A deep convolutional neural network-based automatic delineation strategy for multiple brain metastases stereotactic radiosurgery

Liu, Yan; Stojadinović, Strahinja; Hrycushko, Brian; Wardak, Zabihullah; Lau, Steven; Lu, Weiguo; Yan, Yulong; Jiang, Steve B; Zhen, Xin; Timmerman, Robert; Nedzi, Lucien A.; Gu, Xuejun

doi:10.1371/journal.pone.0185844

Cited by 136 publications

(114 citation statements)

References 23 publications

Supporting

Mentioning

110

Contrasting

Unclassified

Order By: Relevance

“…While the current benchmark for brain metastases segmentation employs MRI imaging only [1,6], CT imaging is an essential reference for clinical treatment planning due to its spatial accuracy. We thus proposed a deep learning framework using multimodal imaging (MRI+CT) and ensemble neural networks for brain metastases detection and segmentation.…”

Section: Deep Learning Frameworkmentioning

confidence: 99%

“…However, such manually contouring process can be very time-consuming and suffer from large inter and intra-reader variability [11]. Driven by the ever-increasing capability of deep learning, automated segmentation of BMs using neural networks has been recently proposed [1,6]. Previous works on computer-aided segmentation of BMs used only MRI as an imaging input.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multimodal Volume-Aware Detection and Segmentation for Brain Metastases Radiosurgery

Weng

et al. 2019

Artificial Intelligence in Radiation Therapy

View full text Add to dashboard Cite

Stereotactic radiosurgery (SRS), which delivers high doses of irradiation in a single or few shots to small targets, has been a standard of care for brain metastases. While very effective, SRS currently requires manually intensive delineation of tumors. In this work, we present a deep learning approach for automated detection and segmentation of brain metastases using multimodal imaging and ensemble neural networks. In order to address small and multiple brain metastases, we further propose a volume-aware Dice loss which optimizes model performance using the information of lesion size. This work surpasses current benchmark levels and demonstrates a reliable AI-assisted system for SRS treatment planning for multiple brain metastases.

show abstract

Section: Deep Learning Frameworkmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Multimodal Volume-Aware Detection and Segmentation for Brain Metastases Radiosurgery

Weng

et al. 2019

Artificial Intelligence in Radiation Therapy

View full text Add to dashboard Cite

show abstract

“…DeepMedic" [13], with the expectation of improved BM segmentation accuracy and higher computational efficiency. The approach was validated with both the BRATS database [14] and their post-contrast T1-weighted MRI collection of brain metastases with a mean tumor volume of 672 mm 3 .…”

mentioning

confidence: 99%

Automated Brain Metastases Detection Framework for T1-Weighted Contrast-Enhanced 3D MRI

Dikici

Ryu

Demirer

et al. 2020

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

Brain Metastases (BM) complicate 20-40% of cancer cases. BM lesions can present as punctate (1 mm) foci, requiring high-precision Magnetic Resonance Imaging (MRI) in order to prevent inadequate or delayed BM treatment. However, BM lesion detection remains challenging partly due to their structural similarities to normal structures (e.g., vasculature). We propose a BMdetection framework using a single-sequence gadoliniumenhanced T1-weighted 3D MRI dataset. The framework focuses on detection of smaller (< 15 mm) BM lesions and consists of: (1) candidate-selection stage, using Laplacian of Gaussian approach for highlighting parts of a MRI volume holding higher BM occurrence probabilities, and (2) detection stage that iteratively processes cropped region-of-interest volumes centered by candidates using a custom-built 3D convolutional neural network ("CropNet"). Data is augmented extensively during training via a pipeline consisting of random gamma correction and elastic deformation stages; the framework thereby maintains its invariance for a plausible range of BM shape and intensity representations. This approach is tested using five-fold cross-validation on 217 datasets from 158 patients, with training and testing groups randomized per patient to eliminate learning bias. The BM database included lesions with a mean diameter of ~5.4 mm and a mean volume of ~160 mm 3 . For 90% BM-detection sensitivity, the framework produced on average 9.12 falsepositive BM detections per patient (standard deviation of 3.49); for 85% sensitivity, the average number of falsepositives declined to 5.85. Comparative analysis showed that the framework produces comparable BM-detection accuracy with the state-of-art approaches validated for significantly larger lesions.Index Terms-magnetic resonance imaging, brain metastases, convolutional neural networks, deep learning, scale-space representations, computer-aided detection, medical image analysis.

show abstract

“…An algorithm was devised using convolutional neural network for segmentation of brain metastases from MRI [35]. Image patches were fed to the net-work for voxel-wise classification which made the setup efficient for segmenting small lesions.…”

Section: Radiomics Using Deep Learningmentioning

confidence: 99%

A Survey on Recent Advancements for AI Enabled Radiomics in Neuro-Oncology

Anwar

Altaf

Rafique

et al. 2020

Radiomics and Radiogenomics in Neuro-Oncology

View full text Add to dashboard Cite

Artificial intelligence (AI) enabled radiomics has evolved immensely especially in the field of oncology. Radiomics provide assistance in diagnosis of cancer, planning of treatment strategy, and prediction of survival. Radiomics in neuro-oncology has progressed significantly in the recent past. Deep learning has outperformed conventional machine learning methods in most image-based applications. Convolutional neural networks (CNNs) have seen some popularity in radiomics, since they do not require hand-crafted features and can automatically extract features during the learning process. In this regard, it is observed that CNN based radiomics could provide state-of-the-art results in neuro-oncology, similar to the recent success of such methods in a wide spectrum of medical image analysis applications. Herein we present a review of the most recent best practices and establish the future trends for AI enabled radiomics in neuro-oncology.

show abstract

A deep convolutional neural network-based automatic delineation strategy for multiple brain metastases stereotactic radiosurgery

Cited by 136 publications

References 23 publications

Multimodal Volume-Aware Detection and Segmentation for Brain Metastases Radiosurgery

Multimodal Volume-Aware Detection and Segmentation for Brain Metastases Radiosurgery

Automated Brain Metastases Detection Framework for T1-Weighted Contrast-Enhanced 3D MRI

A Survey on Recent Advancements for AI Enabled Radiomics in Neuro-Oncology

Contact Info

Product

Resources

About