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

BACKGROUND: Lymph node metastasis (LNM) is the principal risk factor for poor outcomes in early-stage cervical cancer. Radiomics may offer a noninvasive way for predicting the stage of LNM. PURPOSE: To evaluate a radiomic signature of LN involvement based on sagittal T1 contrast-enhanced (CE) and T2 MRI sequences. STUDY TYPE: Retrospective. POPULATION: In all, 143 patients were randomly divided into two primary and validation cohorts with 100 patients in the primary cohort and 43 patients in the validation cohort. FIELD STRENGTH/SEQUENCE: T1 CE and T2 MRI sequences at 3T. ASSESSMENT: The gold standard of LN status was based on histologic results. A radiologist with 10 years of experience used the ITK-SNAP software for 3D manual segmentation. A senior radiologist with 15 years of experience validated all segmentations. The area under the receiver operating characteristics curve (ROC AUC), classification accuracy, sensitivity, and specificity were used between LNM and non-LNM groups. STATISTICAL TESTS: A total of 970 radiomic features and seven clinical characteristics were extracted. Minimum redundancy / maximum relevance and support vector machine algorithms were applied to select features and construct a radiomic signature. The Mann-Whitney U-test and the chi-square test were used to test the performance of clinical characteristics and potential prognostic outcomes. The results were used to assess the quantitative discrimination performance of the SVM-based radiomic signature. RESULTS: The radiomic signatures allowed good discrimination between LNM and non-LNM groups. The ROC AUC was 0.753 (95% confidence interval [CI], 0.656-0.850) in the primary cohort and 0.754 (95% CI, 0584-0.924) in the validation cohort. DATA CONCLUSIONS: A multiple-sequence MRI radiomic signature can be used as a noninvasive biomarker for preoperative assessment of LN status and potentially influence the therapeutic decision-making in early-stage cervical cancer patients. LEVEL OF EVIDENCE: 3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;49:304-310.

Radiomics in multiple sclerosis and neuromyelitis optica spectrum disorder.

OBJECTIVE: To develop and validate an individual radiomics nomogram for differential diagnosis between multiple sclerosis (MS) and neuromyelitis optica spectrum disorder (NMOSD). METHODS: We retrospectively collected 67 MS and 68 NMOSD with spinal cord lesions as a primary cohort and prospectively recruited 28 MS and 26 NMOSD patients as a validation cohort. Radiomic features were extracted from the spinal cord lesions. A prediction model for differentiating MS and NMOSD was built by combining the radiomic features with several clinical and routine MRI measurements. The performance of the model was assessed with respect to its calibration plot and clinical discrimination in the primary and validation cohorts. RESULTS: Nine radiomics features extracted from an initial set of 485, predominantly reflecting lesion heterogeneity, combined with lesion length, patient sex, and EDSS, were selected to build the model for differentiating MS and NMOSD. The areas under the ROC curves (AUC) for differentiating the two diseases were 0.8808 and 0.7115, for the primary and validation cohort, respectively. This model demonstrated good calibration (C-index was 0.906 and 0.802 in primary and validation cohort). CONCLUSIONS: A validated nomogram that incorporates the radiomic signature of spinal cord lesions, as well as cord lesion length, sex, and EDSS score, can usefully differentiate MS and NMOSD. KEY POINTS: • Radiomic features of spinal cord lesions in MS and NMOSD were different. • Radiomic signatures can capture pathological alterations and help differentiate MS and NMOSD.

Prognostic value of computed tomography radiomics features in patients with gastric cancer following curative resection.

OBJECTIVES: The present study aimed to investigate the clinical prognostic significance of radiomics signature (R-signature) in patients with gastric cancer who had undergone radical resection. METHODS: A total of 181 patients with gastric cancer who had undergone radical resection were enrolled in this retrospective study. The association between the R-signature and overall survival (OS) was assessed in the primary cohort and verified in the validation cohort. Furthermore, the performance of a radiomics nomogram integrating the R-signature and significant clinicopathological risk factors was evaluated. RESULTS: The R-signature, which consisted of six imaging features, stratified patients with gastric cancer who had undergone radical resection into two prognostic risk groups in both cohorts. The radiomics nomogram incorporating R-signature and significant clinicopathological risk factors (T stage, N stage, and differentiation) exhibited significant prognostic superiority over clinical nomogram and R-signature alone (Harrell concordance index, 0.82 vs 0.71 and 0.82 vs 0.74, respectively, p < 0.001 in both analyses). All calibration curves showed remarkable consistency between predicted and actual survival, and decision curve analysis verified the usefulness of the radiomics nomogram for clinical practice. CONCLUSIONS: The R-signature could be used to stratify patients with gastric cancer following radical resection into high- and low-risk groups. Furthermore, the radiomics nomogram provided better predictive accuracy than other predictive models and might aid clinicians with therapeutic decision-making and patient counseling. KEY POINTS: • Radiomics can stratify the gastric cancer patients following radical resection into high- and low-risk groups. • Radiomics can improve the prognostic value of TNM staging system. • Radiomics may facilitate personalized treatment of gastric cancer patients.

