Staging liver fibrosis: comparison of radiomics model and fusion model based on multiparametric MRI in patients with chronic liver disease

OBJECTIVES

To develop and compare radiomics model and fusion model based on multiple MR parameters for staging liver fibrosis in patients with chronic liver disease.

MATERIALS AND METHODS

Patients with chronic liver disease who underwent multiparametric abdominal MRI were included in this retrospective study. Multiparametric MR images were imported into 3D-Slicer software for drawing bounding boxes on MR images. By using a 3D-Slicer extension of SlicerRadiomics, radiomics features were extracted from these MR images. The z-score normalization method was used for post-processing radiomics features. The least absolute shrinkage and selection operator method (LASSO) was performed for selecting significant radiomics features. The logistic regression analysis was used for building the radiomics model. A fusion model was built by integrating serum fibrosis biomarkers of aspartate transaminase-to-platelet ratio index (APRI) and the fibrosis-4 index (FIB-4) with radiomics signatures.

RESULTS

In the training cohort, AUCs of radiomics and fusion model were 0.707-0.842 and 0.718-0.854 for differentiating different groups. In the testing cohort, AUCs were 0.514-0.724 and 0.609-0.728. For the training cohort, there was no significant difference of AUCs between radiomics and fusion model (p > 0.05). For the testing cohort, AUCs of fusion model were higher than those of radiomics model in differentiating F1-3 vs. F4 and F1-2 vs. F4 (p = 0.011 & 0.042).

CONCLUSIONS

Radiomics model and fusion model based on multiparametric MRI exhibited the feasibility for staging liver fibrosis in patients with CLD, and APRI and FIB-4 could improve the diagnostic performance of radiomics model in differentiating F1-3 vs. F4 and F1-2 vs. F4.

Automated breast volume scanner based Radiomics for non-invasively prediction of lymphovascular invasion status in breast cancer

PURPOSE

Lymphovascular invasion (LVI) indicates resistance to preoperative adjuvant chemotherapy and a poor prognosis and can only be diagnosed by postoperative pathological examinations in breast cancer. Thus, a technique for preoperative diagnosis of LVI is urgently needed. We aim to explore the ability of an automated breast volume scanner (ABVS)-based radiomics model to noninvasively predict the LVI status in breast cancer.

METHODS

We conducted a retrospective analysis of data from 335 patients diagnosed with T1-3 breast cancer between October 2019 and September 2022. The patients were divided into training cohort and validation cohort with a ratio of 7:3. For each patient, 5901 radiomics features were extracted from ABVS images. Feature selection was performed using LASSO method. We created machine learning models for different feature sets with support vector machine algorithm to predict LVI. And significant clinicopathologic factors were identified by univariate and multivariate logistic regression to combine with three radiomics signatures as to develop a fusion model.

RESULTS

The three SVM-based prediction models, demonstrated relatively high efficacy in identifying LVI of breast cancer, with AUCs of 79.00%, 80.00% and 79.40% and an accuracy of 71.00%, 80.00% and 75.00% in the validation cohort for AP, SP and CP plane image. The fusion model achieved the highest AUC of 87.90% and an accuracy of 85.00% in the validation cohort.

CONCLUSIONS

The combination of radiomics features from ABVS images and an SVM prediction model showed promising performance for preoperative noninvasive prediction of LVI in breast cancer.

A radiomics model for predicting the response to methylprednisolone in brain necrosis after radiotherapy for nasopharyngeal carcinoma

BACKGROUND

Methylprednisolone is recommended as the front-line therapy for radiation-induced brain necrosis (RN) after radiotherapy for nasopharyngeal carcinoma. However, some patients fail to benefit from methylprednisolone or even progress. This study aimed to develop and validate a radiomic model to predict the response to methylprednisolone in RN.

METHODS

Sixty-six patients receiving methylprednisolone were enrolled. In total, 961 radiomic features were extracted from the pre-treatment magnetic resonance imagings of the brain. Least absolute shrinkage and selection operator regression was then applied to construct the radiomics signature. Combined with independent clinical predictors, a radiomics model was built with multivariate logistic regression analysis. Discrimination, calibration and clinical usefulness of the model were assessed. The model was internally validated using 10-fold cross-validation.

RESULTS

The radiomics signature consisted of 16 selected features and achieved favorable discrimination performance. The radiomics model incorporating the radiomics signature and the duration between radiotherapy and RN diagnosis, yielded an AUC of 0.966 and an optimism-corrected AUC of 0.967 via 10-fold cross-validation, which also revealed good discrimination. Calibration curves showed good agreement. Decision curve analysis confirmed the clinical utility of the model.

CONCLUSIONS

The presented radiomics model can be conveniently used to facilitate individualized prediction of the response to methylprednisolone in patients with RN.

A radiomics-based model on non-contrast CT for predicting cirrhosis: make the most of image data

Background

To establish and validate a radiomics-based model for predicting liver cirrhosis in patients with hepatitis B virus (HBV) by using non-contrast computed tomography (CT).

Methods

This retrospective study developed a radiomics-based model in a training cohort of 144 HBV-infected patients. Radiomic features were extracted from abdominal non-contrast CT scans. Features selection was performed with the least absolute shrinkage and operator (LASSO) method based on highly reproducible features. Support vector machine (SVM) was adopted to build a radiomics signature. Multivariate logistic regression analysis was used to establish a radiomics-based nomogram that integrated radiomics signature and other independent clinical predictors. Performance of models was evaluated through discrimination ability, calibration and clinical benefits. An internal validation was conducted in 150 consecutive patients.

Results

The radiomics signature comprised 25 cirrhosis-related features and showed significant differences between cirrhosis and non-cirrhosis cohorts (P < 0.001). A radiomics-based nomogram that integrates radiomics signature, alanine transaminase, aspartate aminotransferase, globulin and international normalized ratio showed great calibration and discrimination ability in the training cohort (area under the curve [AUC]: 0.915) and the validation cohort (AUC: 0.872). Decision curve analysis confirmed the most clinical benefits can be provided by the nomogram compared with other methods.

Conclusions

Our developed radiomics-based nomogram can successfully diagnose the status of cirrhosis in HBV-infected patients, that may help clinical decision-making.