematic review aims to provide an overview of the available evidence of radiomics for clinical outcome measures in GEP-NETs, to understand which applications hold the most promise and which areas lack evidence.

METHODS

PubMed, Embase, and Wiley/Cochrane Library databases were searched and a forward and backward reference check of the identified studies was executed. Inclusion criteria were (1) patients with GEP-NETs and (2) radiomics analysis on CT, MRI or PET. Two reviewers independently agreed on eligibility and assessed methodological quality with the radiomics quality score (RQS) and extracted outcome data.

RESULTS

In total, 1364 unique studies were identified and 45 were included for analysis. Most studies focused on GEP-NET grade and differential diagnosis of GEP-NETs from other neoplasms, while only a minority analysed treatment response or long-term outcomes. Several studies were able to predict tumour grade or to differentiate GEP-NETs from other lesions with a good performance (AUCs 0.74-0.96 and AUCs 0.80-0.99, respectively). Only one study developed a model to predict recurrence in pancreas NETs (AUC 0.77). The included studies reached a mean RQS of 18%.

CONCLUSION

Although radiomics for GEP-NETs is still a relatively new area, some promising models have been developed. Future research should focus on developing robust models for clinically relevant aims such as prediction of response or long-term outcome in GEP-NET, since evidence for these aims is still scarce.

KEY POINTS

• The majority of radiomics studies in gastroenteropancreatic neuroendocrine tumours is of low quality. • Most evidence for radiomics is available for the identification of tumour grade or differentiation of gastroenteropancreatic neuroendocrine tumours from other neoplasms. • Radiomics for the prediction of response or long-term outcome in gastroenteropancreatic neuroendocrine tumours warrants further research.

Novel Harmonization Method for Multi-Centric Radiomic Studies in Non-Small Cell Lung Cancer

The purpose of this multi-centric work was to investigate the relationship between radiomic features extracted from pre-treatment computed tomography (CT), positron emission tomography (PET) imaging, and clinical outcomes for stereotactic body radiation therapy (SBRT) in early-stage non-small cell lung cancer (NSCLC). One-hundred and seventeen patients who received SBRT for early-stage NSCLC were retrospectively identified from seven Italian centers. The tumor was identified on pre-treatment free-breathing CT and PET images, from which we extracted 3004 quantitative radiomic features. The primary outcome was 24-month progression-free-survival (PFS) based on cancer recurrence (local/non-local) following SBRT. A harmonization technique was proposed for CT features considering lesion and contralateral healthy lung tissues using the LASSO algorithm as a feature selector. Models with harmonized CT features (B models) demonstrated better performances compared to the ones using only original CT features (C models). A linear support vector machine (SVM) with harmonized CT and PET features (A1 model) showed an area under the curve (AUC) of 0.77 (0.63–0.85) for predicting the primary outcome in an external validation cohort. The addition of clinical features did not enhance the model performance. This study provided the basis for validating our novel CT data harmonization strategy, involving delta radiomics. The harmonized radiomic models demonstrated the capability to properly predict patient prognosis.

The Usefulness of Radiomics Methodology for Developing Descriptive and Prognostic Image-Based Phenotyping in the Aging Population: Results From a Small Feasibility Study

Background: Radiomics is an emerging field that translates medical images into quantitative data to enable phenotypic profiling of human disease. In this retrospective study, we asked whether it is possible to use image-based phenotyping to describe and determine prognostic factors in the aging population.

Methods: A radiomic frailty cohort with 101 patients was included in the analysis (65 ± 15 years, 55 men). A total of 44 texture features were extracted from the segmented muscle area of the ultrasound images of the anterior thigh. Univariate and multivariate analyses were performed to assess the image data sets and clinical data.

Results: Our results showed that the heterogeneity of muscle was associated with an increased incidence of hearing impairment, stroke, myocardial infarction, dementia/memory loss, and falls in the following two years. Regression analysis revealed a muscle radiomic model with 87.1% correct predictive value with good sensitivity and moderate specificity (p = 0.001).

Conclusion: It is possible to develop and identify image-based phenotypes in the elderly population. The muscle radiomic model needs to further be validated. Future studies correlated with biological data (genomics, transcriptomics, metabolomics, etc.) will give further insights into the biological basis and molecular processes of the developed radiomic model.

Auto-Segmentation Ultrasound-Based Radiomics Technology to Stratify Patient With Diabetic Kidney Disease: A Multi-Center Retrospective Study

Background

An increasing proportion of patients with diabetic kidney disease (DKD) has been observed among incident hemodialysis patients in large cities, which is consistent with the continuous growth of diabetes in the past 20 years.

Purpose

In this multicenter retrospective study, we developed a deep learning (DL)-based automatic segmentation and radiomics technology to stratify patients with DKD and evaluate the possibility of clinical application across centers.

Materials and Methods

The research participants were enrolled retrospectively and separated into three parts: training, validation, and independent test datasets for further analysis. DeepLabV3+ network, PyRadiomics package, and least absolute shrinkage and selection operator were used for segmentation, extraction of radiomics variables, and regression, respectively.

Results

A total of 499 patients from three centers were enrolled in this study including 246 patients with type II diabetes mellitus (T2DM) and 253 patients with DKD. The mean intersection-over-union (Miou) and mean pixel accuracy (mPA) of automatic segmentation of the data from the three medical centers were 0.812 ± 0.003, 0.781 ± 0.009, 0.805 ± 0.020 and 0.890 ± 0.004, 0.870 ± 0.002, 0.893 ± 0.007, respectively. The variables from the renal parenchyma and sinus provided different information for the diagnosis and follow-up of DKD. The area under the curve (AUC) of the radiomics model for differentiating between DKD and T2DM patients was 0.674 ± 0.074 and for differentiating between the high and low stages of DKD was 0.803 ± 0.037.

Conclusion

In this study, we developed a DL-based automatic segmentation, radiomics technology to stratify patients with DKD. The DL technology was proposed to achieve fast and accurate anatomical-level segmentation in the kidney, and an ultrasound-based radiomics model can achieve high diagnostic performance in the diagnosis and follow-up of patients with DKD.

Imaging approaches and radiomics: toward a new era of ultraprecision radioimmunotherapy?

Strong rationale and a growing number of preclinical and clinical studies support combining radiotherapy and immunotherapy to improve patient outcomes. However, several critical questions remain, such as the identification of patients who will benefit from immunotherapy and the identification of the best modalities of treatment to optimize patient response. Imaging biomarkers and radiomics have recently emerged as promising tools for the non-invasive assessment of the whole disease of the patient, allowing comprehensive analysis of the tumor microenvironment, the spatial heterogeneity of the disease and its temporal changes. This review presents the potential applications of medical imaging and the challenges to address, in order to help clinicians choose the optimal modalities of both radiotherapy and immunotherapy, to predict patient's outcomes and to assess response to these promising combinations.

Virtual Biopsy in Soft Tissue Sarcoma. How Close Are We?

A shift in radiology to a data-driven specialty has been unlocked by synergistic developments in imaging biomarkers (IB) and computational science. This is advancing the capability to deliver "virtual biopsies" within oncology. The ability to non-invasively probe tumour biology both spatially and temporally would fulfil the potential of imaging to inform management of complex tumours; improving diagnostic accuracy, providing new insights into inter- and intra-tumoral heterogeneity and individualised treatment planning and monitoring. Soft tissue sarcomas (STS) are rare tumours of mesenchymal origin with over 150 histological subtypes and notorious heterogeneity. The combination of inter- and intra-tumoural heterogeneity and the rarity of the disease remain major barriers to effective treatments. We provide an overview of the process of successful IB development, the key imaging and computational advancements in STS including quantitative magnetic resonance imaging, radiomics and artificial intelligence, and the studies to date that have explored the potential biological surrogates to imaging metrics. We discuss the promising future directions of IBs in STS and illustrate how the routine clinical implementation of a virtual biopsy has the potential to revolutionise the management of this group of complex cancers and improve clinical outcomes.

Machine Learning Based on MRI DWI Radiomics Features for Prognostic Prediction in Nasopharyngeal Carcinoma

Simple Summary

In the past, radiomics studies of nasopharyngeal carcinoma (NPC) were only based on basic MR sequences. Previous studies have shown that radiomics methods based on T2-weighted imaging and contrast-enhanced T1-weighted imaging have been successfully used to improve the prognosis of patients with nasopharyngeal carcinoma. The purpose of this study was to explore the predictive efficacy of radiomics analyses based on readout-segmented echo-planar diffusion-weighted imaging (RESOLVE-DWI) which quantitatively reflects the diffusion motion of water molecules for prognosis evaluation in nasopharyngeal carcinoma. Several prognostic radiomics models were established by using diffusion-weighted imaging, apparent diffusion coefficient maps, T2-weighted and contrast-enhanced T1-weighted imaging to predict the risk of recurrence or metastasis of nasopharyngeal carcinoma, and the predictive effects of different models were compared. The results show that the model based on MRI DWI can successfully predict the prognosis of patients with nasopharyngeal carcinoma and has higher predictive efficiency than the model based on the conventional sequence, which suggests MRI DWI-radiomics can provide a useful and alternative approach for survival estimation.

Abstract

Purpose: This study aimed to explore the predictive efficacy of radiomics analyses based on readout-segmented echo-planar diffusion-weighted imaging (RESOLVE-DWI) for prognosis evaluation in nasopharyngeal carcinoma in order to provide further information for clinical decision making and intervention. Methods: A total of 154 patients with untreated NPC confirmed by pathological examination were enrolled, and the pretreatment magnetic resonance image (MRI)—including diffusion-weighted imaging (DWI), apparent diffusion coefficient (ADC) maps, T2-weighted imaging (T2WI), and contrast-enhanced T1-weighted imaging (CE-T1WI)—was collected. The Random Forest (RF) algorithm selected radiomics features and established the machine-learning models. Five models, namely model 1 (DWI + ADC), model 2 (T2WI + CE-T1WI), model 3 (DWI + ADC + T2WI), model 4 (DWI + ADC + CE-T1WI), and model 5 (DWI + ADC + T2WI + CE-T1WI), were constructed. The average area under the curve (AUC) of the validation set was determined in order to compare the predictive efficacy for prognosis evaluation. Results: After adjusting the parameters, the RF machine learning models based on extracted imaging features from different sequence combinations were obtained. The invalidation sets of model 1 (DWI + ADC) yielded the highest average AUC of 0.80 (95% CI: 0.79–0.81). The average AUCs of the model 2, 3, 4, and 5 invalidation sets were 0.72 (95% CI: 0.71–0.74), 0.66 (95% CI: 0.64–0.68), 0.74 (95% CI: 0.73–0.75), and 0.75 (95% CI: 0.74–0.76), respectively. Conclusion: A radiomics model derived from the MRI DWI of patients with nasopharyngeal carcinoma was generated in order to evaluate the risk of recurrence and metastasis. The model based on MRI DWI can provide an alternative approach for survival estimation, and can reveal more information for clinical decision-making and intervention.

A Multiparametric Method Based on Clinical and CT-Based Radiomics to Predict the Expression of p53 and VEGF in Patients With Spinal Giant Cell Tumor of Bone

Purpose

This project aimed to assess the significance of vascular endothelial growth factor (VEGF) and p53 for predicting progression-free survival (PFS) in patients with spinal giant cell tumor of bone (GCTB) and to construct models for predicting these two biomarkers based on clinical and computer tomography (CT) radiomics to identify high-risk patients for improving treatment.

Material and Methods

A retrospective study was performed from April 2009 to January 2019. A total of 80 patients with spinal GCTB who underwent surgery in our institution were identified. VEGF and p53 expression and clinical and general imaging information were collected. Multivariate Cox regression models were used to verify the prognostic factors. The radiomics features were extracted from the regions of interest (ROIs) in preoperative CT, and then important features were selected by the SVM to build classification models, evaluated by 10-fold crossvalidation. The clinical variables were processed using the same method to build a conventional model for comparison.

Results

The immunohistochemistry of 80 patients was obtained: 49 with high-VEGF and 31 with low-VEGF, 68 with wild-type p53, and 12 with mutant p53. p53 and VEGF were independent prognostic factors affecting PFS found in multivariate Cox regression analysis. For VEGF, the Spinal Instability Neoplastic Score (SINS) was greater in the high than low groups, p &lt; 0.001. For p53, SINS (p = 0.030) and Enneking stage (p = 0.017) were higher in mutant than wild-type groups. The VEGF radiomics model built using 3 features achieved an area under the curve (AUC) of 0.88, and the p53 radiomics model built using 4 features had an AUC of 0.79. The conventional model built using SINS, and the Enneking stage had a slightly lower AUC of 0.81 for VEGF and 0.72 for p53.