A multi-view co-training network for semi-supervised medical image-based prognostic prediction.

Prognostic prediction has long been a hotspot in disease analysis and management, and the development of image-based prognostic prediction models has significant clinical implications for current personalized treatment strategies. The main challenge in prognostic prediction is to model a regression problem based on censored observations, and semi-supervised learning has the potential to play an important role in improving the utilization efficiency of censored data. However, there are yet few effective semi-supervised paradigms to be applied. In this paper, we propose a semi-supervised co-training deep neural network incorporating a support vector regression layer for survival time estimation (Co-DeepSVS) that improves the efficiency in utilizing censored data for prognostic prediction. First, we introduce a support vector regression layer in deep neural networks to deal with censored data and directly predict survival time, and more importantly to calculate the labeling confidence of each case. Then, we apply a semi-supervised multi-view co-training framework to achieve accurate prognostic prediction, where labeling confidence estimation with prior knowledge of pseudo time is conducted for each view. Experimental results demonstrate that the proposed Co-DeepSVS has a promising prognostic ability and surpasses most widely used methods on a multi-phase CT dataset. Besides, the introduction of SVR layer makes the model more robust in the presence of follow-up bias.

Development and Validation of a Deep Learning Model to Screen for Trisomy 21 During the First Trimester From Nuchal Ultrasonographic Images.

Importance: Accurate screening of trisomy 21 in the first trimester can provide an early opportunity for decision-making regarding reproductive choices. Objective: To develop and validate a deep learning model for screening fetuses with trisomy 21 based on ultrasonographic images. Design, Setting, and Participants: This diagnostic study used data from all available cases and controls enrolled at 2 hospitals in China between January 2009 and September 2020. Two-dimensional images of the midsagittal plane of the fetal face in singleton pregnancies with gestational age more than 11 weeks and less than 14 weeks were examined. Observers were blinded to subjective fetus nuchal translucency (NT) marker measurements. A convolutional neural network was developed to construct a deep learning model. Data augmentation was applied to generate more data. Different groups were randomly selected as training and validation sets to assess the robustness of the deep learning model. The fetal NT was shown and measured. Each detection of trisomy 21 was confirmed by chorionic villus sampling or amniocentesis. Data were analyzed from March 1, 2021, to January 3, 2022. Main Outcomes and Measures: The primary outcome was detection of fetuses with trisomy 21. The receiver operating characteristic curve, metrics of accuracy, area under the curve (AUC), sensitivity, and specificity were used for model performance evaluation. Results: A total of 822 case and control participants (mean [SD] age, 31.9 [4.6] years) were enrolled in the study, including 550 participants (mean [SD] age, 31.7 [4.7] years) in the training set and 272 participants (mean [SD] age, 32.3 [4.7] years) in the validation set. The deep learning model showed good performance for trisomy 21 screening in the training (AUC, 0.98; 95% CI, 0.97-0.99) and validation (AUC, 0.95; 95% CI, 0.93-0.98) sets. The deep learning model had better detective performance for fetuses with trisomy 21 than the model with NT marker and maternal age (training: AUC, 0.82; 95% CI, 0.77-0.86; validation: AUC, 0.73; 95% CI, 0.66-0.80). Conclusions and Relevance: These findings suggest that this deep learning model accurately screened fetuses with trisomy 21, which indicates that the model is a potential tool to facilitate universal primary screening for trisomy 21.

Knowledge-guided multi-task attention network for survival risk prediction using multi-center computed tomography images.

Accurate preoperative prediction of overall survival (OS) risk of human cancers based on CT images is greatly significant for personalized treatment. Deep learning methods have been widely explored to improve automated prediction of OS risk. However, the accuracy of OS risk prediction has been limited by prior existing methods. To facilitate capturing survival-related information, we proposed a novel knowledge-guided multi-task network with tailored attention modules for OS risk prediction and prediction of clinical stages simultaneously. The network exploits useful information contained in multiple learning tasks to improve prediction of OS risk. Three multi-center datasets, including two gastric cancer datasets with 459 patients, and a public American lung cancer dataset with 422 patients, are used to evaluate our proposed network. The results show that our proposed network can boost its performance by capturing and sharing information from other predictions of clinical stages. Our method outperforms the state-of-the-art methods with the highest geometrical metric. Furthermore, our method shows better prognostic value with the highest hazard ratio for stratifying patients into high- and low-risk groups. Therefore, our proposed method may be exploited as a potential tool for the improvement of personalized treatment.

