Deep-Learning Detection of Cancer Metastases to the Brain on MRI

Overview

Journal J Magn Reson Imaging

Date 2020 Mar 14

PMID 32167652

Citations 48

Authors

Min Zhang

Geoffrey S Young

Huai Chen

Jing Li

Lei Qin

J Ricardo McFaline-Figueroa

David A Reardon

Xinhua Cao

Xian Wu

Xiaoyin Xu

Affiliations

Soon will be listed here.

Abstract

Background: Approximately one-fourth of all cancer metastases are found in the brain. MRI is the primary technique for detection of brain metastasis, planning of radiotherapy, and the monitoring of treatment response. Progress in tumor treatment now requires detection of new or growing metastases at the small subcentimeter size, when these therapies are most effective.

Purpose: To develop a deep-learning-based approach for finding brain metastasis on MRI.

Study Type: Retrospective.

Sequence: Axial postcontrast 3D T -weighted imaging.

Field Strength: 1.5T and 3T.

Population: A total of 361 scans of 121 patients were used to train and test the Faster region-based convolutional neural network (Faster R-CNN): 1565 lesions in 270 scans of 73 patients for training; 488 lesions in 91 scans of 48 patients for testing. From the 48 outputs of Faster R-CNN, 212 lesions in 46 scans of 18 patients were used for training the RUSBoost algorithm (MatLab) and 276 lesions in 45 scans of 30 patients for testing.

Assessment: Two radiologists diagnosed and supervised annotation of metastases on brain MRI as ground truth. This data were used to produce a 2-step pipeline consisting of a Faster R-CNN for detecting abnormal hyperintensity that may represent brain metastasis and a RUSBoost classifier to reduce the number of false-positive foci detected.

Statistical Tests: The performance of the algorithm was evaluated by using sensitivity, false-positive rate, and receiver's operating characteristic (ROC) curves. The detection performance was assessed both per-metastases and per-slice.

Results: Testing on held-out brain MRI data demonstrated 96% sensitivity and 20 false-positive metastases per scan. The results showed an 87.1% sensitivity and 0.24 false-positive metastases per slice. The area under the ROC curve was 0.79.

Conclusion: Our results showed that deep-learning-based computer-aided detection (CAD) had the potential of detecting brain metastases with high sensitivity and reasonable specificity.

Level Of Evidence: 3 TECHNICAL EFFICACY STAGE: 2 J. Magn. Reson. Imaging 2020;52:1227-1236.

Citing Articles

Enhancing deep learning methods for brain metastasis detection through cross-technique annotations on SPACE MRI.

Wald T, Hamm B, Holzschuh J, Shafie R, Kudak A, Kovacs B Eur Radiol Exp. 2025; 9(1):15.

PMID: 39913077 PMC: 11802942. DOI: 10.1186/s41747-025-00554-5.

A review of deep learning for brain tumor analysis in MRI.

Dorfner F, Patel J, Kalpathy-Cramer J, Gerstner E, Bridge C NPJ Precis Oncol. 2025; 9(1):2.

PMID: 39753730 PMC: 11698745. DOI: 10.1038/s41698-024-00789-2.

Automatic detection and multi-component segmentation of brain metastases in longitudinal MRI.

Andrearczyk V, Schiappacasse L, Abler D, Wodzinski M, Hottinger A, Raccaud M Sci Rep. 2024; 14(1):31603.

PMID: 39738168 PMC: 11686181. DOI: 10.1038/s41598-024-78865-7.

Computational Modeling and AI in Radiation Neuro-Oncology and Radiosurgery.

Lee C, Yang H, Wu H, Lin Y, Lu C, Peng S Adv Exp Med Biol. 2024; 1462:307-322.

PMID: 39523273 DOI: 10.1007/978-3-031-64892-2_18.

Advances in imaging modalities for spinal tumors.

Takamiya S, Malvea A, Ishaque A, Pedro K, Fehlings M Neurooncol Adv. 2024; 6(Suppl 3):iii13-iii27.

PMID: 39430391 PMC: 11485884. DOI: 10.1093/noajnl/vdae045.

References

Greto D, Scoccianti S, Compagnucci A, Arilli C, Casati M, Francolini G . Gamma Knife Radiosurgery in the management of single and multiple brain metastases. Clin Neurol Neurosurg. 2016; 141:43-7. DOI: 10.1016/j.clineuro.2015.12.009. View

Shen D, Wu G, Zhang D, Suzuki K, Wang F, Yan P . Machine learning in medical imaging. Comput Med Imaging Graph. 2015; 41:1-2. DOI: 10.1016/j.compmedimag.2015.02.001. View

Park J, Kim E . Contrast-enhanced, three-dimensional, whole-brain, black-blood imaging: application to small brain metastases. Magn Reson Med. 2010; 63(3):553-61. DOI: 10.1002/mrm.22261. View