Conclusion

p53 and VEGF are associated with prognosis in patients with spinal GCTB, and the radiomics analysis based on preoperative CT provides a feasible method for the evaluation of these two biomarkers, which may aid in choosing better management strategies.

MRI-Based Radiomics and Urine Creatinine for the Differentiation of Renal Angiomyolipoma With Minimal Fat From Renal Cell Carcinoma: A Preliminary Study

Objectives

Standard magnetic resonance imaging (MRI) techniques are different to distinguish minimal fat angiomyolipoma (mf-AML) with minimal fat from renal cell carcinoma (RCC). Here we aimed to evaluate the diagnostic performance of MRI-based radiomics in the differentiation of fat-poor AMLs from other renal neoplasms.

Methods

A total of 69 patients with solid renal tumors without macroscopic fat and with a pathologic diagnosis of RCC (n=50) or mf-AML (n=19) who underwent conventional MRI and intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI) were included. Clinical data including age, sex, tumor location, urine creatinine, and urea nitrogen were collected from medical records. The apparent diffusion coefficient (ADC), pure diffusion coefficient (D), pseudodiffusion coefficient (D*), and perfusion fraction (f) were measured from renal tumors. We used the ITK-SNAP software to manually delineate the regions of interest on T2-weighted imaging (T2WI) and IVIM-DWI from the largest cross-sectional area of the tumor. We extracted 396 radiomics features by the Analysis Kit software for each MR sequence. The hand-crafted features were selected by using the Pearson correlation analysis and least absolute shrinkage and selection operator (LASSO). Diagnostic models were built by logistic regression analysis. Receiver operating characteristic curve analysis was performed using five-fold cross-validation and the mean area under the curve (AUC) values were calculated and compared between the models to obtain the optimal model for the differentiation of mf-AML and RCC. Decision curve analysis (DCA) was used to evaluate the clinical utility of the models.

Results

Clinical model based on urine creatinine achieved an AUC of 0.802 (95%CI: 0.761-0.843). IVIM-based model based on f value achieved an AUC of 0.692 (95%CI: 0.627-0.757). T2WI-radiomics model achieved an AUC of 0.883 (95%CI: 0.852-0.914). IVIM-radiomics model achieved an AUC of 0.874 (95%CI: 0.841-0.907). Combined radiomics model achieved an AUC of 0.919 (95%CI: 0.894-0.944). Clinical-radiomics model yielded the best performance, with an AUC of 0.931 (95%CI: 0.907-0.955). The calibration curve and DCA confirmed that the clinical-radiomics model had a good consistency and clinical usefulness.

Conclusion

The clinical-radiomics model may be served as a noninvasive diagnostic tool to differentiate mf-AML with RCC, which might facilitate the clinical decision-making process.

Preoperative prediction of the degree of differentiation of lung adenocarcinoma presenting as sub-solid or solid nodules with a radiomics nomogram

AIM

To develop and validate a radiomics nomogram for prediction of degree of differentiation in lung adenocarcinoma presenting as sub-solid or solid nodules.

MATERIALS AND METHODS

A total of 438 patients with histopathologically confirmed adenocarcinoma (248 non-poorly differentiated and 190 poorly differentiated) were divided into training cohort (n=235) and internal validation cohort (n=203) according to surgery sequence. Sixty patients form public TCIA dataset were selected for external validation. One thousand, two hundred and eighteen radiomics features were extracted from each volumetric region of interest and a least absolute shrinkage and selection operator logistic regression was applied to select meaningful radiomic features for building a radiomics score (Rad-score) model. A nomogram model incorporating the Rad-score and type was established after multivariable logistic regression. The discrimination efficiency, calibration efficacy, and clinical utility value of the nomogram were evaluated.

RESULTS

The Rad-score model could predict the differentiation degree of lung adenocarcinoma with an area under the curve (AUC) of 0.83 (95% confidence interval [CI]: 0.78-0.89) in the internal validation cohort. The AUC of the nomogram and radiographic model was 0.86 (95% CI: 0.80-0.91), 0.78 (95% CI: 0.72-0.84) in the internal validation cohort respectively. The AUC of the nomogram in the external validation cohort was 0.73 (95% CI: 0.58-0.88). Delong's test showed that the nomogram performed better than radiographic features alone (p=0.001).

CONCLUSIONS

The proposed radiomics nomogram has the potential to predict the differentiation degree of lung adenocarcinoma preoperatively.

[18F]FDG-PET/CT radiomics and artificial intelligence in lung cancer: Technical aspects and potential clinical applications

Lung cancer is the second most common cancer and the leading cause of cancer-related death worldwide. Molecular imaging using [¹⁸F]fluorodeoxyglucose Positron Emission Tomography and/or Computed Tomography ([¹⁸F]FDG-PET/CT) plays an essential role in the diagnosis, evaluation of response to treatment, and prediction of outcomes. The images are evaluated using qualitative and conventional quantitative indices. However, there is far more information embedded in the images, which can be extracted by sophisticated algorithms. Recently, the concept of uncovering and analyzing the invisible data extracted from medical images, called radiomics, is gaining more attention. Currently, [¹⁸F]FDG-PET/CT radiomics is growingly evaluated in lung cancer to discover if it enhances the diagnostic performance or implication of [¹⁸F]FDG-PET/CT in the management of lung cancer. In this review, we provide a short overview of the technical aspects, as they are discussed in different articles of this special issue. We mainly focus on the diagnostic performance of the [¹⁸F]FDG-PET/CT-based radiomics and the role of artificial intelligence in non-small cell lung cancer, impacting the early detection, staging, prediction of tumor subtypes, biomarkers, and patient's outcomes.

Deep learning for locally advanced nasopharyngeal carcinoma prognostication based on pre- and post-treatment MRI

PURPOSE

We aimed to predict the prognosis of advanced nasopharyngeal carcinoma (stage Ⅲ-Ⅳa) using Pre- and Post-treatment MR images based on deep learning (DL).

METHODS

A total of 206 patients with primary nasopharyngeal carcinoma who were diagnosed and treated at the Renmin Hospital of Wuhan University between June 2012 and January 2018 were retrospectively selected. A rectangular region of interest (ROI), which included the tumor area, surrounding tissues and organs, was delineated on each Pre- and Post-treatment MR image. Two Inception-Resnet-V2 based transfer learning models, named Pre-model and Post-model, were trained with the Pre-treatment images and the Post-treatment images, respectively. In addition, an ensemble learning model based on the Pre-model and Post-models was established. The three established models were evaluated by receiver operating characteristic curve (ROC), confusion matrix, and Harrell's concordance indices (C-index). High-risk-related gradient-weighted class activation mapping (Grad-CAM) images were developed according to the DL models.