Multi-Focus Network to Decode Imaging Phenotype for Overall Survival Prediction of Gastric Cancer Patients.

Gastric cancer (GC) is the third leading cause of cancer-associated deaths globally. Accurate risk prediction of the overall survival (OS) for GC patients shows significant prognostic value, which helps identify and classify patients into different risk groups to benefit from personalized treatment. Many methods based on machine learning algorithms have been widely explored to predict the risk of OS. However, the accuracy of risk prediction has been limited and remains a challenge with existing methods. Few studies have proposed a framework and pay attention to the low-level and high-level features separately for the risk prediction of OS based on computed tomography images of GC patients. To achieve high accuracy, we propose a multi-focus fusion convolutional neural network. The network focuses on low-level and high-level features, where a subnet to focus on lower-level features and the other enhanced subnet with lateral connection to focus on higher-level semantic features. Three independent datasets of 640 GC patients are used to assess our method. Our proposed network is evaluated by metrics of the concordance index and hazard ratio. Our network outperforms state-of-the-art methods with the highest concordance index and hazard ratio in independent validation and test sets. Our results prove that our architecture can unify the separate low-level and high-level features into a single framework, and can be a powerful method for accurate risk prediction of OS.

Multi-Habitat Based Radiomics for the Prediction of Treatment Response to Concurrent Chemotherapy and Radiation Therapy in Locally Advanced Cervical Cancer.

Objectives: To develop a radiomic model based on multiparametric magnetic resonance imaging (MRI) for predicting treatment response prior to commencing concurrent chemotherapy and radiation therapy (CCRT) for locally advanced cervical cancer. Materials and methods: The retrospective study enrolled 120 patients (allocated to a training or a test set) with locally advanced cervical cancer who underwent CCRT between December 2014 and June 2017. All patients enrolled underwent MRI with nine sequences before treatment and again at the end of the fourth week of treatment. Responses were evaluated by MRI according to RECIST standards, and patients were divided into a responder group or non-responder group. For every MRI sequence, a total of 114 radiomic features were extracted from the outlined tumor habitat. On the training set, the least absolute shrinkage and selection operator method was used to select key features and to construct nine habitat signatures. Then, three kinds of machine learning models were compared and applied to integrate these predictive signatures and the clinical characteristics into a radiomic model. The discrimination ability, reliability, and calibration of our radiomic model were evaluated. Results: The radiomic model, which consisted of three habitat signatures from sagittal T2 image, axial T1 enhanced-MRI image, and ADC image, respectively, has shown good predictive performance, with area under the curve of 0.820 (95% CI: 0.713-0.927) in the training set and 0.798 (95% CI: 0.678-0.917) in the test set. Meanwhile, the model proved to perform better than each single signature or clinical characteristic. Conclusions: A radiomic model employing features from multiple tumor habitats held the ability for predicting treatment response in patients with locally advanced cervical cancer before commencing CCRT. These results illustrated a potential new tool for improving medical decision-making and therapeutic strategies.

Development and validation of a CT-based radiomic nomogram for preoperative prediction of early recurrence in advanced gastric cancer.

BACKGROUND: In the clinical management of advanced gastric cancer (AGC), preoperative identification of early recurrence after curative resection is essential. Thus, we aimed to create a CT-based radiomic model to predict early recurrence in AGC patients preoperatively. MATERIALS AND METHODS: We enrolled 669 consecutive patients (302 in the training set, 219 in the internal test set and 148 in the external test set) with clinicopathologically confirmed AGC from two centers. Radiomic features were extracted from preoperative diagnostic CT images. Machine learning methods were applied to shrink feature size and build a predictive radiomic signature. We incorporated the radiomic signature and clinical risk factors into a nomogram using multivariable logistic regression analysis. The area under the curve (AUC) of operating characteristics (ROC), accuracy, and calibration curves were assessed to evaluate the nomogram's performance in discriminating early recurrence. RESULTS: A radiomic signature, including three hand crafted features and six deep learning features, was significantly associated with early recurrence (p-value <0.0001 for all sets). In addition, clinical N stage, carbohydrate antigen 199 levels, carcinoembryonic antigen levels, and Borrmann type were considered useful predictors for early recurrence. The nomogram, combining all these predictors, showed powerful prognostic ability in the training set and two test sets with AUCs of 0.831 (95% CI, 0.786-0.876), 0.826 (0.772-0.880) and 0.806 (0.732-0.881), respectively. The predicted risk yielded good agreement with the observed recurrence probability. CONCLUSIONS: By incorporating a radiomic signature and clinical risk factors, we created a radiomic nomogram to predict early recurrence in patients with AGC, preoperatively, which may serve as a potential tool to guide personalized treatment.