Yoo H, Jung E, Nam B, Shin S, Gwak H, Kim M . Growth rate of newly developed metastatic brain tumors after thoracotomy in patients with non-small cell lung cancer. Lung Cancer. 2010; 71(2):205-8. DOI: 10.1016/j.lungcan.2010.05.013. View

Lippitz B, Lindquist C, Paddick I, Peterson D, ONeill K, Beaney R . Stereotactic radiosurgery in the treatment of brain metastases: the current evidence. Cancer Treat Rev. 2013; 40(1):48-59. DOI: 10.1016/j.ctrv.2013.05.002. View

Menze B, Jakab A, Bauer S, Kalpathy-Cramer J, Farahani K, Kirby J . The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS). IEEE Trans Med Imaging. 2014; 34(10):1993-2024. PMC: 4833122. DOI: 10.1109/TMI.2014.2377694. View

Xiao Y, Wu J, Lin Z, Zhao X . A deep learning-based multi-model ensemble method for cancer prediction. Comput Methods Programs Biomed. 2017; 153:1-9. DOI: 10.1016/j.cmpb.2017.09.005. View

Sharma H, Zerbe N, Klempert I, Hellwich O, Hufnagl P . Deep convolutional neural networks for automatic classification of gastric carcinoma using whole slide images in digital histopathology. Comput Med Imaging Graph. 2017; 61:2-13. DOI: 10.1016/j.compmedimag.2017.06.001. View

Perez-Ramirez U, Arana E, Moratal D . Brain metastases detection on MR by means of three-dimensional tumor-appearance template matching. J Magn Reson Imaging. 2016; 44(3):642-52. DOI: 10.1002/jmri.25207. View

10.

Grovik E, Yi D, Iv M, Tong E, Rubin D, Zaharchuk G . Deep learning enables automatic detection and segmentation of brain metastases on multisequence MRI. J Magn Reson Imaging. 2019; 51(1):175-182. PMC: 7199496. DOI: 10.1002/jmri.26766. View

11.

Sunwoo L, Kim Y, Choi S, Kim K, Kang J, Kang Y . Computer-aided detection of brain metastasis on 3D MR imaging: Observer performance study. PLoS One. 2017; 12(6):e0178265. PMC: 5464563. DOI: 10.1371/journal.pone.0178265. View

12.

Park H, Chiang V, Knisely J, Raldow A, Yu J . Stereotactic radiosurgery with or without whole-brain radiotherapy for brain metastases: an update. Expert Rev Anticancer Ther. 2011; 11(11):1731-8. DOI: 10.1586/era.11.165. View

13.

Devoid H, McTyre E, Page B, Metheny-Barlow L, Ruiz J, Chan M . Recent advances in radiosurgical management of brain metastases. Front Biosci (Schol Ed). 2015; 8(1):203-14. DOI: 10.2741/s458. View

14.

Firmino M, Morais A, Mendoca R, Dantas M, Hekis H, Valentim R . Computer-aided detection system for lung cancer in computed tomography scans: review and future prospects. Biomed Eng Online. 2014; 13:41. PMC: 3995505. DOI: 10.1186/1475-925X-13-41. View

15.

Segonne F, Dale A, Busa E, Glessner M, Salat D, Hahn H . A hybrid approach to the skull stripping problem in MRI. Neuroimage. 2004; 22(3):1060-75. DOI: 10.1016/j.neuroimage.2004.03.032. View

16.

Ren S, He K, Girshick R, Sun J . Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Trans Pattern Anal Mach Intell. 2016; 39(6):1137-1149. DOI: 10.1109/TPAMI.2016.2577031. View

17.

Oh Y, Taylor S, Bekele B, Debnam J, Allen P, Suki D . Number of metastatic sites is a strong predictor of survival in patients with nonsmall cell lung cancer with or without brain metastases. Cancer. 2009; 115(13):2930-8. DOI: 10.1002/cncr.24333. View

18.

Doi K . Computer-aided diagnosis in medical imaging: historical review, current status and future potential. Comput Med Imaging Graph. 2007; 31(4-5):198-211. PMC: 1955762. DOI: 10.1016/j.compmedimag.2007.02.002. View

19.

Linskey M, Andrews D, Asher A, Burri S, Kondziolka D, Robinson P . The role of stereotactic radiosurgery in the management of patients with newly diagnosed brain metastases: a systematic review and evidence-based clinical practice guideline. J Neurooncol. 2009; 96(1):45-68. PMC: 2808519. DOI: 10.1007/s11060-009-0073-4. View

20.

Charron O, Lallement A, Jarnet D, Noblet V, Clavier J, Meyer P . Automatic detection and segmentation of brain metastases on multimodal MR images with a deep convolutional neural network. Comput Biol Med. 2018; 95:43-54. DOI: 10.1016/j.compbiomed.2018.02.004. View