RESULTS

The Pre-model, Post-model, and ensemble model displayed a C-index of 0.717 (95% CI: 0.639 to 0.795), 0.811 (95% CI: 0.745-0.877), 0.830 (95% CI: 0.767-0.893), and AUC of 0.741 (95% CI: 0.584-0.900), 0.806 (95% CI: 0.670-0.942), and 0.842 (95% CI: 0.718-0.967) for the test cohort, respectively. In comparison with the models, the performance of Post-model was better than the performance of Pre-model, which indicated the importance of Post-treatment images for prognosis prediction. All three DL models performed better than the TNM staging system (0.723, 95% CI: 0.567-0.879). The captured features presented on Grad-CAM images suggested that the areas around the tumor and lymph nodes were related to the prognosis of the tumor.

CONCLUSIONS

The three established DL models based on Pre- and Post-treatment MR images have a better performance than TNM staging. Post-treatment MR images are of great significance for prognosis prediction and could contribute to clinical decision-making.

Can radiomic analysis of a single-phase dual-energy CT improve the diagnostic accuracy of differentiating enhancing from non-enhancing small renal lesions?

BACKGROUND

The value of dual-energy computed tomography (DECT)-based radiomics in renal lesions is unknown.

PURPOSE

To develop DECT-based radiomic models and assess their incremental values in comparison to conventional measurements for differentiating enhancing from non-enhancing small renal lesions.

MATERIAL AND METHODS

A total of 349 patients with 519 small renal lesions (390 non-enhancing, 129 enhancing) who underwent contrast-enhanced nephrographic phase DECT examinations between June 2013 and January 2020 on multiple DECT platforms were retrospectively recruited. Cohort A included all lesions, while cohort B included Bosniak II-IV and solid enhancing renal lesions. Radiomic models were built with features selected by the least absolute shrinkage and selection operator regression (LASSO). ROC analyses were performed to compare the diagnostic accuracy among conventional and radiomic models for predicting enhancing renal lesions.

RESULTS

The individual iodine concentration (IC), normalized IC, mean attenuation on 75-keV images, radiomic model of iodine images, 75-keV images and a combined model integrating all the above-mentioned features all demonstrated high AUCs for predicting renal lesion enhancement in cohort A (AUCs = 0.934-0.979) as well as in the test dataset (AUCs = 0.892-0.962) of cohort B (P values with Bonferroni correction >0.003). The AUC (0.864) of mean attenuation on 75-keV images was significantly lower than those of other models (all P values ≤0.001) except the radiomic model of 75-keV images (P = 0.038) in the training dataset of cohort B.

CONCLUSION

No incremental value was found by adding radiomic and machine learning analyses to iodine images for differentiating enhancing from non-enhancing renal lesions.

Quantitative classification and radiomics of [18F]FDG-PET/CT in indeterminate thyroid nodules

PURPOSE

To evaluate whether quantitative [18F]FDG-PET/CT assessment, including radiomic analysis of [18F]FDG-positive thyroid nodules, improved the preoperative differentiation of indeterminate thyroid nodules of non-Hürthle cell and Hürthle cell cytology.

METHODS

Prospectively included patients with a Bethesda III or IV thyroid nodule underwent [18F]FDG-PET/CT imaging. Receiver operating characteristic (ROC) curve analysis was performed for standardised uptake values (SUV) and SUV-ratios, including assessment of SUV cut-offs at which a malignant/borderline neoplasm was reliably ruled out (≥ 95% sensitivity). [18F]FDG-positive scans were included in radiomic analysis. After segmentation at 50% of SUVpeak, 107 radiomic features were extracted from [18F]FDG-PET and low-dose CT images. Elastic net regression classifiers were trained in a 20-times repeated random split. Dimensionality reduction was incorporated into the splits. Predictive performance of radiomics was presented as mean area under the ROC curve (AUC) across the test sets.

RESULTS

Of 123 included patients, 84 (68%) index nodules were visually [18F]FDG-positive. The malignant/borderline rate was 27% (33/123). SUV-metrices showed AUCs ranging from 0.705 (95% CI, 0.601-0.810) to 0.729 (0.633-0.824), 0.708 (0.580-0.835) to 0.757 (0.650-0.864), and 0.533 (0.320-0.747) to 0.700 (0.502-0.898) in all (n = 123), non-Hürthle (n = 94), and Hürthle cell (n = 29) nodules, respectively. At SUVmax, SUVpeak, SUVmax-ratio, and SUVpeak-ratio cut-offs of 2.1 g/mL, 1.6 g/mL, 1.2, and 0.9, respectively, sensitivity of [18F]FDG-PET/CT was 95.8% (95% CI, 78.9-99.9%) in non-Hürthle cell nodules. In Hürthle cell nodules, cut-offs of 5.2 g/mL, 4.7 g/mL, 3.4, and 2.8, respectively, resulted in 100% sensitivity (95% CI, 66.4-100%). Radiomic analysis of 84 (68%) [18F]FDG-positive nodules showed a mean test set AUC of 0.445 (95% CI, 0.290-0.600) for the PET model.

CONCLUSION

Quantitative [18F]FDG-PET/CT assessment ruled out malignancy in indeterminate thyroid nodules. Distinctive, higher SUV cut-offs should be applied in Hürthle cell nodules to optimize rule-out ability. Radiomic analysis did not contribute to the additional differentiation of [18F]FDG-positive nodules.

TRIAL REGISTRATION NUMBER

This trial is registered with ClinicalTrials.gov: NCT02208544 (5 August 2014), https://clinicaltrials.gov/ct2/show/NCT02208544 .

CT-Based Radiomics and Deep Learning for BRCA Mutation and Progression-Free Survival Prediction in Ovarian Cancer Using a Multicentric Dataset

PURPOSE

Build predictive radiomic models for early relapse and BRCA mutation based on a multicentric database of high-grade serous ovarian cancer (HGSOC) and validate them in a test set coming from different institutions.

METHODS

Preoperative CTs of patients with HGSOC treated at four referral centers were retrospectively acquired and manually segmented. Hand-crafted features and deep radiomics features were extracted respectively by dedicated software (MODDICOM) and a dedicated convolutional neural network (CNN). Features were selected with and without prior harmonization (ComBat harmonization), and models were built using different machine learning algorithms, including clinical variables.

RESULTS

We included 218 patients. Radiomic models showed low performance in predicting both BRCA mutation (AUC in test set between 0.46 and 0.59) and 1-year relapse (AUC in test set between 0.46 and 0.56); deep learning models demonstrated similar results (AUC in the test of 0.48 for BRCA and 0.50 for relapse). The inclusion of clinical variables improved the performance of the radiomic models to predict BRCA mutation (AUC in the test set of 0.74).

CONCLUSIONS

In our multicentric dataset, representative of a real-life clinical scenario, we could not find a good radiomic predicting model for PFS and BRCA mutational status, with both traditional radiomics and deep learning, but the combination of clinical and radiomic models improved model performance for the prediction of BRCA mutation. These findings highlight the need for standardization through the whole radiomic pipelines and robust multicentric external validations of results.

Feasibility of a Clinical-Radiomics Model to Predict the Outcomes of Acute Ischemic Stroke

Objective

To develop a model incorporating radiomic features and clinical factors to accurately predict acute ischemic stroke (AIS) outcomes.

Materials and Methods

Data from 522 AIS patients (382 male [73.2%]; mean age ± standard deviation, 58.9 ± 11.5 years) were randomly divided into the training (n = 311) and validation cohorts (n = 211). According to the modified Rankin Scale (mRS) at 6 months after hospital discharge, prognosis was dichotomized into good (mRS ≤ 2) and poor (mRS > 2); 1310 radiomics features were extracted from diffusion-weighted imaging and apparent diffusion coefficient maps. The minimum redundancy maximum relevance algorithm and the least absolute shrinkage and selection operator logistic regression method were implemented to select the features and establish a radiomics model. Univariable and multivariable logistic regression analyses were performed to identify the clinical factors and construct a clinical model. Ultimately, a multivariable logistic regression analysis incorporating independent clinical factors and radiomics score was implemented to establish the final combined prediction model using a backward step-down selection procedure, and a clinical-radiomics nomogram was developed. The models were evaluated using calibration, receiver operating characteristic (ROC), and decision curve analyses.

Results

Age, sex, stroke history, diabetes, baseline mRS, baseline National Institutes of Health Stroke Scale score, and radiomics score were independent predictors of AIS outcomes. The area under the ROC curve of the clinical-radiomics model was 0.868 (95% confidence interval, 0.825–0.910) in the training cohort and 0.890 (0.844–0.936) in the validation cohort, which was significantly larger than that of the clinical or radiomics models. The clinical radiomics nomogram was well calibrated (p > 0.05). The decision curve analysis indicated its clinical usefulness.

Conclusion

The clinical-radiomics model outperformed individual clinical or radiomics models and achieved satisfactory performance in predicting AIS outcomes.

Combined model of radiomics, clinical, and imaging features for differentiating focal pneumonia-like lung cancer from pulmonary inflammatory lesions: an exploratory study

Background

Only few studies have focused on differentiating focal pneumonia-like lung cancer (F-PLC) from focal pulmonary inflammatory lesion (F-PIL). This exploratory study aimed to evaluate the clinical value of a combined model incorporating computed tomography (CT)-based radiomics signatures, clinical factors, and CT morphological features for distinguishing F-PLC and F-PIL.

Methods

In total, 396 patients pathologically diagnosed with F-PLC and F-PIL from two medical institutions between January 2015 and May 2021 were retrospectively analyzed. Patients from center 1 were included in the training (n = 242) and internal validation (n = 104) cohorts. Moreover, patients from center 2 were classified under the external validation cohort (n = 50). The clinical and CT morphological characteristics of both groups were compared first. And then, a clinical model incorporating clinical and CT morphological features, a radiomics model reflecting the radiomics signature of lung lesions, and a combined model were developed and validated, respectively.

Results

Age, gender, smoking history, respiratory symptoms, air bronchogram, necrosis, and pleural attachment differed significantly between the F-PLC and F-PIL groups (all P < 0.05). For the clinical model, age, necrosis, and pleural attachment were the most effective factors to differentiate F-PIL from F-PLC, with the area under the curves (AUCs) of 0.838, 0.819, and 0.717 in the training and internal and external validation cohorts, respectively. For the radiomics model, five radiomics features were found to be significantly related to the identification of F-PLC and F-PIL (all P < 0.001), with the AUCs of 0.804, 0.877, and 0.734 in the training and internal and external validation cohorts, respectively. For the combined model, five radiomics features, age, necrosis, and pleural attachment were independent predictors for distinguishing between F-PLC and F-PIL, with the AUCs of 0.915, 0.899, and 0.805 in the training and internal and external validation cohorts, respectively. The combined model exhibited a better performance than had the clinical and radiomics models.

Conclusions

The combined model, which incorporates CT-based radiomics signatures, clinical factors, and CT morphological characteristics, is effective in differentiating F-PLC from F-PIL.

The value of radiomics based on dual-energy CT for differentiating benign from malignant solitary pulmonary nodules

OBJECTIVE

To investigate the value of monochromatic dual-energy CT (DECT) images based on radiomics in differentiating benign from malignant solitary pulmonary nodules.

MATERIALS AND METHODS

This retrospective study was approved by the institutional review board, and informed consent was waived. Pathologically confirmed lung nodules smaller than 3 cm with integrated arterial phase and venous phase (AP and VP) gemstone spectral imaging were retrospectively identified. After extracting the radiomic features of each case, principal component analysis (PCA) was used for feature selection, and after training with the logistic regression method, three classification models (Model_AP, Model_VP and Model_Combination) were constructed. The performance was assessed by the area under the receiver operating curve (AUC), and the efficacy of the models was validated using an independent cohort.

RESULTS

A total of 153 patients were included and divided into a training cohort (n = 107) and a validation cohort (n = 46). A total of 1130 radiomic features were extracted from each case. The PCA method selected 22, 25 and 35 principal components to construct the three models. The diagnostic accuracy of Model_AP, Model_VP and Model_Combination was 0.8043, 0.6739, and 0.7826 in the validation set, with AUCs of 0.8148 (95% CI 0.682-0.948), 0.7485 (95% CI 0.602-0.895), and 0.8772 (95% CI 0.780-0.974), respectively. The DeLong test showed that there were significant differences in the AUCs between Model_AP and Model_Combination (P = 0.0396) and between Model_VP and Model_Combination (P = 0.0465). However, the difference in AUCs between Model_AP and Model_VP was not significant (P = 0.5061). These results demonstrate that Model_Combination shows a better performance than the other models. Decision curve analysis proved the clinical utility of this model.

CONCLUSIONS

We developed a radiomics model based on monochromatic DECT images to identify solitary pulmonary nodules. This model could serve as an effective tool for discriminating benign from malignant pulmonary nodules in patients. The combination of arterial phase and venous phase imaging could significantly improve the model performance.

Deep learning combined with radiomics for the classification of enlarged cervical lymph nodes

PURPOSE

To investigate the application of deep learning combined with traditional radiomics methods for classifying enlarged cervical lymph nodes.

METHODS

The clinical and computed tomography (CT) imaging data of 276 patients with enlarged cervical lymph nodes (150 with lymph-node metastasis, 65 with lymphoma, and 61 with benign lymphadenopathy) who were treated at the hospital from January 2015 to January 2021 were retrospectively analysed. The patients were randomly divided into a training group and a test group at a ratio of 8:2. The radiomics features were extracted using one-by-one convolution and neural network activation, filtered with the least absolute shrinkage and selection operator (LASSO) model, and used to construct a discrimination model with PyTorch. Then, the performance of the model was compared with the radiologists' diagnostic performance. The neural network model was evaluated using the area under the receiver-operator characteristic curve (AUC), and the accuracy, sensitivity, and specificity were analysed.

RESULTS

A total of 102 features, comprising five traditional radiomic features and 97 deep learning features, were selected with LASSO and used to construct a discrimination model, which achieved a total accuracy of 87.50%. The AUC value, specificity, and sensitivity were, respectively, 0.92, 92.30%, and 90.00% for metastatic lymph nodes, 0.87, 95.45%, and 83.33% for benign lymphadenopathy, and 0.88, 90.47%, and 85.71% for lymphoma. The accuracies of the radiologists' diagnoses were 62.68% and 62.68%. The diagnostic performance of the model was significantly different from that of the radiologists (p < 0.05).

CONCLUSION

CT-based deep learning combined with the traditional radiomics methods has a high diagnostic value for the classification of cervical enlarged lymph nodes.

Multi-Size Deep Learning Based Preoperative Computed Tomography Signature for Prognosis Prediction of Colorectal Cancer

Background: Preoperative and postoperative evaluation of colorectal cancer (CRC) patients is crucial for subsequent treatment guidance. Our study aims to provide a timely and rapid assessment of the prognosis of CRC patients with deep learning according to non-invasive preoperative computed tomography (CT) and explore the underlying biological explanations.

Methods: A total of 808 CRC patients with preoperative CT (development cohort: n = 426, validation cohort: n = 382) were enrolled in our study. We proposed a novel end-to-end Multi-Size Convolutional Neural Network (MSCNN) to predict the risk of CRC recurrence with CT images (CT signature). The prognostic performance of CT signature was evaluated by Kaplan-Meier curve. An integrated nomogram was constructed to improve the clinical utility of CT signature by combining with other clinicopathologic factors. Further visualization and correlation analysis for CT deep features with paired gene expression profiles were performed to reveal the molecular characteristics of CRC tumors learned by MSCNN in radiographic imaging.

Results: The Kaplan-Meier analysis showed that CT signature was a significant prognostic factor for CRC disease-free survival (DFS) prediction [development cohort: hazard ratio (HR): 50.7, 95% CI: 28.4–90.6, p &lt; 0.001; validation cohort: HR: 2.04, 95% CI: 1.44–2.89, p &lt; 0.001]. Multivariable analysis confirmed the independence prognostic value of CT signature (development cohort: HR: 30.7, 95% CI: 19.8–69.3, p &lt; 0.001; validation cohort: HR: 1.83, 95% CI: 1.19–2.83, p = 0.006). Dimension reduction and visualization of CT deep features demonstrated a high correlation with the prognosis of CRC patients. Functional pathway analysis further indicated that CRC patients with high CT signature presented down-regulation of several immunology pathways. Correlation analysis found that CT deep features were mainly associated with activation of metabolic and proliferative pathways.

Conclusions: Our deep learning based preoperative CT signature can effectively predict prognosis of CRC patients. Integration analysis of multi-omic data revealed that some molecular characteristics of CRC tumor can be captured by deep learning in CT images.

Identifying factors that may influence the classification performance of radiomics models using contrast-enhanced mammography (CEM) images

Background

Radiomics plays an important role in the field of oncology. Few studies have focused on the identification of factors that may influence the classification performance of radiomics models. The goal of this study was to use contrast-enhanced mammography (CEM) images to identify factors that may potentially influence the performance of radiomics models in diagnosing breast lesions.

Methods

A total of 157 women with 161 breast lesions were included. Least absolute shrinkage and selection operator (LASSO) regression and the random forest (RF) algorithm were employed to construct radiomics models. The classification result for each lesion was obtained by using 100 rounds of five-fold cross-validation. The image features interpreted by the radiologists were used in the exploratory factor analyses. Univariate and multivariate analyses were performed to determine the association between the image features and misclassification. Additional exploratory analyses were performed to examine the findings.

Results

Among the lesions misclassified by both LASSO and RF ≥ 20% of the iterations in the cross-validation and those misclassified by both algorithms ≤5% of the iterations, univariate analysis showed that larger lesion size and the presence of rim artifacts and/or ripple artifacts were associated with more misclassifications among benign lesions, and smaller lesion size was associated with more misclassifications among malignant lesions (all p <  0.050). Multivariate analysis showed that smaller lesion size (odds ratio [OR] = 0.699, p = 0.002) and the presence of air trapping artifacts (OR = 35.568, p = 0.025) were factors that may lead to misclassification among malignant lesions. Additional exploratory analyses showed that benign lesions with rim artifacts and small malignant lesions (< 20 mm) with air trapping artifacts were misclassified by approximately 50% more in rate compared with benign and malignant lesions without these factors.

Conclusions

Lesion size and artifacts in CEM images may affect the diagnostic performance of radiomics models. The classification results for lesions presenting with certain factors may be less reliable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40644-022-00460-8.

Ultrasound-based radiomics analysis for preoperative prediction of central and lateral cervical lymph node metastasis in papillary thyroid carcinoma: a multi-institutional study

Background

An accurate preoperative assessment of cervical lymph node metastasis (LNM) is important for choosing an optimal therapeutic strategy for papillary thyroid carcinoma (PTC) patients. This study aimed to develop and validate two ultrasound (US) nomograms for the individual prediction of central and lateral compartment LNM in patients with PTC.