Multiscale imaging of colitis in mice using confocal laser endomicroscopy, light-sheet fluorescence microscopy, and magnetic resonance imaging.

The objective of our study is to develop a multimodality approach by combining magnetic resonance imaging (MRI) and optical imaging methods to assess acute murine colitis at the macro- and microscopic level. In vivo MRI is used to measure the cross-sectional areas of colons at the macroscopic level. Dual-color confocal laser endomicroscopy (CLE) allows in vivo examination of the fluorescently labeled epithelial cells and microvessels in the mucosa with a spatial resolution of ∼1.4 µm during ongoing endoscopy. To further validate the structural changes of the colons in three-dimensions, ex vivo light-sheet fluorescence microscopy (LSFM) is applied for in-toto imaging of cleared colon sections. MRI, LSFM, and CLE findings are significantly correlated with histological scoring (p < 0.01) and the inflammation-associated activity index (p < 0.01). Our multimodality imaging technique permits visualization of mucosa in colitis at different scales, which can enhance our understanding of the pathogenesis of inflammatory bowel diseases.

Using biparametric MRI radiomics signature to differentiate between benign and malignant prostate lesions.

PURPOSE: To investigate the efficiency of radiomics signature in discriminating between benign and malignant prostate lesions with similar biparametric magnetic resonance imaging (bp-MRI) findings. EXPERIMENTAL DESIGN: Our study consisted of 331 patients underwent bp-MRI before pathological examination from January 2013 to November 2016. Radiomics features were extracted from peripheral zone (PZ), transition zone (TZ), and lesion areas segmented on images obtained by T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and its derivative apparent-diffusion coefficient (ADC) imaging. The individual prediction model, built using the clinical data and biparametric MRI features (Bp signature), was prepared using data of 232 patients and validated using data of 99 patients. The predictive performance was calculated and demonstrated using receiver operating characteristic (ROC) curves, calibration curves, and decision curves. RESULTS: The Bp signature, based on the six selected radiomics features of bp-MRI, showed better discrimination in the validation cohort (area under the curve [AUC], 0.92) than on each subcategory images (AUC, 0.81 on T2WI; AUC, 0.77 on DWI; AUC, 0.89 on ADC). The differential diagnostic efficiency was poorer with the clinical model (AUC, 0.73), built using the selected independent clinical risk factors with statistical significance (P < 0.05), than with the Bp signature. Discrimination efficiency improved when including the Bp signature and clinical factors [i.e., the individual prediction model (AUC, 0.93)]. CONCLUSION: The Bp signature, based on bp-MRI, performed better than each single imaging modality. The individual prediction model including the radiomics signatures and clinical factors showed better preoperative diagnostic performance, which could contribute to clinical individualized treatment.

Quantitative Biomarkers for Prediction of Epidermal Growth Factor Receptor Mutation in Non-Small Cell Lung Cancer.

OBJECTIVES: To predict epidermal growth factor receptor (EGFR) mutation status using quantitative radiomic biomarkers and representative clinical variables. METHODS: The study included 180 patients diagnosed as of non-small cell lung cancer (NSCLC) with their pre-therapy computed tomography (CT) scans. Using a radiomic method, 485 features that reflect the heterogeneity and phenotype of tumors were extracted. Afterwards, these radiomic features were used for predicting epidermal growth factor receptor (EGFR) mutation status by a least absolute shrinkage and selection operator (LASSO) based on multivariable logistic regression. As a result, we found that radiomic features have prognostic ability in EGFR mutation status prediction. In addition, we used radiomic nomogram and calibration curve to test the performance of the model. RESULTS: Multivariate analysis revealed that the radiomic features had the potential to build a prediction model for EGFR mutation. The area under the receiver operating characteristic curve (AUC) for the training cohort was 0.8618, and the AUC for the validation cohort was 0.8725, which were superior to prediction model that used clinical variables alone. CONCLUSION: Radiomic features are better predictors of EGFR mutation status than conventional semantic CT image features or clinical variables to help doctors to decide who need EGFR tyrosine kinase inhibitor (TKI) treatment.