Patient-specific model-based segmentation of brain tumors in 3D intraoperative ultrasound images

Ilunga-Mbuyamba, Elisee; Aviña-Cervantes, Juan Gabriel; Lindner, Dirk; Arlt, Felix; Ituna-Yudonago, J.F.; Chalopin, Claire

doi:10.1007/s11548-018-1703-0

Cited by 13 publications

(16 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ritschel et al 9 and Ilunga-Mbuyamba et al 10 proposed iUS tumor segmentation methods that were evaluated on 13 and 14 cases, respectively (see table 1 and figure 3). The method proposed by Ritschel et al obtained a mean Dice score of 0.73.…”

Section: Comparison With Existing High Grade Tumor Segmentation Methodsmentioning

confidence: 99%

“…The loss function to train the models is the Dice loss function (equation 1) as it is commonly used with segmentation neural networks on unbalanced datasets. 16 loss(y true , y 10 14 0.77 (0.09) Methods R. 9 and I. 10 were evaluated on high grade tumors.…”

Section: Segmentation Networkmentioning

confidence: 99%

“…16 loss(y true , y 10 14 0.77 (0.09) Methods R. 9 and I. 10 were evaluated on high grade tumors. The numbers in brackets are standard deviations.…”

Section: Segmentation Networkmentioning

confidence: 99%

“…8 However, very few works have addressed this problem in iUS images. 9,10 In particular, these works focused on high grade tumors only. To the best of our knowledge, no method has been applied to low grade brain tumors in iUS, which are even more challenging to segment than high grade tumors due to a lower contrast.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Automatic segmentation of brain tumor in intraoperative ultrasound images using 3D U-Net

Carton

Chabanas

Munkvold

et al. 2020

Medical Imaging 2020: Image-Guided Procedures, Robotic Interventions, and Modeling

View full text Add to dashboard Cite

Because of the deformation of the brain during neurosurgery, intraoperative imaging can be used to visualize the actual location of the brain structures. These images are used for image-guided navigation as well as determining whether the resection is complete and localizing the remaining tumor tissue. Intraoperative ultrasound (iUS) is a convenient modality with short acquisition times. However, iUS images are difficult to interpret because of the noise and artifacts. In particular, tumor tissue is difficult to distinguish from healthy tissue and it is very difficult to delimit tumors in iUS images. In this paper, we propose an automatic method to segment low grade brain tumors in iUS images using a 2-D and 3-D U-Net. We trained the networks on three folds with twelve training cases and five test cases each. The obtained results are promising, with a median Dice score of 0.72. The volume differences between the estimated and ground truth segmentations were similar to the intra-rater volume differences. While these results are preliminary, they suggest that deep learning methods can be successfully applied to tumor segmentation in intraoperative images.

show abstract

Section: Comparison With Existing High Grade Tumor Segmentation Methodsmentioning

confidence: 99%

Section: Segmentation Networkmentioning

confidence: 99%

“…16 loss(y true , y 10 14 0.77 (0.09) Methods R. 9 and I. 10 were evaluated on high grade tumors. The numbers in brackets are standard deviations.…”

Section: Segmentation Networkmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automatic segmentation of brain tumor in intraoperative ultrasound images using 3D U-Net

Carton

Chabanas

Munkvold

et al. 2020

Medical Imaging 2020: Image-Guided Procedures, Robotic Interventions, and Modeling

View full text Add to dashboard Cite

show abstract

“…In the contemporary era, IOUS imaging is used in neuro-oncological surgery to investigate the spatiotemporal change of the tumor morphology during the operation. B-mode is the most commonly-used modality [ 12 ]. The solid part of the tumor appears hyperechogenic compared to the surrounding cerebral parenchyma, while cysts appear hypoechoic.…”

Section: Introductionmentioning

confidence: 99%

New Hope in Brain Glioma Surgery: The Role of Intraoperative Ultrasound. A Review

Pino¹,

Imperato

Musca³

et al. 2018

Brain Sciences

View full text Add to dashboard Cite

Maximal safe resection represents the gold standard for surgery of malignant brain tumors. As regards gross-total resection, accurate localization and precise delineation of the tumor margins are required. Intraoperative diagnostic imaging (Intra-Operative Magnetic Resonance-IOMR, Intra-Operative Computed Tomography-IOCT, Intra-Operative Ultrasound-IOUS) and dyes (fluorescence) have become relevant in brain tumor surgery, allowing for a more radical and safer tumor resection. IOUS guidance for brain tumor surgery is accurate in distinguishing tumor from normal parenchyma, and it allows a real-time intraoperative visualization. We aim to evaluate the role of IOUS in gliomas surgery and to outline specific strategies to maximize its efficacy. We performed a literature research through the Pubmed database by selecting each article which was focused on the use of IOUS in brain tumor surgery, and in particular in glioma surgery, published in the last 15 years (from 2003 to 2018). We selected 39 papers concerning the use of IOUS in brain tumor surgery, including gliomas. IOUS exerts a notable attraction due to its low cost, minimal interruption of the operational flow, and lack of radiation exposure. Our literature review shows that increasing the use of ultrasound in brain tumors allows more radical resections, thus giving rise to increases in survival.

show abstract

Open access segmentations of intraoperative brain tumor ultrasound images

Behboodi,

Carton,

Chabanas

et al. 2024

Medical Physics

View full text Add to dashboard Cite

PurposeRegistration and segmentation of magnetic resonance (MR) and ultrasound (US) images could play an essential role in surgical planning and resectioning brain tumors. However, validating these techniques is challenging due to the scarcity of publicly accessible sources with high‐quality ground truth information. To this end, we propose a unique set of segmentations (RESECT‐SEG) of cerebral structures from the previously published RESECT dataset to encourage a more rigorous development and assessment of image‐processing techniques for neurosurgery.Acquisition and Validation MethodsThe RESECT database consists of MR and intraoperative US (iUS) images of 23 patients who underwent brain tumor resection surgeries. The proposed RESECT‐SEG dataset contains segmentations of tumor tissues, sulci, falx cerebri, and resection cavity of the RESECT iUS images. Two highly experienced neurosurgeons validated the quality of the segmentations.Data Format and Usage NotesSegmentations are provided in 3D NIFTI format in the OSF open‐science platform: https://osf.io/jv8bk.Potential ApplicationsThe proposed RESECT‐SEG dataset includes segmentations of real‐world clinical US brain images that could be used to develop and evaluate segmentation and registration methods. Eventually, this dataset could further improve the quality of image guidance in neurosurgery.

show abstract

Patient-specific model-based segmentation of brain tumors in 3D intraoperative ultrasound images

Cited by 13 publications

References 43 publications

Automatic segmentation of brain tumor in intraoperative ultrasound images using 3D U-Net

Automatic segmentation of brain tumor in intraoperative ultrasound images using 3D U-Net

New Hope in Brain Glioma Surgery: The Role of Intraoperative Ultrasound. A Review

Open access segmentations of intraoperative brain tumor ultrasound images

Contact Info

Product

Resources

About