Dataset Information

Radiomics-Derived Data by Contrast Enhanced Magnetic Resonance in RAS Mutations Detection in Colorectal Liver Metastases.

ABSTRACT: Purpose: To assess the association of RAS mutation status and radiomics-derived data by Contrast Enhanced-Magnetic Resonance Imaging (CE-MRI) in liver metastases. Materials and Methods: 76 patients (36 women and 40 men; 59 years of mean age and 36-80 years as range) were included in this retrospective study. Texture metrics and parameters based on lesion morphology were calculated. Per-patient univariate and multivariate analysis were made. Wilcoxon-Mann-Whitney U test, receiver operating characteristic (ROC) analysis, pattern recognition approaches with features selection approaches were considered. Results: Significant results were obtained for texture features while morphological parameters had not significant results to classify RAS mutation. The results showed that using a univariate analysis was not possible to discriminate accurately the RAS mutation status. Instead, considering a multivariate analysis and classification approaches, a KNN exclusively with texture parameters as predictors reached the best results (AUC of 0.84 and an accuracy of 76.9% with 90.0% of sensitivity and 67.8% of specificity on training set and an accuracy of 87.5% with 91.7% of sensitivity and 83.3% of specificity on external validation cohort). Conclusions: Texture parameters derived by CE-MRI and combined using multivariate analysis and patter recognition approaches could allow stratifying the patients according to RAS mutation status.

SUBMITTER: Granata V

PROVIDER: S-EPMC7865653 | biostudies-literature | 2021 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Radiomics-Derived Data by Contrast Enhanced Magnetic Resonance in RAS Mutations Detection in Colorectal Liver Metastases.

Granata Vincenza V Fusco Roberta R Avallone Antonio A De Stefano Alfonso A Ottaiano Alessandro A Sbordone Carolina C Brunese Luca L Izzo Francesco F Petrillo Antonella A

Cancers 20210125 3

<i>Purpose</i>: To assess the association of RAS mutation status and radiomics-derived data by Contrast Enhanced-Magnetic Resonance Imaging (CE-MRI) in liver metastases. <i>Materials and Methods</i>: 76 patients (36 women and 40 men; 59 years of mean age and 36-80 years as range) were included in this retrospective study. Texture metrics and parameters based on lesion morphology were calculated. Per-patient univariate and multivariate analysis were made. Wilcoxon-Mann-Whitney U test, receiver op ...[more]

PMID: 33504085

Similar Datasets

Project description:BackgroundPerihepatic fluorouracil encapsulated lesions (FELs) can result in potentially confusing computed tomography (CT) and magnetic resonance imaging (MRI) features in postoperative examinations of gastrointestinal tumors. This retrospective study aimed to summarize the typical imaging features of FELs and determine the best imaging modality to distinguish FELs from liver metastases for junior residents.MethodsPatients with FELs who had undergone gastrointestinal tumor surgery in Tongji Hospital from January 2016 to June 2022 were evaluated. The imaging features of FELs were summarized by two senior radiologists. Contrast-enhanced CT (CECT) was used as the primary follow-up tool for postoperative gastrointestinal tumor patients. Patients with FELs and available CECT and MRI examinations were matched with patients with liver metastases based on gender and age and presented in chronological order in a 2:1 ratio. Different imaging modality combinations were used for further evaluation, including a CECT group (modality Ⅰ), CECT and nonenhanced MRI group (modality Ⅱ) and CECT with all MRI sequences group (modality Ⅲ). Subsequently, two junior residents blindly evaluated three groups following a 4-week interval based on a 5-point scale (1= definite benign lesion, 2= probable benign lesion, 3= indeterminate, 4= probable liver metastasis, 5= definite liver metastasis).ResultsImaging features of 33 patients with 36 FELs were analyzed. CECT and dynamic contrast-enhanced MRI (DCE-MRI) showed no enhancement in most lesions. Additionally, 20 patients with FELs meeting the requirements were matched with 40 patients with liver metastases. The highest sensitivity, specificity, and consistency for identifying liver metastases were achieved using a combination of CECT and MRI encompassing all sequences yielded, including modality Ⅰ (reader 1: 72.0% and 17.4%; reader 2: 62.0% and 17.4%; kappa value 0.295), modality Ⅱ (reader 1: 88.0% and 8.7%; reader 2: 92.0% and 34.8%; kappa value 0.259), and modality Ⅲ (reader 1: 98.0% and 34.8%; reader 2: 92.0% and 39.1%; kappa value 0.680).ConclusionsFELs are typically non-enhancing lesions. In our study, two junior residents could best distinguish FELs from liver metastases using CECT with all MRI sequences.

Project description:BackgroundBreast cancer is a heterogeneous disease with different morphological and biological characteristics. The molecular subtypes of breast cancer are closely related to the treatment and prognosis of patients. In order to predict the luminal type of breast cancer in a noninvasive manner, our study developed and validated a radiomics nomogram combining clinical factors with a radiomics score based on the features of the intratumoral subregion to distinguish between luminal and nonluminal breast cancer.MethodsFrom January 2018 to January 2020, 153 women with clinically and pathologically diagnosed breast cancer with an average age of 50.08 years were retrospectively analyzed. Using a semiautomatic segmentation method, the whole tumor was divided into 3 subregions on the basis of the time required for the contrast agent to reach its peak; 540 features were extracted from 3 subregions and the whole tumor region. Subsequently, 2 machine learning classifiers were developed. The least absolute shrinkage and selection operator method was used for feature selection and radiomics score (Rad-score) construction. Moreover, multivariable logistic regression analysis was applied to select independent factors from the Rad-score and clinical factors to establish a prediction model in the form of a nomogram. The performance of the nomogram was evaluated through calibration, discrimination, and clinical usefulness.ResultsThe prediction performance of texture features from the rapid subregion was the best in the 3 intratumoral subregions, and the area under the receiver operating characteristic curve (AUC) values in the training and validation cohort were 0.805 (95% CI: 0.719-0.892) and 0.737 (95% CI: 0.581-0.893), respectively. The Rad-score, consisting of 5 features from the rapid subregion, was associated with the luminal type of breast cancer (P=0.001 and P=0.035 in the training and validation cohorts, respectively). The predictors included in the personalized prediction nomogram included Rad-score, human epidermal growth factor receptor 2 (HER2) status, and tumor histological grade. The nomogram showed good discrimination, with an area under the receiver operating characteristic curve in the training and validation cohorts of 0.830 (95% CI: 0.746-0.896) and 0.879 (95% CI: 0.748-0.957), respectively. The calibration curve of the 2 cohorts and decision curve analysis demonstrated that the nomogram had good calibration and clinical usefulness.ConclusionsWe proposed a nomogram model that combined clinical factors and Rad-score, which showed good performance in predicting the luminal type of breast cancer.

Dataset Information

Radiomics-Derived Data by Contrast Enhanced Magnetic Resonance in RAS Mutations Detection in Colorectal Liver Metastases.

Publications

Radiomics-Derived Data by Contrast Enhanced Magnetic Resonance in RAS Mutations Detection in Colorectal Liver Metastases.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets