Background: This study investigated the efficacy of radiomics to predict survival outcome for locally advanced breast cancer (LABC) patients and the association of radiomics with tumor heterogeneity and microenvironment. Methods: Patients with LABC from 2010 to 2015 were retrospectively reviewed. Radiomics features were extracted from enhanced MRI. We constructed the radiomics score using lasso and assessed its prognostic value. An external validation cohort from The Cancer Imaging Archive was used to assess phenotype reproducibility. Sequencing data from TCGA and our center were applied to reveal genomic landscape of different radiomics score groups. Tumor infiltrating lymphocytes map and bioinformatics methods were applied to evaluate the heterogeneity of tumor microenvironment. Computational histopathology was also applied. Results: A total of 278 patients were divided into training cohort and validation cohort. Radiomics score was constructed and significantly associated with disease-free survival (DFS) of the patients in training cohort, validation cohort and external validation cohort (p < 0.001, p = 0.014 and p = 0.041, respectively). The radiomics-based nomogram showed better predictive performance of DFS compared with TNM model. Distinct gene expression patterns were identified. Immunophenotype and immune cell composition was different in each radiomics score group. The link between radiomics and computational histopathology was revealed. Conclusions: The radiomics score could effectively predict prognosis of LABC after neoadjuvant chemotherapy and radiotherapy. Radiomics revealed heterogeneity of tumor cell and tumor microenvironment and holds great potential to facilitate individualized DFS estimation and guide personalized care.
Objective: The aim of the study was to construct and validate a nomogram for differentiating follicular thyroid carcinoma (FTC) from follicular thyroid adenoma (FTA). Methods: Two hundred patients with pathologically confirmed thyroid follicular neoplasms were retrospectively analyzed. The patients were randomly divided into a training set (n = 140) and validation set (n = 60). Baseline data including demographics, CT (computed tomography) signs, and radiomic features were analyzed. Predictive models were developed and compared to build a nomogram. The predictive effectiveness of it was evaluated by the area under receiver operating characteristic curve (AUC). Results: The CT model, radiomic model and combination model showed excellent discrimination (AUCs [95% confidence interval] = 0.847 [0.766–0.928], 0.863 [0.746–0.932], 0.913 [0.850–0.975]). The nomogram based on the combination model showed remarkable discrimination in the training and validation sets. The calibration curves suggested good consistency between actual observation and prediction. Conclusions: This study proposed a nomogram that can accurately and intuitively predict the malignancy potential of follicular thyroid neoplasms.
Objectives: To investigate the value of CT-based radiomics analysis in preoperatively discriminating pancreatic mucinous cystic neoplasms (MCN) and atypical serous cystadenomas (ASCN).Methods: A total of 103 MCN and 113 ASCN patients who underwent surgery were retrospectively enrolled. A total of 764 radiomics features were extracted from preoperative CT images. The optimal features were selected by Mann-Whitney U test and minimum redundancy and maximum relevance method. The radiomics score (Rad-score) was then built using random forest algorithm. Radiological/clinical features were also assessed for each patient. Multivariable logistic regression was used to construct a radiological model. The performance of the Rad-score and the radiological model was evaluated using 10-fold cross-validation for area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy.Results: Ten screened optimal features were identified and the Rad-score was then built based on them. The radiological model was built based on four radiological/clinical factors. In the 10-fold cross-validation, the Rad-score was proved to be robust and reliable (average AUC: 0.784, sensitivity: 0.847, specificity: 0.745, PPV: 0.767, NPV: 0.849, accuracy: 0.793). The radiological model performed slightly less well in classification (average AUC: average AUC: 0.734 sensitivity: 0.748, specificity: 0.705, PPV: 0.732, NPV: 0.798, accuracy: 0.728.Conclusions: The CT-based radiomics analysis provided promising performance for preoperatively discriminating MCN from ASCN and showed good potential in improving diagnostic power, which may serve as a novel tool for guiding clinical decision-making for these patients.
Contrast-enhanced fluid-attenuated inversion-recovery (FLAIR) magnetic resonance imaging (MRI) has been reported to have higher sensitivity for detecting leptomeningeal disease compared with contrastenhanced T1-weighted MRI (CE T1WI). However, currently there are no studies showing the potential value of clinical applications of contrast-enhanced FLAIR (CE FLAIR) sequence in diagnosing intracranial tumors in a larger group of patients. The purpose of this study was to evaluate the diagnostic value of CE FLAIR in comparison with CE T1WI for intracranial tumors and to provide more information for clinical diagnosis and therapy. One hundred and four consecutive cases of intracranial tumors referred for CE brain MRI were analyzed with regard to FLAIR and T1WI pre- and post-administration of Gd-DTPA. The CE FLAIR and CE T1WI were evaluated independently by two radiologists for the number of examinations with one or more enhanced lesions, the number and location of enhanced lesions per examination, signal-to-noise ratio (SNR) and contrast-enhancement ratio (CER) of lesions, as well as the size and extent of the enhanced lesions. In 98 of 104 cases, enhanced lesions were seen both on the FLAIR and T1W images. More lesions were seen on CE T1WI (n=120) than those on CE FLAIR sequence (n=117), but no differences of statistical significance were found between the two sequences (P>0.05). Four lesions were revealed only on the CE FLAIR images whereas 7 lesions were only found on CE T1WI. Enhanced lesions located in the cerebral hemisphere or the forth ventricle were revealed much more on CE T1WI than on CE FLAIR images. However, CE FLAIR images may be useful in showing superficial abnormalities and those located in the sulcus or lateral ventricle. The CER and contrast-to-noise ratio (CNR) on CE T1WI was significantly higher (t=7.10, P=0.00; t=9.67, P=0.00, respectively), but grey matter/white matter contrast was lower (t=2.46, P=0.02) than those on CE FLAIR images. The SNR did not show any statistically significant difference between the two sequences (t=1.1, P=0.27). The size and extent of lesions on the CE FLAIR images were significantly larger than those on CE T1WI (t=4.13, P=0.00). CE FLAIR and CE T1WI may complement each other in showing intracranial tumors and the CE FLAIR sequence should be selected as a routine MRI sequence.