Methods

A total of 720 PTC patients from 3 institutions were enrolled in this study. They were categorized into a primary cohort, an internal validation, and two external validation cohorts. Radiomics features were extracted from conventional US images. LASSO regression was used to select optimized features to construct the radiomics signature. Two nomograms integrating independent clinical variables and radiomics signature were established with multivariate logistic regression. The performance of the nomograms was assessed with regard to discrimination, calibration, and clinical usefulness.

Results

The radiomics scores were significantly higher in patients with central/lateral LNM. A radiomics nomogram indicated good discrimination for central compartment LNM, with an area under the curve (AUC) of 0.875 in the training set, the corresponding value in the validation sets were 0.856, 0.870 and 0.870, respectively. Another nomogram for predicting lateral LNM also demonstrated good performance with an AUC of 0.938 and 0.905 in the training and internal validation cohorts, respectively. The AUC for the two external validation cohorts were 0.881 and 0.903, respectively. The clinical utility of the nomograms was confirmed by the decision curve analysis.

Conclusion

The nomograms proposed here have favorable performance for preoperatively predicting cervical LNM, hold promise for optimizing the personalized treatment, and might greatly facilitate the decision-making in clinical practice.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12880-022-00809-2.

Characterization of breast tumors using machine learning based upon multiparametric magnetic resonance imaging features

Magnetic resonance imaging (MRI) is playing an important role in the classification of breast tumors. MRI can be used to obtain multiparametric (mp) information, such as structural, hemodynamic, and physiological information. Quantitative analysis of mp-MRI data has shown potential in improving the accuracy of breast tumor classification. In general, a large set of quantitative and texture features can be generated depending upon the type of methodology used. A suitable combination of selected quantitative and texture features can further improve the accuracy of tumor classification. Machine learning (ML) classifiers based upon features derived from MRI data have shown potential in tumor classification. There is a need for further research studies on selecting an appropriate combination of features and evaluating the performance of different ML classifiers for accurate classification of breast tumors. The objective of the current study was to develop and optimize an ML framework based upon mp-MRI features for the characterization of breast tumors (malignant vs. benign and low- vs. high-grade). This study included the breast mp-MRI data of 60 female patients with histopathology results. A total of 128 features were extracted from the mp-MRI tumor data followed by features selection. Five ML classifiers were evaluated for tumor classification using 10-fold crossvalidation with 10 repetitions. The support vector machine (SVM) classifier based on optimum features selected using a wrapper method with an adaptive boosting (AdaBoost) technique provided the highest sensitivity (0.96 ± 0.03), specificity (0.92 ± 0.09), and accuracy (94% ± 2.91%) in the classification of malignant versus benign tumors. This method also provided the highest sensitivity (0.94 ± 0.07), specificity (0.80 ± 0.05), and accuracy (90% ± 5.48%) in the classification of low- versus high-grade tumors. These findings suggest that the SVM classifier outperformed other ML methods in the binary classification of breast tumors.

Development and validation of novel radiomics-based nomograms for the prediction of EGFR mutations and Ki-67 proliferation index in non-small cell lung cancer

BACKGROUND

We developed and validated novel radiomics-based nomograms to identify epidermal growth factor receptor (EGFR) mutations and the Ki-67 proliferation index of non-small cell lung cancer.

METHODS

We enrolled 132 patients with histologically verified non-small cell lung cancer from four hospital institutions who underwent computed tomography (CT) scans. EGFR mutations and the Ki-67 proliferation index were measured from tumor tissues. A total of 1,287 radiomic features were extracted, and a three-stage feature selection method was implemented to acquire the most valuable radiomic features. Finally, the radiomic scores and nomograms of the two tasks were established and tested. Receiver operating characteristic curves, calibration curves, and decision curves were used to evaluate their prediction performance and clinical utility.

RESULTS

In task [1], smoking status and histological type were significantly associated with EGFR mutations. After feature selection, 10 features were used to establish radiomic score, which showed good performance [area under the curve (AUC) =0.800] in the validation cohort. The radiomic nomogram had an AUC of 0.798 (95% CI: 0.664 to 0.931) with a C-index of 0.798 in the validation cohort. In task [2], gender, smoking status, histological type, and stage showed a significant correlation with Ki-67 proliferation index expression. A total of 28 features were selected to develop a radiomic score, with an AUC of 0.820 in the validation cohort. The final nomogram showed an AUC of 0.828 (95% CI: 0.703 to 0.953) with a C-index of 0.828 in the validation cohort.

CONCLUSIONS

EGFR mutations and Ki-67 proliferation index in non-small cell lung cancer can be predicted efficiently by the novel radiomic scores and nomograms.

A Machine Learning Model Based on PET/CT Radiomics and Clinical Characteristics Predicts Tumor Immune Profiles in Non-Small Cell Lung Cancer: A Retrospective Multicohort Study

Background

The tumor immune microenvironment (TIME) phenotypes have been reported to mainly impact the efficacy of immunotherapy. Given the increasing use of immunotherapy in cancers, knowing an individual's TIME phenotypes could be helpful in screening patients who are more likely to respond to immunotherapy. Our study intended to establish, validate, and apply a machine learning model to predict TIME profiles in non-small cell lung cancer (NSCLC) by using 18F-FDG PET/CT radiomics and clinical characteristics.

Methods

The RNA-seq data of 1145 NSCLC patients from The Cancer Genome Atlas (TCGA) cohort were analyzed. Then, 221 NSCLC patients from Daping Hospital (DPH) cohort received18F-FDG PET/CT scans before treatment and CD8 expression of the tumor samples were tested. The Artificial Intelligence Kit software was used to extract radiomic features of PET/CT images and develop a radiomics signature. The models were established by radiomics, clinical features, and radiomics-clinical combination, respectively, the performance of which was calculated by receiver operating curves (ROCs) and compared by DeLong test. Moreover, based on radiomics score (Rad-score) and clinical features, a nomogram was established. Finally, we applied the combined model to evaluate TIME phenotypes of NSCLC patients in The Cancer Imaging Archive (TCIA) cohort (n = 39).

Results

TCGA data showed CD8 expression could represent the TIME profiles in NSCLC. In DPH cohort, PET/CT radiomics model outperformed CT model (AUC: 0.907 vs. 0.861, P = 0.0314) to predict CD8 expression. Further, PET/CT radiomics-clinical combined model (AUC = 0.932) outperformed PET/CT radiomics model (AUC = 0.907, P = 0.0326) or clinical model (AUC = 0.868, P = 0.0036) to predict CD8 expression. In the TCIA cohort, the predicted CD8-high group had significantly higher immune scores and more activated immune pathways than the predicted CD8-low group (P = 0.0421).

