Multi parametric magnetic resonance imaging for radiation treatment planning

Radiomics Based on Multimodal magnetic resonance imaging for the Differential Diagnosis of Benign and Malignant Vertebral Compression Fractures

OBJECTIVES

Recent studies have indicated that radiomics may have excellent performance and clinical application prospects in the differential diagnosis of benign and malignant vertebral compression fractures (VCFs). However, multimodal magnetic resonance imaging (MRI)-based radiomics model is rarely used in the differential diagnosis of benign and malignant VCFs, and is limited to lumbar. Herein, this study intends to develop and validate MRI radiomics models for differential diagnoses of benign and malignant VCFs in patients.

METHODS

This cross-sectional study involved 151 adult patients diagnosed with VCF in The First Affiliated Hospital of Soochow University in 2016-2021. The study was conducted in three steps: (i) the original MRI images were segmented, and the region of interest (ROI) was marked out; (ii) among the extracted features, those features with Pearson's correlation coefficient lower than 0.9 and the top 15 with the highest variance and Lasso regression coefficient less than and more than 0 were selected; (iii) MRI images and combined data were studied by logistic regression, decision tree, random forest and extreme gradient boosting (XGBoost) models in training set and the test set (ratio of 8:2), respectively; and the models were further verified and evaluated for the differential diagnosis performance. The evaluated indexes included area under receiver (AUC) of operating characteristic curve, accuracy, sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), and 95% confidence intervals (CIs). The AUCs were used to assess the predictive performance of different machine learning modes for benign and malignant VCFs.

RESULTS

A total of 1144 radiomics features, and 14 clinical features were extracted. Finally, 12 radiomics features were included in the radiomics model, and 12 radiomics features with 14 clinical features were included in the combined model. In the radiomics model, the differential diagnosis performance in the logistic regression model with the AUC of 0.905 ± 0.026, accuracy of 0.817 ± 0.057, sensitivity of 0.831 ± 0.065, and negative predictive value of 0.813 ± 0.042, was superior to the other three. In the combined model, XGBoost model had the superior differential diagnosis performance with specificity (0.979 ± 0.026) and positive predictive value (0.971 ± 0.035).

CONCLUSION

The multimodal MRI-based radiomics model performed well in the differential diagnosis of benign and malignant VCFs, which may provide a tool for clinicians to differentially diagnose VCFs.

Evaluating contouring accuracy and dosimetry impact of current MRI-guided adaptive radiation therapy for brain metastases: a retrospective study

BACKGROUND

Magnetic resonance imaging (MRI) guided adaptive radiotherapy (MRgART) has gained increasing attention, showing clinical advantages over conventional radiotherapy. However, there are concerns regarding online target delineation and modification accuracy. In our study, we aimed to investigate the accuracy of brain metastases (BMs) contouring and its impact on dosimetry in 1.5 T MRI-guided online adaptive fractionated stereotactic radiotherapy (FSRT).

METHODS

Eighteen patients with 64 BMs were retrospectively evaluated. Pre-treatment 3.0 T MRI scans (gadolinium contrast-enhanced T1w, T1c) and initial 1.5 T MR-Linac scans (non-enhanced online-T1, T2, and FLAIR) were used for gross target volume (GTV) contouring. Five radiation oncologists independently contoured GTVs on pre-treatment T1c and initial online-T1, T2, and FLAIR images. We assessed intra-observer and inter-observer variations and analysed the dosimetry impact through treatment planning based on GTVs generated by online MRI, simulating the current online adaptive radiotherapy practice.

RESULTS

The average Dice Similarity Coefficient (DSC) for inter-observer comparison were 0.79, 0.54, 0.59, and 0.64 for pre-treatment T1c, online-T1, T2, and FLAIR, respectively. Inter-observer variations were significantly smaller for the 3.0 T pre-treatment T1c than for the contrast-free online 1.5 T MR scans (P < 0.001). Compared to the T1c contours, the average DSC index of intra-observer contouring was 0.52‒0.55 for online MRIs. For BMs larger than 3 cm3, visible on all image sets, the average DSC indices were 0.69, 0.71 and 0.64 for online-T1, T2, and FLAIR, respectively, compared to the pre-treatment T1c contour. For BMs < 3 cm3, the average visibility rates were 22.3%, 41.3%, and 51.8% for online-T1, T2, and FLAIR, respectively. Simulated adaptive planning showed an average prescription dose coverage of 63.4‒66.9% when evaluated by ground truth planning target volumes (PTVs) generated on pre-treatment T1c, reducing it from over 99% coverage by PTVs generated on online MRIs.

CONCLUSIONS

The accuracy of online target contouring was unsatisfactory for the current MRI-guided online adaptive FSRT. Small lesions had poor visibility on 1.5 T non-contrast-enhanced MR-Linac images. Contour inaccuracies caused a one-third drop in prescription dose coverage for the target volume. Future studies should explore the feasibility of contrast agent administration during daily treatment in MRI-guided online adaptive FSRT procedures.