Volume isotropic simultaneous interleaved black- and bright-blood imaging: a novel sequence for contrast-enhanced screening of brain metastasis

Stereotactic radiotherapy of brain metastases: clinical impact of three-dimensional SPACE imaging for 3T-MRI-based treatment planning

Purpose

For planning CyberKnife stereotactic radiosurgery (CK SRS) of brain metastases (BM), it is essential to precisely determine the exact number and location of BM in MRI. Recent MR studies suggest the superiority of contrast-enhanced 3D fast spin echo SPACE (sampling perfection with application-optimized contrast by using different flip angle evolutions) images over 3D gradient echo (GE) T1-weighted MPRAGE (magnetization-prepared rapid gradient echo) images for detecting small BM. The aim of this study is to test the usability of the SPACE sequence for MRI-based radiation treatment planning and its impact on changing treatment.

Methods

For MRI-based radiation treatment planning using 3T MRI in 199 patients with cerebral oligometastases, we compared the detectability of BM in post-gadolinium SPACE images, post-gadolinium MPRAGE images, and post-gadolinium late-phase MPRAGE images.

Results

When SPACE images were used for MRI-based radiation treatment planning, 29.8% and 16.9% more BM, respectively, were detected and included in treatment planning than in the post-gadolinium MPRAGE images and the post-gadolinium late-phase MPRAGE images (post-gadolinium MPRAGE imaging: n_total = 681, mean ± SD 3.4 ± 4.2; post-gadolinium SPACE imaging: n_total = 884, mean ± SD 4.4 ± 6.0; post-gadolinium late-phase MPRAGE imaging: n_total = 796, mean ± SD 4.0 ± 5.3; P_{post-gadolinium SPACE imaging versus post-gadolinium MPRAGE imaging} < 0.0001, P_{post-gadolinium SPACE imaging versus post-gadolinium late-phase MPRAGE imaging}< 0.0001).

Conclusion

For 3T MRI-based treatment planning of stereotactic radiosurgery of BM, we recommend the use of post-gadolinium SPACE imaging rather than post-gadolinium MPRAGE imaging.

A deep convolutional neural network-based automatic detection of brain metastases with and without blood vessel suppression

OBJECTIVES

To develop an automated model to detect brain metastases using a convolutional neural network (CNN) and volume isotropic simultaneous interleaved bright-blood and black-blood examination (VISIBLE) and to compare its diagnostic performance with the observer test.

METHODS

This retrospective study included patients with clinical suspicion of brain metastases imaged with VISIBLE from March 2016 to July 2019 to create a model. Images with and without blood vessel suppression were used for training an existing CNN (DeepMedic). Diagnostic performance was evaluated using sensitivity and false-positive results per case (FPs/case). We compared the diagnostic performance of the CNN model with that of the twelve radiologists.

RESULTS

Fifty patients (30 males and 20 females; age range 29-86 years; mean 63.3 ± 12.8 years; a total of 165 metastases) who were clinically diagnosed with brain metastasis on follow-up were used for the training. The sensitivity of our model was 91.7%, which was higher than that of the observer test (mean ± standard deviation; 88.7 ± 3.7%). The number of FPs/case in our model was 1.5, which was greater than that by the observer test (0.17 ± 0.09).

CONCLUSIONS

Compared to radiologists, our model created by VISIBLE and CNN to diagnose brain metastases showed higher sensitivity. The number of FPs/case by our model was greater than that by the observer test of radiologists; however, it was less than that in most of the previous studies with deep learning.

KEY POINTS

• Our convolutional neural network based on bright-blood and black-blood examination to diagnose brain metastases showed a higher sensitivity than that by the observer test. • The number of false-positives/case by our model was greater than that by the previous observer test; however, it was less than those from most previous studies. • In our model, false-positives were found in the vessels, choroid plexus, and image noise or unknown causes.

3D MR sequence capable of simultaneous image acquisitions with and without blood vessel suppression: utility in diagnosing brain metastases

OBJECTIVE

Volume isotropic simultaneous interleaved bright- and black-blood examination (VISIBLE) is a recently developed 3D MR sequence that provides simultaneous acquisitions of images with blood vessel suppression (Black) and images without it (Bright). Our purpose was to evaluate the usefulness of VISIBLE in detecting brain metastases.

METHODS

This prospective study included patients with suspected brain metastasis imaged with both VISIBLE and MPRAGE. From a data set, we compared the number of visualized blood vessels and the lesion-to-normal contrast-to-noise ratio (CNR) in 60 patients. We also performed an observer test to compare their diagnostic performance with VISIBLE, MPRAGE and only Black in 34 patients. Diagnostic performance was evaluated using a figure of merit (FOM), sensitivity, false-positive results per case (FPs/case) and reading time.

RESULTS

The number of vessels was significantly fewer in Black compared to MPRAGE and Bright (P < 0.0001). CNR was significantly higher with both Black and Bright than with MPRAGE (P < 0.005). In the observer test, significantly higher sensitivity (P < 0.0001) and FOM (P < 0.0001), significantly shorter reading time (P = 0.0001) and similar FPs/case were achieved with VISIBLE compared to MPRAGE. Compared to only Black, VISIBLE resulted in comparable sensitivity, but significantly fewer FPs/case (P = 0.0008).

CONCLUSION

VISIBLE can improve radiologists' diagnostic performance for brain metastasis.

KEY POINTS

• VISIBLE can achieve higher sensitivity and shorter reading time than MPRAGE. • VISIBLE can achieve lower false-positive rates than blood vessel suppressed images. • Compared to MPRAGE, VISIBLE can improve diagnostic performance for brain metastasis.