Conclusion

Our study indicates that 18F-FDG PET/CT radiomics-clinical combined model could be a clinically practical method to non-invasively detect the tumor immune status in NSCLCs.

Prediction of Non-Response to Neoadjuvant Chemoradiotherapy in Esophageal Cancer Patients with 18F-FDG PET Radiomics Based Machine Learning Classification

Background: Approximately 26% of esophageal cancer (EC) patients do not respond to neoadjuvant chemoradiotherapy (nCRT), emphasizing the need for pre-treatment selection. The aim of this study was to predict non-response using a radiomic model on baseline ¹⁸F-FDG PET. Methods: Retrospectively, 143 ¹⁸F-FDG PET radiomic features were extracted from 199 EC patients (T1N1-3M0/T2–4aN0-3M0) treated between 2009 and 2019. Non-response (n = 57; 29%) was defined as Mandard Tumor Regression Grade 4–5 (n = 44; 22%) or interval progression (n = 13; 7%). Randomly, 139 patients (70%) were allocated to explore all combinations of 24 feature selection strategies and 6 classification methods towards the cross-validated average precision (AP). The predictive value of the best-performing model, i.e AP and area under the ROC curve analysis (AUC), was evaluated on an independent test subset of 60 patients (30%). Results: The best performing model had an AP (mean ± SD) of 0.47 ± 0.06 on the training subset, achieved by a support vector machine classifier trained on five principal components of relevant clinical and radiomic features. The model was externally validated with an AP of 0.66 and an AUC of 0.67. Conclusion: In the present study, the best-performing model on pre-treatment ¹⁸F-FDG PET radiomics and clinical features had a small clinical benefit to identify non-responders to nCRT in EC.

A CT-based radiomics model to predict subsequent brain metastasis in patients with ALK-rearranged non-small cell lung cancer undergoing crizotinib treatment

Background

Brain metastasis (BM) comprises the most common reason for crizotinib failure in patients with anaplastic lymphoma kinase (ALK)‐rearranged non–small cell lung cancer (NSCLC). We hypothesize that its occurrence could be predicted by a computed tomography (CT)‐based radiomics model, therefore, allowing for selection of enriched patient populations for prevention therapies.

Methods

A total of 75 eligible patients were enrolled from Sun Yat‐sen University Cancer Center between June 2014 and September 2019. The primary endpoint was brain metastasis‐free survival (BMFS), estimated from the initiation of crizotinib to the date of the occurrence of BM. Patients were randomly divided into two cohorts for model training (n = 51) and validation (n = 24), respectively. A radiomics signature was constructed based on features extracted from chest CT before crizotinib treatment. Clinical model was developed using the Cox proportional hazards model. Log‐rank test was performed to describe the difference of BMFS risk.

Results

Patients with low radiomics score had significantly longer BMFS than those with higher, both in the training cohort (p = 0.019) and validation cohort (p = 0.048). The nomogram combining smoking history and the radiomics signature showed good performance for the estimation of BMFS, both in the training (concordance index [C‐index], 0.762; 95% confidence interval [CI], 0.663–0.861) and validation cohort (C‐index, 0.724; 95% CI, 0.601–0.847).

Conclusion

We have developed a CT‐based radiomics model to predict subsequent BM in patients with non‐brain metastatic NSCLC undergoing crizotinib treatment. Selection of an enriched patient population at high BM risk will facilitate the design of clinical trials or strategies to prevent BM.

Radiomics in medical imaging: pitfalls and challenges in clinical management

BACKGROUND

Radiomics and radiogenomics are two words that recur often in language of radiologists, nuclear doctors and medical physicists especially in oncology field. Radiomics is the technique of medical images analysis to extract quantitative data that are not detected by human eye.

METHODS

This article is a narrative review on Radiomics in Medical Imaging. In particular, the review exposes the process, the limitations related to radiomics, and future prospects are discussed.

RESULTS

Several studies showed that radiomics is very promising. However, there were some critical issues: poor standardization and generalization of radiomics results, data-quality control, repeatability, reproducibility, database balancing and issues related to model overfitting.

CONCLUSIONS

Radiomics procedure should made considered all pitfalls and challenges to obtain robust and reproducible results that could be generalized in other patients cohort.

Computed tomography-based radiomic to predict resectability in locally advanced pancreatic cancer treated with chemotherapy and radiotherapy

BACKGROUND

Surgical resection after neoadjuvant treatment is the main driver for improved survival in locally advanced pancreatic cancer (LAPC). However, the diagnostic performance of computed tomography (CT) imaging to evaluate the residual tumour burden at restaging after neoadjuvant therapy is low due to the difficulty in distinguishing neoplastic tissue from fibrous scar or inflammation. In this context, radiomics has gained popularity over conventional imaging as a complementary clinical tool capable of providing additional, unprecedented information regarding the intratumor heterogeneity and the residual neoplastic tissue, potentially serving in the therapeutic decision-making process.

AIM

To assess the capability of radiomic features to predict surgical resection in LAPC treated with neoadjuvant chemotherapy and radiotherapy.

METHODS

Patients with LAPC treated with intensive chemotherapy followed by ablative radiation therapy were retrospectively reviewed. One thousand six hundred and fifty-five radiomic features were extracted from planning CT inside the gross tumour volume. Both extracted features and clinical data contribute to create and validate the predictive model of resectability status. Patients were repeatedly divided into training and validation sets. The discriminating performance of each model, obtained applying a LASSO regression analysis, was assessed with the area under the receiver operating characteristic curve (AUC). The validated model was applied to the entire dataset to obtain the most significant features.

RESULTS

Seventy-one patients were included in the analysis. Median age was 65 years and 57.8% of patients were male. All patients underwent induction chemotherapy followed by ablative radiotherapy, and 19 (26.8%) ultimately received surgical resection. After the first step of variable selections, a predictive model of resectability was developed with a median AUC for training and validation sets of 0.862 (95%CI: 0.792-0.921) and 0.853 (95%CI: 0.706-0.960), respectively. The validated model was applied to the entire dataset and 4 features were selected to build the model with predictive performance as measured using AUC of 0.944 (95%CI: 0.892-0.996).

CONCLUSION

The present radiomic model could help predict resectability in LAPC after neoadjuvant chemotherapy and radiotherapy, potentially integrating clinical and morphological parameters in predicting surgical resection.